e

in
nl
O
ty
Introduction to
Data Base Management Systems

r si
ve
ni
U
ity
m
)A
(c
 e
in
© Amity University Press

All Rights Reserved

nl
No parts of this publication may be reproduced, stored in a retrieval system or transmitted
in any form or by any means, electronic, mechanical, photocopying, recording or otherwise
without the prior permission of the publisher.

O
Advisory Committee

ty
Chairman : Ms. Monica Agarwal
Members : Prof. Arun Bisaria

si
Dr. Priya Mary Mathew
Prof. Aindril De
r
Mr. Alok Awtans
ve
Dr. Coral J Barboza
Dr. Monica Rose
Mr. Sachit Paliwal
ni

SLM Review Committee

Mr. Gaurav Agarwal
U

Ms. Nitika Khanna

Ms. Rashmi Saxena
Ms. Renu Singh
ity

Ms. Mona Chaudhary

m
)A
(c

Published by Amity University Press for exclusive use of Amity Directorate of Distance and Online Education,
Amity University, Noida-201313
 Contents

e
Page No.

in
Module – 1 : Introduction to DBMS 01
1.1 DBMS Concepts
1.1.1 Introduction

nl
1.1.2 An overview of Database Management System
1.1.3 Database System Vs File System
1.1.4 Components of DBMS

O
1.1.5 Advantages of DBMS
1.1.6 Types of Database System
1.2 DBMS Architecture

ty
2.1.1 Database Three Level Architecture
2.1.2 Schema and Instance
2.1.3 Data Independence

si
2.1.4 Database Language and Interfaces
2.1.5 Data Models
2.1.6 Data Definitions Language, DML
1.3 Database Users
r
ve
3.1.1 Types of Database Users
3.1.2 Responsibility of Database Administrator

Module - 2: Relational Database & ER Model 69

ni

2.1 Relational Model

2.1.1 Introduction
2.1.2 Relational System Concept
U

2.1.3 Properties of Relation

2.1.4 CODD’s Rules
2.1.5 Relational Model
ity

2.1.6 Data Modeling using Entity Relationship Model

2.1.7 Elements of E-R Model
2.1.8 Entity and Entity Types, Entity Set
2.1.9 Attribute Concept and Types of Attributes
m

2.1.10 Relationship Types

2.1.11 Role and Structural Constraints
)A

2.2 Enhanced Entity -Relationship Model

2.2.1 Enhanced E-R Model Concepts
2.2.2 Subclasses, Superclasses and Inheritance
2..2.3 Generalization
2.2.4 Specialization
(c

2.2.5 Constraints and Characteristics of of EER Model

 2.2.6 Modeling of UNION types using Categories
Case Study

e
Module - 3: Relational Model Objects 90

in
3.1 Understanding Relational Model
1.1.1 Domains and Relation
1.1.2 Keys Concepts and Its Type

nl
1.1.3 Relational Model Constraints
1.1.4 Domain Constraints

O
1.1.5 Key Constraints
1.1.6 Referential Key Constraints
1.1.7 Update Operations and Dealing with Constraints Violations
3.2 Relational Model Operations and SQL Language

ty
2.1.1 Relational Operators
2.1.2 Set Operations on Relation

si
2.1.3 Join Operations
2.1.4 SQL Language and Its Characteristics
2.1.5 Data Definition Language and its Features
r
2.1.6 Data Manipulation Language and its Features
ve
2.1.7 Data Definitions Commands-Create, Alter, Truncate, Drop
2.1.8 Data Manipulation Commands-Select, update, delete
2.1.9 SQL String Functions
2.1.10 Views
ni

2..1.11 SQL Joins

Module - 4 : Database Design 159

U

4.1 Functional Dependencies

4.1.1 Introduction
4.1.2 Rules of Functional Dependency
4.1.3 Types of Functional Dependencies
ity

4.2 Normalization
4.2.1 Design of Database Schema Anomalies
4.2.2 Process of Normalization
m

4.2.3 First Normal Form

4.2.4 Second Normal Form
4.2.5 Third Normal Formal
)A

4.2.6 Boyce -CODD Normal Form

4.2.7 Multivalued Dependency
4.2.8 Forth Normal Form
4.2.9 Join Dependency
(c

4.2.10 Fifth Normal Form

CASE STUDY
 Module -5: Data Recovery & Protection 203
5.1 Concurrency Control Techniques

e
5.1.1 Introduction

in
5.1.2 Transaction and System Concept
5.1.3 Properties of Transactions
5.1.4 Locking Concepts and Types of Locking

nl
5.1.5 Two Phase Locking Techniques
5.1.6 Concurrency Control Based on Time Stamping

O
5.1.7 Multivesion Concurrency Control
5.1.8 Granularity of Data Items and Multiple Granularity Locking
5.1.9 Define Schedule and Serializibility

ty
5.1.10 Characterizing Schedules Based on Serializability
5.2 Database Recovery and Security
5.2.1 Recovery Concepts

si
5.2.2 Recovery Techniques Based on Deferred Update
5.2.3 Recovery Techniques Based on Immediate Update
5.2.4 Shadow Paging r
ve
5.2.5 System Recovery
5.2.6 Media Recovery
5.2.7 Database Security
ni
U
ity
m
)A
(c
(c
)A
m
ity
U
ni
ve
r si
ty
O
nl
in
e
 Introduction to Data Base Management Systems 1

Module – 1 : Introduction to DBMS

Notes

e
Structure:

in
1.1 DBMS Concepts
1.1.1 Introduction

nl
1.1.2 An overview of Database Management System
1.1.3 Database System Vs File System

O
1.1.4 Components of DBMS
1.1.5 Advantages of DBMS
1.1.6 Types of Database System

ty
1.2 DBMS Architecture
2.1.1 Database Three Level Architecture

si
2.1.2 Schema and Instance
2.1.3 Data Independence
2.1.4 Database Language and Interfaces
2.1.5 Data Models
r
ve
2.1.6 Data Definitions Language, DML
1.3 Database Users
3.1.1 Types of Database Users
ni

3.1.2 Responsibility of Database Administrator

U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 2 Introduction to Data Base Management Systems

Unit - 1.1: DBMS Concepts

Notes

e
Objectives

in
At the end of this unit, you will be able to:

●● Learn about Database Management System

nl
●● Understand Database System Vs. File System
●● Know Components and Advantages of DBMS

O
●● Learn the Types of Database System

1.1.1 Introduction
A database management device (or DBMS) is basically nothing extra than an

ty
automatic statistics-keeping gadget. Users of the device are given centres to carry
out several styles of operations on this kind of device for either manipulation of the
information within the database or the management of the database structure itself.

si
Database management systems (DBMSs) are categorized in line with their data
structures or sorts. There are several types of databases that may be used on a
mainframe to take advantage of: inverted listing, hierarchic, community, or relational.
r
Mainframe web sites generally tend to use a hierarchical version when the information
ve
structure (not information values) of the data wanted for a software is pretty static. For
example, an invoice of fabric (bom) database structure constantly has an excessive
stage meeting element quantity, and numerous levels of additives with subcomponents.
The structure usually has an issue forecast, value, and pricing facts, and so forth. The
structure of the statistics for a bom software not often changes, and new statistics
ni

factors (now not values) are not often recognized. An application usually starts at the
pinnacle with the meeting component number, and is going all the way down to the
element additives. Hierarchical and relational database structures have no unusual
U

advantages. RDBMS has the extra, extensive benefit over the hierarchical database
of being non-navigational. Via navigational, we suggest that in a hierarchical database,
the software programmer need to recognize the structure of the database. The program
should comprise precise logic to navigate from the foundation phase to the desired
ity

infant segments containing the desired attributes or elements. This system should still
get the right of entry to the intervening segments, despite the fact that they are not
wanted.
m

Important Terminology
Records: Records is not anything however information and facts are stored or
loose flowing over a network, commonly it is raw and unprocessed. For instance: when
)A

you go to any website, they may save you IP cope with, this is data, in go back they
could add a cookie in your browser, marking you that you visited the website, that is
information, your name, it is information, your age, it’s data. Data turns into information
whilst it’s miles processed, turning it into something significant. Like, primarily based
on the cookie records stored on person’s browser, if an internet site can examine that
(c

typically men of age 20-25 go to us extra that is information, derived from the statistics
accumulated.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 3

DBMS: A DBMS is a software that lets in advent, definition and manipulation of

Notes

e
database, permitting customers to save, process and analyse records effortlessly.
DBMS offers us with an interface or a tool, to perform various operations like creating
database, storing information in it, updating records, growing tables within the database

in
and plenty extra. DBMS also offers safety and security to the databases. It additionally
continues statistics consistency in case of more than one users. Right here are a few
examples of famous DBMS used nowadays:

nl
◌◌ MySQL
◌◌ Oracle

O
◌◌ SQL Server
◌◌ IBM DB2
◌◌ PostgreSQL

ty
◌◌ Amazon SimpleDB (cloud based) etc.
Database: database is a set of inter-associated information which facilitates
in green retrieval, insertion and deletion of data from database and organizes the

si
information inside the shape of tables, perspectives, schemas, reports and many
others. As an example, college database organizes the facts approximately college
students, school, and admin body of workers etc. Which helps in efficient retrieval,
insertion and deletion of data from it.
r
ve
DDL is short name of Data Definition Language, which deals with database
schemas and descriptions, of how the data should reside in the database.

●● CREATE: to create a database and its objects like (table, index, views, store
procedure, function, and triggers)
ni

●● ALTER: alters the structure of the existing database

●● DROP: delete objects from the database
U

●● TRUNCATE: remove all records from a table, including all spaces allocated for the
records are removed
●● COMMENT: add comments to the data dictionary
ity

●● RENAME: rename an object

DML is short name of Data Manipulation Language which deals with data
manipulation and includes most common SQL statements such SELECT, INSERT,
UPDATE, DELETE, etc., and it is used to store, modify, retrieve, delete and update data
m

in a database.

●● SELECT: retrieve data from a database

)A

●● INSERT: insert data into a table

●● UPDATE: updates existing data within a table
●● DELETE: Delete all records from a database table
●● MERGE: UPSERT operation (insert or update)
(c

●● CALL: call a PL/SQL or Java subprogram

●● EXPLAIN PLAN: interpretation of the data access path

Amity Directorate of Distance & Online Education

 4 Introduction to Data Base Management Systems

●● LOCK TABLE: concurrency Control

Notes

e
Database Management System: the software program that is used to manipulate
database is known as database management system (DBMS). As an instance, mySQL,

in
oracle and so forth. Are popular commercial DBMS used in exceptional packages?
DBMS permits customers the following responsibilities:

◌◌ Data definition: it enables in introduction, change and removal of definitions

nl
that outline the employer of records in database.
◌◌ Data updation: it allows in insertion, amendment and deletion of the actual
records in the database.

O
◌◌ DATA retrieval: it allows in retrieval of information from the database which
may be used by packages for diverse functions.
User administration: it enables in registering and tracking users, imposing facts

ty
security, tracking performance, preserving information integrity, managing concurrency
manipulate and getting better information corrupted by way of surprising failure.

What structures exist in a relational database? Relational databases include the

si
subsequent systems:

Database: A database is a logical grouping of facts. It incorporates a set of

associated desk spaces and index spaces. Normally, a database contains all of the
r
facts this is associated with one application or with a collection of associated programs.
ve
You could have a payroll database or a stock database, as an instance:

Table: A table is a logical shape made of rows and columns. Rows haven’t any
fixed order, so in case you retrieve information you would possibly need to type the
facts. The order of the columns is the order distinct whilst the desk turned into created
ni

via the database administrator. On the intersection of each column and row is a specific
information object referred to as a value, or, extra exactly, an atomic price. A table is
called with a high-degree qualifier of the owner’s consumer identity accompanied by
U

using the table name, as an instance check. Dept. or prod. Dept. There are three kinds
of tables:

◌◌ A base table this is created and holds chronic information

ity

◌◌ A transient table that shops intermediate question results

◌◌ An outcomes table that is back when you question tables
m
)A
(c

Figure 1. Example of a DB2 table (department table)

Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 5

In this table we use:

Notes

e
Columns:

in
The ordered set of columns are deptno, deptname, mgrno, and admrdept. All of the
records in a given column should be of the same statistics type.

◌◌ Rows: each row carries statistics for an unmarried department.

nl
◌◌ Values: at the intersection of a column and row is a cost. As an instance,
making plans is the fee of the deptname column inside the row for department
b01.

O
Indexes:
We realize that facts are saved in the form of information. Each report has a key
discipline, which helps it to be identified uniquely.

ty
Indexing is an information structure technique to effectively retrieve records from
the database files based on a few attributes on which the indexing has been performed.

si
An index is used for 2 functions:

◌◌ For performance, to retrieve information values more quickly

◌◌ For uniqueness. Through creating an index on an employee’s name, you
r
could retrieve information greater fast for that worker than by means of
ve
scanning the complete table. Also, by creating a completely unique index
on an employee quantity, db2 will enforce the distinctiveness of every fee. A
unique index is the only way db2 can put into effect forte. Growing an index
robotically creates the index space, the facts set that consists of the index.
ni

Indexing is defined based totally on its indexing attributes. Indexing can be of

the following sorts −
◌◌ Number one index − primary index is described on an ordered records file.
U

The statistics record is ordered on a key discipline. The important thing

discipline is usually the number one key of the relation.
◌◌ Secondary index − secondary index may be generated from an area that is a
candidate key and has a completely unique cost in each record, or a non-key
ity

with replica values.

◌◌ Clustering index − clustering index is defined on an ordered statistics file. The
information document is ordered on a non-key subject. Ordered indexing is of
sorts −
m

 Dense index
 Sparse index
)A

Dense index
In dense index, there may be an index document for each search key fee in the
database. This makes searching quicker however requires more space to store index
records itself. Index statistics comprise search key value and a pointer to the real record
(c

on the disk

Amity Directorate of Distance & Online Education

 6 Introduction to Data Base Management Systems

Notes

e
in
Sparse Index

nl
In sparse index, index records aren’t created for every seek key. An index file
here includes a search key and an actual pointer to the facts on the disk. To look a
document, we first continue via index document and reach on the actual region of

O
the information. If the data we are searching out is not wherein we at once reach with
the aid of following the index, then the gadget starts sequential seek till the preferred
records is discovered.

ty
Multilevel Index
r si
Index information comprise seek-key values and statistics suggestions. Multilevel
ve
index is stored at the disk at the side of the actual database files. As the scale of the
database grows, so does the scale of the indices. There’s an enormous want to hold
the index information inside the fundamental reminiscence that allows you to accelerate
the quest operations. If single-degree index is used, then a large length index cannot be
ni

kept in memory which ends up in a couple of disk accesses.

U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 7

Keys:
Notes

e
A secret’s one or extra columns that are diagnosed as such within the advent of a
desk or index, or inside the definition of referential integrity.

in
Primary key:
A table can most effective have one number one key as it defines the entity. There

nl
are the requirements for a number one key:

◌◌ It must have a fee, this is, it cannot be null.

◌◌ It must be specific, that is, it should have a completely unique index defined

O
on it. Particular key
We already realize that a primary key must be specific, however it is feasible to
have multiple unique key in a table. In our EMP table instance, the worker quantity is

ty
described because the number one key and is therefore unique. If we additionally had
a social safety value in our table, hopefully that price would be specific. To assure this,
you could create a completely unique index on the social security column.

si
Foreign key
A foreign secret is a key this is laid out in a referential integrity constraint to make
r
its lifestyles dependent on a primary or precise key (discern key) in every other table.
The instance given is that of an employee’s work branch variety referring to the number
ve
one key described on the branch wide variety in the dept. table. This constraint is part
of the definition of the table.

1.1.2 An Overview of Database Management System

ni

Database control gadget or DBMS in short refers to the technology of storing and
retrieving customer’s statistics with utmost performance alongside appropriate security
U

features. This tutorial explains the basics of DBMS which includes its architecture,
information models, facts schemas, records independence, e-r version, relation version,
relational database design, and storage and document structure and plenty more.

Why to examine DBMS? Traditionally, Data become organized in file codecs.

ity

DBMS changed into a brand new idea then, and all the research became carried out
to make it triumph over the deficiencies in traditional fashion of records control. A
contemporary DBMS has the subsequent traits −

●● Actual-world entity − a modern DBMS is greater sensible and uses real-global

m

entities to design its structure. It makes use of the behaviour and attributes too.
As an instance, a college database may additionally use students as an entity and
their age as an attribute.
)A

●● Relation-based totally tables − DBMS lets in entities and members of the family
among them to form tables. A user can understand the structure of a database just
by searching at the table names.
●● Isolation of facts and application − a database system is totally special than its
(c

facts. A database is an active entity, while facts is said to be passive, on which

the database works and organizes. DBMS also shops metadata, that is records
approximately information, to ease its very own system.
Amity Directorate of Distance & Online Education
 8 Introduction to Data Base Management Systems

●● Less redundancy − DBMS follows the regulations of normalization, which splits a

Notes

e
relation whilst any of its attributes is having redundancy in values. Normalization is
a mathematically rich and medical technique that reduces information redundancy.

in
●● Consistency − consistency is a state wherein every relation in a database stays
constant. There exist techniques and strategies, that may come across strive
of leaving database in inconsistent state. A DBMS can offer extra consistency

nl
compared to earlier forms of information storing applications like document-
processing systems.
●● Question language − DBMS is ready with query language, which makes it more

O
green to retrieve and manipulate statistics. A person can apply as many and as
special filtering alternatives as required to retrieve a set of records. Historically it
was now not feasible where record-processing machine changed into used.
●● Multiuser and concurrent get entry to − DBMS helps multi-person environment and

ty
lets in them to get admission to and manipulate facts in parallel. Although there
are regulations on transactions while users try to cope with the identical statistics
object, but users are constantly blind to them.

si
●● A couple of views − DBMS gives a couple of perspectives for specific users. A
person who’s within the income department can have a one-of-a-kind view
of database than someone operating within the production department. This
r
selection enables the users to have a pay attention view of the database
ve
consistent with their necessities.
●● Security − capabilities like multiple perspectives provide security to a point where
customers are not able to get right of entry to statistics of different customers and
departments. DBMS gives strategies to impose constraints even as getting into
ni

statistics into the database and retrieving the equal at a later degree.
DBMS offers many special ranges of protection features, which enables multiple
users to have extraordinary perspectives with specific features. As an instance, a
U

person within the income department cannot see the information that belongs to the
purchase branch. Additionally, it could additionally be controlled how lots information
of the sales department must be exhibited to the consumer. Considering a DBMS isn’t
stored at the disk as conventional report structures, it’s far very hard for miscreants to
ity

interrupt the code.

Users
An average DBMS has users with distinctive rights and permissions who use it for
m

distinctive purposes. A few customers retrieve records and a few returned it up. The
users of a DBMS may be widely categorised as follows −

●● Administrators − directors preserve the DBMS and are liable for administrating the
)A

database. They are accountable to look after its utilization and by whom it have
to be used. They devise get entry to profiles for users and apply barriers to keep
isolation and pressure security. Directors additionally look after DBMS resources
like system license, required tools, and other software program and hardware
(c

associated renovation.
●● Designers − designers are the group of individuals who simply paintings at the
designing a part of the database. They hold a near watch on what information
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 9

should be saved and in what format. They perceive and layout the entire set of
Notes

e
entities, family members, constraints, and perspectives.
●● Give up customers − give up users are individuals who absolutely obtain the

in
advantages of getting a DBMS. Quit customers can variety from simple visitors
who be aware of the logs or market quotes to state-of-the-art users inclusive of
enterprise analysts.

nl
O
ty
Advantages of DBMS
Database is a group of related information and statistics is a group of information

si
and figures that can be processed to provide records. Normally statistics represents
recordable records. Facts aids in generating facts, that’s based totally on data. For
example, if we have statistics about marks received via all students, we are able to then

r
conclude about toppers and average marks. A database management gadget stores
statistics in this sort of way that it turns into easier to retrieve, manipulate, and convey
ve
records. Following are the important characteristics and packages of DBMS.

●● ACID − DBMS follows the concepts of atomicity, consistency, isolation, and

sturdiness (generally shortened as acid). Those concepts are applied on
transactions, which control statistics in a database. Acid residences assist the
ni

database live healthy in multi-transactional environments and in case of failure.

●● Multiuser and concurrent get admission to − DBMS helps multi-person
U

surroundings and permits them to access and control statistics in parallel.

Although there are regulations on transactions whilst customers try and manage
the equal information item, but customers are continually unaware of them.
●● A couple of perspectives − DBMS offers more than one views for exclusive users.
ity

A consumer who’s in the sales branch will have an exclusive view of database
than a person running within the manufacturing department. This feature enables
the users to have a concentrate view of the database in keeping with their
requirements.
m

●● Safety − functions like a couple of views offer protection to some extent where
customers are not able to get admission to facts of other users and departments.
DBMS gives techniques to impose constraints at the same time as entering data
)A

into the database and retrieving the equal at a later stage. DBMS gives many
exclusive levels of protection capabilities, which enables multiple customers to
have one of a kind views with different capabilities. As an example, a person within
the income department can’t see the information that belongs to the acquisition
department. Moreover, it could additionally be managed how plenty statistics of
(c

the income branch have to be displayed to the consumer. In view that a DBMS
is not saved on the disk as traditional document structures, it is very tough for
miscreants to break the code.
Amity Directorate of Distance & Online Education
 10 Introduction to Data Base Management Systems

A database management system has following characteristics:

Notes

e
◌◌ Data stored into tables: data is in no way directly saved into the database.
Data is saved into tables, created in the database. DBMS additionally allows

in
to have relationships between tables which makes the statistics greater
meaningful and related. You can without problems apprehend what form
of information is stored where by searching at all of the tables created in a

nl
database
◌◌ Reduced redundancy: within the modern-day international hard drives are
very cheap, but earlier while tough drives were too pricey, pointless repetition

O
of records in database became a massive problem. But DBMS follows
normalisation which divides the records in such a way that repetition is
minimum.
◌◌ Information consistency: on stay statistics, I. E. Records this is being

ty
continuously updated and delivered, preserving the consistency of facts can
end up a mission. But DBMS handles it all with the aid of itself.
◌◌ Aid more than one consumer and concurrent access: DBMS allows multiple

si
users to work on it (replace, insert, and delete information) at the identical
time and still manages to keep the statistics consistency.
◌◌ Question language: DBMS offers users with a simple question language,
r
using which facts may be without difficulty fetched, inserted, deleted and
ve
updated in a database.
◌◌ Safety: the DBMS also looks after the safety of information, shielding the
statistics from un-approved get admission to. In a normal DBMS, we can
create consumer accounts with one of a kind get entry to permissions, the
ni

usage of which we can without problems comfortable our facts with the aid of
proscribing user access.
◌◌ DBMS supports transactions, which lets in us to higher manage and manage
U

records integrity in actual global packages wherein multi-threading is

significantly used.

1.1.3 Database System Vs File System

ity

Document system
A file machine is a way of arranging the documents in a storage medium like a
tough disk, pen force, DVD, and so forth. It helps you to organize the facts and permits
m

smooth retrieval of documents while they may be required. It by and large consists of
different forms of files like mp3, mp4, txt, doc, and so forth which can be grouped into
directories. A document system permits you to handle the way of analysing and writing
data to the storage medium. It’s far without delay hooked up into the pc with the running
)A

structures consisting of home windows and Linux.

Functions of a file system

●● It lets you keep facts in a group of files.
(c

●● Files data are dependent on each other.

●● C/C++ languages have been used to layout the documents.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 11

●● Shared report device aid

Notes

e
●● Speedy document system restoration.

in
DBMS:
Database management gadget (DBMS) is a software program for storing and
retrieving person’s facts whilst thinking about suitable safety features. It includes a set

nl
of programs that control the database. The DBMS accepts the request for facts from
a utility and instructs the DBMS engine to provide the precise statistics. In massive
structures, a DBMS helps customers and different 1/3-birthday celebration software to

O
store and retrieve facts. Features of DBMS

◌◌ A user-handy catalogue of statistics

◌◌ Transaction help

ty
◌◌ Concurrency manipulate with healing offerings
◌◌ Authorization services
◌◌ The price of records is the equal at all locations.

si
◌◌ Offers help for records communication
◌◌ Unbiased application offerings
◌◌
r
Permits multiple customers to share a record at the identical time
ve
Basis File System DBMS
Structure File system is a software that DBMS is a software for managing
maganes and organizes the files the database.
in a storage medium within a
ni

computer.
Data Redundant data can be present In DBMS there is no redundant data.
Redundancy in a file system.
U

Backup and It doesn’t provide backup and It provides backup and recovery of
Recovery recovery of data if it is lost. data evene if it is lost.
Query There is no efficient query Efficient query processing is there in
ity

processing processing in file system DBMS

Constistency There is less data consistency in There is more data consistency
file system. because of the process of
normalization.
m

Compiexity It is less compiex as compared It has more complisity n handling as

to DBMS compared to file system
Security File systems provide less DBMS has more security
)A

Constraints security in comparison to DBMS mechanisms as compared to file

system.
Cost It is less expensive than DBMS It has a comaratively higher cost
that a file system
Data There is no data independence In DBMS data independence exists
(c

independence

Amity Directorate of Distance & Online Education

 12 Introduction to Data Base Management Systems

1.1.4 Components of DBMS

Notes

e
●● Software program. DBMS is more often than not a software program gadget
that may be taken into consideration as a control console or an interface to have

in
interaction with and control databases. The interfacing also spreads across
actual-global bodily structures that make contributions information to the backend
databases. The os, networking software, and the hardware infrastructure is

nl
worried in creating, getting access to, coping with, and processing the databases.
●● Facts. DBMS includes operational facts, access to database statistics and
metadata as an aid to perform the essential functionality. The statistics may

O
include files with along with index files, administrative data, and information
dictionaries used to represent statistics flows, possession, shape, and
relationships to other facts or items.
●● Tactics. At the same time as not a part of the DBMS software, processes may be

ty
taken into consideration as commands on using DBMS. The documented pointers
help users in designing, enhancing, dealing with, and processing databases.
●● Database languages. These are additives of the DBMS used to get admission

si
to, alter, shop, and retrieve statistics gadgets from databases; specify database
schema; manage user get entry to; and carry out different associated database
control operations. Sorts of DBMS languages consist of information definition
r
language (ddl), facts manipulation language (dml), database get right of entry to
ve
language (dal) and statistics manipulate language (dcl).
●● Query processor. As an essential factor of the DBMS, the query processor
acts as an intermediary among users and the DBMS facts engine as a way to
communicate question requests. When users input an education in sq. Language,
ni

the command is finished from the high-stage language education to a low-stage

language that the underlying system can recognize and method to carry out the
proper DBMS functionality. Further to coaching parsing and translation, the query
U

processor additionally optimizes queries to make sure speedy processing and

accurate outcomes.
●● Runtime database supervisor. A centralized control element of DBMS that
handles functionality related to runtime facts that is usually used for context-
ity

based database access. This factor exams for person authorization to request the
query; approaches the authorised queries; devises an most effective method for
question execution; helps concurrency so that multiple users can simultaneously
paintings on same databases; and ensures integrity of information recorded into
m

the databases.
●● Database supervisor: In contrast to the runtime database manager that handles
queries and information at runtime, the database supervisor performs DBMS
)A

functionality related to the information inside databases. Database supervisor lets

in a fixed of commands to carry out one-of-a-kind DBMS operations that include
creating, deleting, backup, restoring, cloning, and other database upkeep duties.
The database supervisor will also be used to update the database with patches
from carriers.
(c

●● Database engine. This is the middle software thing in the DBMS answer that
performs the centre capabilities related to facts garage and retrieval. A database

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 13

engine is also handy via APIs that allow customers or apps to create, examine,
Notes

e
write, and delete statistics in databases.
●● Reporting. The report generator extracts beneficial statistics from DBMS

in
documents and displays it in established layout based totally on described specs.
This information may be used for further analysis, selection making, or commercial
enterprise intelligence.

nl
1.1.5 Advantages of DBMS
Following are the advantages of DBMS system:

O
1. Information redundancy
Unlike traditional report-system garage, statistics redundancy in DBMS may be very
less or not present. Records redundancy takes place while the same information are

ty
stored unnecessarily at extraordinary locations. Data redundancy is decreased or
removed in DBMS because all facts are saved at a centralized place in preference
to being created by man or woman customers and for every software. For e. G: utility

si
a and alertness b have the equal consumer marvel, and we need to shop non-public
records approximately the user including call, age, deal with, date of start and so on.
Now not to mention, this person has also get entry to one-of-a-kind software, so in
conventional file-based gadget, there’s a need to preserve separate file machine for
r
every of the application to shop consumer’s facts whilst in DBMS method, there will
ve
be simply one centralized place where records can be down streamed to the distinct
software as and while needed.
2. Records inconsistency
ni

In traditional file gadget storage, the changes made by using one consumer in
one utility doesn’t update the changes in different application, given each have the
equal set of details. While this isn’t the case with DBMS structures as there is an
unmarried repository of data that is described once and is accessed with the aid of
U

many customers, and facts are consistent.

3. Data sharing
Fact Sharing is the number one gain of database control structures. DBMS device
ity

allows customers and packages to share statistics with multiple applications and
users. Statistics are stored in one or more servers in the community and that there
may be a few software locking mechanism that forestalls the equal set of facts from
being modified by humans at the identical time. Even as the record device doesn’t
m

have this capability.

4. Information searching
)A

Searching and retrieving of facts may be very smooth in DBMS structures. The want
to put in writing separate packages for each of the search is removed as in the case
with a traditional file-primarily based method. In DBMS, we can write small queries
to look for a couple of statistics at a time from the statistics from db servers. Five.
Statistics protection
(c

DBMS systems provide a robust framework to protect information privateness and

protection. DBMS guarantees that simplest legal users have to get the right of entry
to information and there is a mechanism to outline get entry to privileges.
Amity Directorate of Distance & Online Education
 14 Introduction to Data Base Management Systems

5. Information concurrency
Notes

e
In DBMS, statistics are saved in a single or more servers inside the network and that
there’s some software program locking mechanism that forestalls the identical set of

in
statistics from being modified via two human beings on the identical time.
6. Records integration
Statistics integration is a system of combining the facts living at distinctive locations

nl
and gift the consumer with a unified view of records. DBMS systems permit facts
integration with plenty feasibility.
7. Information get admission to at the same time as in conventional file-primarily based

O
approach, it might take hours to look for very particular data that is probably needed
inside the context of some business emergency, while DBMS reduces this time to 3
seconds. This is a first-rate gain of DBMS due to the fact we will write small queries

ty
as a way to seek the database for you and it’ll retrieve the records within the quickest
manner viable due to its built in looking operations.
8. Selection making

si
Stepped forward statistics sharing and higher-controlled information permit business
to make quality enterprise decisions in order to promote the boom of the organization.
9. Records backup and restoration
r
That is some other advantage of DBMS as it affords a robust framework for statistics
ve
backup, users aren’t required to again up their data periodically and manually, it’s far
mechanically taken care via DBMS. Furthermore, in case of a server crash, DBMS
restores the database to its preceding situation.
ni

10. Records migration

There are a few facts which are access very frequently whilst there are few that
aren’t. So, DBMS provides the capability to get admission to the frequently accessed
U

records as fast as feasible.

11. Low preservation price
Even though a DBMS structure is probably steeply-priced at the time of purchase
ity

however their maintenance involves a very minimum cost.

12. Records loss
It is almost removed with DBMS, one could maintain statistics for thousands of years,
furnished we don’t see the doomsday. Facts safety and very low garage value (as
m

compared to our preceding generations) within the current century cut any possibility
of facts loss.
13. Facts atomicity
)A

An atomic transaction is one wherein all of the database moves occur or none of
them do. It’s far the obligation of DBMS to store a complete transaction within the
database. If any transaction is partially finished, then it rolls backs them. For e. G:
if we make a web buy, money is deducted from our account while if the purchase is
(c

by some means failed, then no money is deducted or if it receives deducted, it gets

lower back within few days.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 15

Notes

e
in
nl
Dangers of DBMS

O
1. Expanded cost
A) Cost of hardware and software program
To keep a huge amount of facts, one desires a massive amount of area. Moreover,

ty
it’ll require greater memory and rapid processing energy to run the DBMS. So,
expensive hardware and software could be wanted that can offer these kind of
facilities. As an end result, the antique report-based system desires to be upgraded.

si
These sophisticated hardware and software program require maintenance which is
very pricey. DBMS requires an excessive preliminary investment for hardware and
software program. A full-size investment primarily based upon the size and capability
r
of the agency is required. Additionally, the company has to pay concurrent annual
maintenance fees.
ve
B) Multiplied value of workforce
DBMS personnel consists of database administrators, software programmers,
database designers, machine analysts, and upkeep personals. All of those are
ni

fantastically knowledgeable and experienced inside the discipline of DBMS. To get

this degree of skilled human beings, one needs to spend a variety of cash. Also,
schooling of DBMS could be very high priced.
U

C) Cost of information conversion

Records conversion may additionally require at any time and the company has to
take this step. Its miles unbelievable that data conversion price is more than the
ity

prices of DBMS hardware and gadget blended. Educated staff is wanted to transform
facts to the brand new gadget. It’s far a key reason that most businesses are still
operating on their old DBMS due to the excessive cost of statistics conversion.
2. Complexity
m

A DBMS fulfils plenty of requirements and it solves many issues related to

databases. But a lot of these functionalities make DBMS an incredibly complicated
software. Builders, designers, dba, and cease-users of database have to have whole
)A

capabilities of DBMS in the event that they want to apply it properly. In the event that
they don’t recognize this complicated gadget then it could reason lack of facts or
database failure. As it calls for plenty of management, so its complexity turns into an
issue and a very big disadvantage of DBMS.
3. Technical staff requirement
(c

Companies have many personnel running for them and these personnel can perform
many other obligations too that are not of their area however it is almost not possible

Amity Directorate of Distance & Online Education

 16 Introduction to Data Base Management Systems

for them to paintings on DBMS. A devoted crew of technical workforce is needed who
Notes

e
can manage DBMS and as an end result, the organization have to pay good-looking
income to them too.

in
4. Database failure facts is the key for any employer, if data is misplaced then the
complete corporation will disintegrate. And as we realize that in DBMS, all of the
files are saved in an unmarried database so the chances of database failure become

nl
more. Any unintentional failure of components may additionally purpose loss of
treasured data. This is simply a massive query mark for big companies and they are
constantly working to clear up this issue.

O
5. Huge length DBMS is made to handle extraordinarily big facts and queries, but due
to its complexity, DBMS has grown to be large in size. Also, it turns into bigger in size
as statistics is fed into it. As an end result, it calls for masses of space and memory
to run its software efficiently.

ty
6. Foreign money upkeep performance is the core need for any software, so it will
become should for DBMS to be modern-day. Because efficiency can best be finished
if your system is present day. Common updates, security features, and the brand new

si
patches have to be achieved. As new threats come every day, so DBMS requires to
be up to date on the everyday basis. DBMS must be up to date in keeping with the
modern state of affairs.
7. r
Tough backup and recuperation database is processed and dealt with through
ve
several users concurrently, so it becomes very tough to get the precise country of
database at the time of failure. It creates many troubles and customers get confused
about what to do next. So, it is very tough to take backup and recover the database
as just one utility application may additionally pose severe troubles.
ni

8. Confidentiality, privateness, and security in DBMS, information is made to be had to

customers from faraway locations, and because it’s miles centralized so possibilities
of abuse are more regularly than in a file-based totally system. If information from the
U

information centre receives corrupted then each consumer of the organization may
be in large hassle.
9. Statistics integrity as database is getting up to date in actual-time, so it’s far required
that information remains accurate in the course of operations. For that a few technical
ity

safeguards are necessary. Several customers trying to update the facts create the
primary danger to data integrity.
10. Decrease efficiency there are numerous software and alertness packages which can
be made only for one precise problem. But DBMS is a multi-use software that is used
m

for many purposes and as an end result, it turns into much less green. Specialised
software is greater green due to the fact they’re produced and optimized for one
hassle.
)A

11. Extended vulnerability as DBMS is able to many things because it is centralized,

however at the same time centralization increases vulnerability. The whole system
shuts down due to the failure of an unmarried component. That is very important
disadvantage of DBMS for those agencies which might be completely depending on
(c

databases.
12. Overall performance conventional documents machine changed into very well for
small corporations as it gave a brilliant overall performance. But DBMS offers
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 17

terrible overall performance for small-scale corporations as its speed is sluggish and
Notes

e
because of that it is nicely applicable for massive firms.
13. Speed as within the previous point, we’ve discussed that DBMS overall performance

in
is a problem with small corporations. Because of which, DBMS runs at very slow
pace for small scale organizations. DBMS is simply appropriate for big companies
and businesses.

nl
1.1.6 Types of Database System

O
ty
r si
ve
1) Centralized database
It is the form of database that shops records at a centralized database gadget. It
comforts the users to get admission to the stored records from exclusive locations via
ni

numerous applications. These programs incorporate the authentication technique to

permit users get entry to information securely. An example of a centralized database
can be primary library that consists of an important database of every library in a
U

college/college.

Advantages of centralized database

◌◌ It has decreased the chance of statistics control, I. E., manipulation of
ity

statistics will now not affect the centre records.

◌◌ Information consistency is maintained because it manages records in an
imperative repository.
◌◌ It affords higher data fine, which enables businesses to establish information
m

standards.
◌◌ It’s far less steeply-priced because fewer providers are required to deal with
the records units.
)A

Disadvantages of centralized database

◌◌ The scale of the centralized database is massive, which increases the
reaction time for fetching the statistics.
(c

◌◌ It is not clean to update such an intensive database system.

◌◌ If any server failure occurs, entire statistics can be lost, which may be a big
loss.
Amity Directorate of Distance & Online Education
 18 Introduction to Data Base Management Systems

2) Disbursed database
Notes

e
In contrast to a centralized database gadget, in allotted systems, information
is sent amongst special database systems of a company. These database systems

in
are related via communication links. Such links help the end-users to get entry to the
information without difficulty.

Examples of the Distributed database are Apache Cassandra, HBase, Ignite, etc.

nl
O
ty
◌◌ r si
Homogeneous ddb: those database structures which execute at the identical
ve
operating gadget and use the equal utility method and convey the identical
hardware devices.
◌◌ Heterogeneous ddb: those database systems which execute on exclusive
working systems underneath distinctive utility tactics, and contains exclusive
ni

hardware devices. Benefits of disbursed database

◌◌ Modular development is feasible in a disbursed database, I. E., the machine
may be accelerated through which include new computers and connecting
U

them to the disbursed machine.

◌◌ One server failure will not affect the whole statistics set. Three)

Relational Database
ity

This database is based at the relational facts model, which shops records inside
the shape of rows (tuple) and columns (attributes), and collectively forms a table
(relation). A relational database uses sq. For storing, manipulating, in addition to
preserving the data. E. F. Codd invented the database in 1970. Every desk inside the
m

database includes a key that makes the statistics unique from others. Examples of
relational databases are MySQL, Microsoft sq. Server, oracle, and so on. Houses of
relational database there are following four normally recognized homes of a relational
)A

version known as acid residences, in which:

A method atomicity: this ensures the records operation will whole either with
fulfilment or with failure. It follows the ‘all or nothing’ strategy. As an instance, a
transaction will both be dedicated or will abort. C way consistency: if we perform
any operation over the facts, its value earlier than and after the operation should be
(c

preserved. As an example, the account balance before and after the transaction must

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 19

be accurate, I. E., it should continue to be conserved. I manner isolation: there can be

Notes

e
concurrent users for gaining access to facts on the same time from the database. As
a consequence, isolation between the records must stay remoted. For example, when
multiple transactions arise on the identical time, one transaction effects need to no

in
longer be seen to the other transactions in the database.

D means durability: it guarantees that after it completes the operation and commits

nl
the facts, facts modifications need to continue to be permanent.

3) Non SQL database

O
Non-SQL now not most Effective Square is a form of database this is used for
storing a huge variety of facts sets. It isn’t always a relational database as it shops
statistics no longer simplest in tabular form but in numerous exclusive approaches.
It got here into existence whilst the call for constructing modern-day applications

ty
improved. Accordingly, No-SQL offered a huge sort of database technologies in
response to the needs.

r si
ve
ni
U
ity

A. Key-value garage: its miles the handiest form of database storage in which it stores
m

each single object as a key (or characteristic call) holding its price, collectively.
B. Record-oriented database: a kind of database used to keep records as json-like
document. It allows developers in storing information by using the identical document-
)A

version layout as used in the application code.

C. Graph databases: it’s far used for storing massive quantities of statistics in a graph-
like structure. Maximum usually, social networking websites use the graph database.
D. Huge-column stores: it is just like the information represented in relational databases.
(c

Right here, records is stored in massive columns together, instead of storing in rows.

Amity Directorate of Distance & Online Education

 20 Introduction to Data Base Management Systems

Benefits of noSQL database

Notes

e
◌◌ It allows true productivity in the application development because it isn’t
always required to shop information in an established format.

in
◌◌ It’s far a higher option for dealing with and dealing with massive statistics sets.
It affords excessive scalability.
◌◌ Users can quickly access records from the database thru key-price.

nl
4) Cloud database
A form of database where facts is stored in a virtual environment and executes

O
over the cloud computing platform. It gives users with diverse cloud computing services
(saas, paas, iaas, and many others.) for gaining access to the database. There are
numerous cloud platforms, however the best alternatives are:

ty
◌◌ Amazon net offerings (AWS)
◌◌ Microsoft azure
◌◌ Kamatera

si
◌◌ Phonixnap
◌◌ Sciencesoft
◌◌ Google cloud sq., and so on.
r
ve
5) Object-orientated databases
The form of database that uses the object-based totally information model
approach for storing facts within the database gadget. The records is represented and
saved as objects which can be much like the objects used within the object-orientated
ni

programming language.

6) Hierarchical databases
U

Its miles the kind of database that stores facts inside the form of discern-children
relationship nodes. Right here, it organizes statistics in a tree-like structure. Facts get
stored within the form of data that are linked through hyperlinks. Each infant record
within the tree will include only one parent. Then again, each determine file can have
ity

more than one toddler information.

m
)A
(c

Hierarchical Database

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 21

7) Network databases
Notes

e
It’s far the database that typically follows the community statistics model. Right
here, the representation of facts is in the shape of nodes related via hyperlinks among

in
them. Not like the hierarchical database, it allows every file to have multiple youngsters
and determine nodes to form a generalized graph shape.

8) Private database

nl
Accumulating and storing statistics at the user’s system defines a personal
database. This database is largely designed for an unmarried consumer.

O
Advantage of personal database
◌◌ It is straightforward and smooth to address.
◌◌ It occupies less storage space as it is small in length.

ty
9) Operational database:
The kind of database which creates and updates the database in real-time. It is

si
basically designed for executing and managing the day by day information operations
in several organizations. For example, an organisation uses operational databases for
handling per day transactions.
r
ve
10) Employer database
Huge groups or enterprises use this database for dealing with a huge amount of
information. It allows agencies to increase and enhance their performance. This kind of
database permits simultaneous access to customers. Blessings of enterprise database:
ni

◌◌ Multi methods are supportable over the organisation database.

◌◌ It allows executing parallel queries on the machine.
U

Check your Understanding

A. Fill in the Blanks

1. A ________ is a software that allows creation, definition and manipulation of

ity

database, allowing users to store, process and analyse data easily.

2. ___________ occurs when the same data are stored unnecessarily at different
places.
3. A __________ enables you to handle the way of reading and writing data to the
m

storage medium.
4. DBMS provides users with a simple ___________, using which data can be easily
fetched, inserted, deleted and updated in a database.
)A

5. __________ is a data structure technique to efficiently retrieve records from the

database files

B. State True or False

(c

1. A table can only have one primary key.

2. Database is a collection of random data.

Amity Directorate of Distance & Online Education

 22 Introduction to Data Base Management Systems

3. DBMS follows Normalisation which divides the data in such a way that repetition is
Notes

e
minimum.
4. DBMS doesn’t provide backup and recovery of data if it is lost.

in
5. Homogeneous DDB are those database systems which execute on different
operating systems under different application procedures, and carries different
hardware devices.

nl
Summary
●● Database control gadget or DBMS in quick refers to the generation of storing

O
and retrieving user’s facts with utmost efficiency along with appropriate security
features. Statistics is not anything but records and records saved or free flowing
over a community, commonly it’s uncooked and unprocessed. For instance: while
you go to any website, they may keep you ip address, this is information, in return

ty
they might upload a cookie in your browser, marking you which you visited the
internet site, that is records, your call, it is records, your age, it’s records.
●● Ddl is brief call of facts definition language, which offers with database schemas

si
and outlines, of how the statistics need to reside within the database. It includes
square statements like modify, drop, truncate, rename and so forth. Dml is short
name of data manipulation language which offers with data manipulation and
r
consists of maximum not unusual sq. Statements such select, insert, update,
ve
delete, etc., and its miles used to shop, adjust, retrieve, delete and replace records
in a database.
●● A current DBMS has the subsequent traits −
●● Actual-global entity − a modern DBMS is greater practical and uses real-
ni

international entities to design its structure. It makes use of the behaviour and
attributes too. For instance, a school database may also use college students as
an entity and their age as an attribute.
U

●● Relation-primarily based tables − DBMS allows entities and family members

amongst them to form tables. A person can recognize the architecture of a
database just with the aid of searching at the desk names.
ity

●● Isolation of records and alertness − a database gadget is totally special than its
facts. A database is an active entity, whereas information is stated to be passive,
on which the database works and organizes. DBMS also stores metadata, which is
records about information, to ease its personal technique.
m

●● Less redundancy − DBMS follows the policies of normalization, which splits a

relation when any of its attributes is having redundancy in values. Normalization is
a mathematically rich and scientific method that reduces records redundancy.
)A

●● Consistency − consistency is a state wherein every relation in a database remains

steady. There exist techniques and that may locate attempt of leaving database
in inconsistent state. A DBMS can offer greater consistency compared to earlier
types of facts storing packages like report-processing structures.
(c

●● Question language − DBMS is equipped with question language, which makes

it greater green to retrieve and manipulate statistics. A consumer can observe
as many and as one of a kind filtering alternatives as required to retrieve a set

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 23

of information. Traditionally it became not possible wherein report-processing

Notes

e
machine turned into used.
●● Multiuser and concurrent access − DBMS supports multi-consumer surroundings

in
and lets in them to get right of entry to and manage facts in parallel. Even though
there are restrictions on transactions when users try to handle the identical
information object, however customers are always ignorant of them.

nl
●● A couple of perspectives − DBMS gives multiple perspectives for exceptional
customers. A user who is inside the sales department may have an exceptional
view of database than someone running in the manufacturing branch. This feature

O
permits the customers to have a concentrate view of the database in line with their
requirements.
●● Protection − features like multiple perspectives offer protection to some extent
where customers are unable to get entry to facts of different customers and

ty
departments. DBMS gives techniques to impose constraints at the same time as
coming into facts into the database and retrieving the identical at a later degree.
DBMS gives many specific ranges of security features, which allows more than

si
one users to have different perspectives with distinctive features. For instance,
a consumer inside the sales branch cannot see the records that belongs to the
acquisition department. Additionally, it could additionally be controlled how a good
r
deal facts of the income branch must be exhibited to the user. Due to the fact that
a DBMS is not saved on the disk as conventional file systems, it is very tough for
ve
miscreants to interrupt the code.
◌◌ Functions of a report machine
◌◌ It lets you shop information in a group of documents.
ni

◌◌ Documents information are dependent on each other.

◌◌ C/C++ languages were used to design the documents.
◌◌ Shared document system help
U

◌◌ Speedy file gadget recovery. Functions of DBMS

◌◌ A consumer-available catalogue of data
◌◌ Transaction help
ity

◌◌ Concurrency manipulate with recuperation services

◌◌ Authorization offerings
◌◌ The fee of statistics is the equal in any respect locations.
m

◌◌ Gives support for information communiqué

◌◌ Unbiased utility offerings Permits a couple of customers to percentage a
report on the identical time
)A

◌◌ Distinction between file system vs. DBMS

◌◌ Blessings of DBMS gadget
1. Information redundancy: in contrast to conventional report-gadget garage, statistics
redundancy in DBMS is very less or no longer present. Records redundancy occurs
(c

when the equal statistics are stored unnecessarily at distinctive places. Records
redundancy is reduced or removed in DBMS because all records are stored at

Amity Directorate of Distance & Online Education

 24 Introduction to Data Base Management Systems

a centralized area in place of being created by using person users and for each
Notes

e
application.
2. Information inconsistency: in conventional record system garage, the modifications

in
made by way of one person in one utility doesn’t update the changes in different
application, given each have the same set of info. Even as this isn’t the case with
DBMS systems as there is a single repository of statistics this is defined as soon as

nl
and is accessed through many users, and data are consistent.
3. Records sharing: facts sharing is the number one advantage of database management
structures. DBMS machine permits customers and programs to percentage statistics

O
with multiple applications and customers. Statistics are saved in one or extra servers
within the network and that there’s a few software locking mechanism that forestalls
the equal set of information from being modified by humans on the equal time. While
the file machine doesn’t have this functionality.

ty
4. Facts looking: looking and retrieving of facts may be very easy in DBMS systems.
The need to write down separate programs for each of the hunt is eliminated as
within the case with a traditional file-based method. In DBMS, we can write small

si
queries to look for multiple data at a time from the records from db servers.
5. Records protection: DBMS structures provide a robust framework to defend records
privacy and security. DBMS guarantees that only legal customers have get right
r
of entry to statistics and there may be a mechanism to outline get right of entry to
ve
privileges.
6. Statistics concurrency : in DBMS, facts are stored in one or extra servers within the
network and that there’s a few software program locking mechanism that forestalls
the equal set of information from being changed by means of two humans on the
ni

equal time.
7. Records integration: facts integration is a process of mixing the information residing
at different places and present the user with a unified view of statistics. DBMS
U

systems permit records integration with an awful lot feasibility.

8. Records get right of entry to: at the same time as in traditional report-primarily based
method, it would take hours to search for very unique records that is probably needed
ity

inside the context of a few enterprise emergency, while DBMS reduces this time to
three seconds. That is an awesome gain of DBMS due to the fact we are able to write
small queries with a view to search the database for you and it’ll retrieve the records
within the quickest manner viable because of its inbuilt searching operations.
m

9. Choice making: improved facts sharing and better-controlled facts permit commercial
enterprise to make great commercial enterprise choices so that it will promote the
increase of the organisation.
)A

10. Facts backup and recuperation : that is any other gain of DBMS because it offers
a strong framework for facts backup, customers aren’t required to again up their
statistics periodically and manually, it’s miles robotically taken care via DBMS.
Furthermore, in case of a server crash, DBMS restores the database to its previous
situation.
(c

11. Statistics migration: there are a few statistics which can be access very often even

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 25

as there are few that aren’t. So, DBMS affords the capability to get entry to the often
Notes

e
accessed information as quickly as possible.
12. Low protection cost: although DBMS systems is probably luxurious at the time of

in
purchase but their upkeep involves a completely minimum cost.
13. Statistics loss is nearly eliminated: with DBMS, you may keep records for heaps of
years, supplied we don’t see the doomsday. Statistics security and very low storage

nl
cost (as compared to our previous generations) in the modern-day century cut any
possibility of records loss.
14. Information atomicity: an atomic transaction is one in which all of the database

O
movements occur or none of them do. It is the duty of DBMS to save a complete
transaction inside the database. If any transaction is partly finished, then it rolls
backs them. Styles of database:

ty
●● Centralized database
It is the form of database that shops statistics at a centralized database gadget.
It comforts the users to get right of entry to the stored facts from distinct places via

si
several packages. Those applications comprise the authentication system to permit
users access facts securely. An example of a centralized database can be primary
library that contains a critical database of each library in a university/university. Benefits
of centralized database
r
ve
●● It has reduced the risk of information control, I. E., manipulation of data will not
affect the centre facts.
●● Records consistency is maintained because it manages statistics in a primary
repository.
ni

●● It provides better information satisfactory, which permits businesses to set up

records standards.
●● IT is much less expensive because fewer providers are required to handle the
U

facts sets. Negative aspects of centralized database

●● The size of the centralized database is big, which will increase the response time
for fetching the records.
ity

●● It isn’t smooth to replace such an intensive database system.

●● If any server failure occurs, complete data could be misplaced, which could be a
large loss.
m

●● distributed database
Unlike a centralized database system, in distributed structures, records is shipped
among exclusive database structures of a business enterprise. These database
)A

systems are linked through conversation links. Such links help the stop-users to get
right of entry to the statistics without difficulty. Examples of the disbursed database are
apache Cassandra, hbase, ignite, and so on.

●● Homogeneous ddb: those database structures which execute on the identical

(c

running device and use the identical application manner and carry the equal
hardware gadgets.
●● Heterogeneous ddb: the ones database structures which execute on distinctive
Amity Directorate of Distance & Online Education
 26 Introduction to Data Base Management Systems

running systems under distinctive application techniques, and carries specific

Notes

e
hardware gadgets. Advantages of disbursed database
●● modular development is feasible in a allotted database, I. E., the device may

in
be multiplied via consisting of new computer systems and connecting them
to the allotted machine.one server failure will now not have an effect on the
whole records set. Three) relational database :this database is based on the

nl
relational records version, which stores information in the form of rows(tuple) and
columns(attributes), and collectively bureaucracy a table(relation). A relational
database uses square for storing, manipulating, in addition to keeping the data. E.
F. Codd invented the database in 1970. Each table within the database carries a

O
key that makes the records specific from others. Examples of relational databases
are MySQL, Microsoft square server, oracle, and so forth. Properties of relational
database

ty
There are following four typically regarded residences of a relational model referred
to as acid residences, in which:

A method atomicity: this guarantees the facts operation will entire either with

si
achievement or with failure. It follows the ‘all or nothing’ approach. For instance, a
transaction will both be committed or will abort.

C means consistency: if we carry out any operation over the information, its fee
r
before and after the operation ought to be preserved. As an instance, the account
ve
balance earlier than and after the transaction must be correct, I. E., it have to remain
conserved.

I way isolation: there can be concurrent customers for having access to

information at the identical time from the database. Accordingly, isolation between the
ni

facts should stay remoted. For instance, when more than one transactions arise on the
identical time, one transaction results have to not be seen to the opposite transactions
in the database.
U

D way sturdiness: it guarantees that after it completes the operation and commits
the statistics, statistics modifications ought to remain everlasting. Four) nosqldatabase:
non-square/not simplest sq. Is a type of database that is used for storing a wide variety
ity

of data sets? It isn’t always a relational database because it stores information not
only in tabular form however in several one of a kind methods. It came into existence
whilst the demand for constructing modern-day applications multiplied. Therefore, nosql
offered an extensive form of database technology in reaction to the demands. We are
able to further divide a nosql database into the subsequent 4 sorts:
m

●● Key-price garage: it’s far the handiest form of database storage where it shops
every unmarried item as a key (or attribute call) conserving its price, together.
)A

B. Report-oriented database: a sort of database used to store data as json-

like report. It facilitates builders in storing information by using the usage of the
equal document-version format as used in the utility code. C. Graph databases:
it is used for storing extensive amounts of facts in a graph-like structure. Most
typically, social networking websites use the graph database. D. Huge-column
(c

shops: its miles much like the statistics represented in relational databases. Here,
information is stored in large columns together, instead of storing in rows.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 27

Benefits of nosql database

Notes

e
●● It enables excellent productiveness within the application development because it
isn’t always required to save information in a based format.

in
●● it’s miles a better choice for handling and dealing with large records units. O it
offers high scalability.
●● Users can speedy access data from the database via key-value.

nl
●● Cloud database: a type of database where statistics is saved in a digital
environment and executes over the cloud computing platform. It gives customers
with various cloud computing services (saas, paas, iaas, and so forth.) for having

O
access to the database. There are numerous cloud platforms, however the first-
class alternatives are:
◌◌ Amazon Net Services(AWS)

ty
◌◌ Microsoft azure
◌◌ kamatera
◌◌ phonixnap

si
◌◌ sciencesoft
◌◌ Google cloud sq., and so on.
◌◌ r
Item-oriented databases: the type of database that makes use of the object-
ve
based information model technique for storing statistics in the database
gadget. The data is represented and saved as objects which can be much like
the objects used inside the item-oriented programming language.
◌◌ Hierarchical databases: it’s far the type of database that shops statistics in
ni

the shape of determine-children relationship nodes. Right here, it organizes

statistics in a tree-like shape. Data get stored in the form of information which
might be linked via hyperlinks. Each baby file in the tree will comprise most
U

effective one determine. However, each parent record will have a couple of
infant information.
◌◌ Community databases: it’s far the database that typically follows the
community facts model. Right here, the representation of records is inside the
ity

form of nodes related via hyperlinks between them. Not like the hierarchical
database, it permits every file to have a couple of youngsters and parent
nodes to shape a generalized graph structure. Nine) non-public database:
accumulating and storing records at the user’s gadget defines a personal
m

database. This database is basically designed for a single consumer. Gain of

private database
◌◌ it is straightforward and smooth to deal with.
)A

◌◌ It occupies much less storage area as it is small in size.

◌◌ Operational database: the kind of database which creates and updates the
database in real-time. It is largely designed for executing and coping with the
everyday data operations in numerous groups. For instance, an organisation
makes use of operational databases for coping with in keeping with day
(c

transactions.
◌◌ Company database: massive businesses or organizations use this database

Amity Directorate of Distance & Online Education

 28 Introduction to Data Base Management Systems

for coping with a large quantity of statistics. It helps businesses to boom and
Notes

e
improve their efficiency. Any such database permits simultaneous get right of
entry to customers.

in
◌◌ Blessings of business enterprise database:
◌◌ Multi strategies are supportable over the agency database.
◌◌ It lets in executing parallel queries at the system.

nl
Activity
1. Prepare a presentation on the topic “Database Management System”.

O
Questions & Exercises
1. What are the characteristics of DBMS?

ty
2. Differentiate between File system and Database system.
3. Describe different types of user in DBMS.
4. What are the components of DBMS?

si
5. Write down the Advantages of DBMS.

Glossary
●● r
DBMS: a DBMS is a software that permits creation, definition and manipulation of
ve
database, permitting customers to save, process and examine data easily. DBMS
presents us with an interface or a device, to perform numerous operations like
creating database, storing information in it, updating statistics, creating tables in
the database and plenty greater. Database: database is a group of inter-related
ni

records which allows in green retrieval, insertion and deletion of statistics from
database and organizes the records inside the shape of tables, perspectives,
and schemas, reviews and so forth. As an instance, college database organizes
U

the information about students, school, and admin body of workers and so forth.
Which facilitates in green retrieval, insertion and deletion of records from it. Ddl is
brief name of statistics definition language, which offers with database schemas
and descriptions, of ways the records need to be living inside the database.
ity

●● Create: to create a database and its objects like (table, index, perspectives, store
system, characteristic, and triggers)
●● Regulate: alters the shape of the prevailing database
m

●● Drop: delete gadgets from the database

●● Truncate: eliminate all statistics from a desk, which include all spaces allotted for
the statistics are eliminated
)A

●● Comment: upload comments to the statistics dictionary

●● Rename: rename an item
DML is brief name of information manipulation language which deals with records
manipulation and includes maximum not unusual sq. Statements such pick, insert,
(c

replace, delete, etc., and it’s miles used to keep, adjust, retrieve, delete and update
information in a database.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 29

◌◌ Select: retrieve statistics from a database

Notes

e
◌◌ Insert: insert facts right into a table
◌◌ Update: updates present statistics inside a table

in
◌◌ Delete: delete all facts from a database table
◌◌ Merge: upset operation (insert or update)
◌◌ Name: name a ply/square or java subprogram

nl
◌◌ Provide an explanation for plan: interpretation of the records get admission to
direction

O
◌◌ Lock desk: concurrency manage
Statistics definition: it facilitates in advent, change and removal of definitions that
define the corporation of records in database. Information updation: it facilitates in
insertion, change and deletion of the actual facts within the database. Facts retrieval:

ty
it allows in retrieval of records from the database which can be utilized by programs for
diverse purposes. Consumer administration: it allows in registering and tracking users,
imposing data security, tracking performance, maintaining records integrity, coping with

si
concurrency manage and convalescing information corrupted through surprising failure.
Table: a desk is a logical structure made up of rows and columns. Rows don’t have any
constant order, so in case you retrieve information you might need to sort the statistics.

r
The order of the columns is the order designated when the desk became created by
means of the database administrator. At the intersection of each column and row is a
ve
specific information item called a value, or, extra exactly, an atomic fee. A desk is known
as with a high-level qualifier of the owner’s consumer identification followed through the
table name, for instance check. Dept. or prod. Dept. There are 3 sorts of tables:
ni

●● A base desk that is created and holds chronic information

●● A temporary table that shops intermediate query outcomes
●● An effects desk that is again while you question tables. Keys: a key’s one or more
U

columns which can be recognized as such within the advent of a desk or index, or
inside the definition of referential integrity. Primary key: a desk can most effective
have one primary key as it defines the entity. There are requirements for a primary
key:
ity

●● It should have a cost, this is, it can’t be null.

●● It ought to be particular, that is, it should have a completely unique index described
on it.
m

Precise key: we already recognize that a number one key need to be specific,
however it’s far feasible to have multiple precise key in a table. In our EMP desk
example, the worker number is described because the number one key and is therefore
)A

specific. If we also had a social protection price in our desk, with any luck that value
could be particular. To assure this, you can create a completely unique index at the
social safety column.

Foreign key: a foreign secret’s a key this is specified in a referential integrity

constraint to make its lifestyles depending on a number one or precise key (figure key)
(c

in some other desk.

Amity Directorate of Distance & Online Education

 30 Introduction to Data Base Management Systems

Reduced redundancy: within the cutting-edge world difficult drives are very cheap,
Notes

e
but earlier when difficult drives were too steeply-priced, useless repetition of statistics
in database changed into a large trouble. However DBMS follows normalisation which
divides the information in such a way that repetition is minimum.

in
Facts consistency: on stay facts, I. E. Facts this is being continuously updated
and added, preserving the consistency of records can end up a task. However, DBMS

nl
handles all of it with the aid of itself.

Further Readings

O
1. Raghu Ramakrishnan and Johannes Gehrke. (2003). Database Management
Systems. Third Edition. McGraw Hill
2. Ramez Elmasri and Shamkant B. Navathe. (2016). Fundamentals of Database
Systems. Seventh Edition. Pearson Education, Inc.

ty
3. C.J. Date. (2004). An Introduction to Database Systems. Pearson Education,
Inc.

si
Check your Understanding-Answers
A. Fill in the Blanks
1. DBMS r
ve
2. Data Redundancy
3. File System
4. Query Language
ni

5. Indexing.
B. True or False
1. True
U

2. False
3. True
ity

4. False
5. False
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 31

Unit -1.2: DBMS Architecture

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Learn about Database Three Level Architecture

nl
●● Know about Schema and instance
●● Understand Database Independence

O
●● Know Database Language and Interfaces
●● Understand about the Data Models
●● Learn about data definitions language and DML

ty
1.2.1 Database Three Level Architecture
DBMS Three Level Architecture Diagram

r si
ve
ni
U
ity
m
)A

In the above diagram,

(c

◌◌ It suggests the structure of DBMS.

Amity Directorate of Distance & Online Education

 32 Introduction to Data Base Management Systems

◌◌ Mapping is the manner of reworking request response between numerous

Notes

e
database ranges of structure.
◌◌ Mapping isn’t always good for small database, because it takes greater time.

in
◌◌ In external / conceptual mapping, DBMS transforms a request on an outside
schema in opposition to the conceptual schema.
◌◌ In conceptual / inner mapping, it is essential to transform the request from the

nl
conceptual to inner stages. This architecture has 3 degrees:
1. Outside stage
2. Conceptual degree

O
3. Internal degree
●● Outside degree

ty
It’s also called view degree. The motive this stage is called “view” is because
numerous customers can view their desired statistics from this degree that is internally
fetched from database with the help of conceptual and inner degree mapping. The user
doesn’t need to recognise the database schema information which include statistics

si
structure, table definition and so forth. User is only involved about information which
is what back lower back to the view stage after it has been fetched from database
(present on the internal degree). Outside degree is the “top degree” of the 3 degree
DBMS architecture. r
ve
●● Conceptual degree
It’s also referred to as logical level. The entire design of the database which
includes relationship amongst statistics, schema of records and many others. Are
ni

described on this degree. Database constraints and safety also are applied in this level
of architecture. This stage is maintained with the aid of dab (database administrator).

●● Internal stage
U

This stage is likewise referred to as bodily stage. This degree describes how the
information is simply stored in the garage gadgets. This stage is likewise liable for
allocating area to the facts. This is the lowest stage of the architecture.
ity

1.2.2 Schema and Instance

Schema:
m

Schema is the general description of the database. The simple structure of the
way the facts may be stored in the database is called schema. A database schema is
the skeleton shape that represents the logical view of the whole database. It defines
how the information is organized and how the family members amongst them are
)A

associated. It formulates all of the constraints that are to be carried out on the facts.
A database schema defines its entities and the connection among them. It includes a
descriptive detail of the database, which may be depicted by way of schema diagrams.
It’s the database designers who design the schema to help programmers apprehend
the database and make it useful. Schema is of three sorts: logical schema, bodily
(c

schema and examine schema. 1. Logical schema – it describes the database designed

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 33

at logical level. 2. Physical schema – it describes the database designed at physical

Notes

e
stage. 3. View schema – it defines the design of the database on the view level.

Example –

in
nl
O
ty
r si
ve
Let’s say a table teacher in our database name school, the teacher table require
the name, DOB, doj in their table so we design a structure as:

Teacher table
ni

Name: String

Doj: date
U

DOB: date

Given below is the schema of the table teacher:

ity

Teacher Subject

Teacher-Name Subject - Name

Teacher - DOB Subject - Code
Teacher - Sub Teacher - Name
Subject - Code
m
)A

Department

Department - Name
Department -ID
Teacher - Name
(c

Amity Directorate of Distance & Online Education

 34 Introduction to Data Base Management Systems

Instances:
Notes

e
Instances s are the gathering of statistics saved at a particular second. The
instances may be modified through certain crud operations as like addition, deletion

in
of statistics. It may be noted that any search query will now not make any kind of
modifications within the instances. A database example is a state of operational
database with statistics at any given time. It includes a snapshot of the database.

nl
Database instances have a tendency to change with time. A DBMS guarantees that
it’s each instance (state) is in a valid kingdom, by using diligently following all the
validations, constraints, and conditions that the database designers have imposed.
Instance –

O
permit’s say a desk trainer in our database whose name is college, suppose the
table has 50 information so the instance of the database has 50 facts for now and the
following day we’re going to upload another fifty records so the next day the instance

ty
have general a hundred information. This is known as an example.

Difference between schema and instance:

si
Schema Instance
It is the overall description of the It is the cllection of information stored in a
database. database at a particular moment.
r
Schema is same for shole database. Data in instances can be changed usng
ve
addition, deletion, updation.
Defines the basic structure of the It is the set of information stored at a particular
database i.e., how the data will be time.
strored in the database
ni

1.2.3 Data Independence

What is Data Independence of DBMS?
U

Data independence is defined as a property of DBMS that lets you alternate the
database schema at one degree of a database system without requiring to change the
schema at the following higher stage. Facts independence helps you to maintain facts
ity

separated from all packages that employ it. You could use this saved information for
computing and presentation. In lots of systems, information independence is a critical
function for additives of the device. Types of statistics independence

In DBMS there are styles of information independence

m

◌◌ Physical records independence

◌◌ Logical records independence.
)A

◌◌ Bodily records independence

Physical records independence lets you separate conceptual ranges from the
inner/physical levels. It allows you to provide a logical description of the database
without the need to specify physical structures. As compared to logical independence,
it is simple to gain physical statistics independence. With bodily independence, you
(c

could easily change the physical storage structures or gadgets with an effect on the
conceptual schema. Any change finished might be absorbed by means of the mapping
among the conceptual and internal levels. Physical facts independence is done via
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 35

the presence of the inner stage of the database after which the transformation from
Notes

e
the conceptual level of the database to the internal degree. Examples of modifications
beneath physical statistics independence Due to bodily independence, any of the below
exchange will no longer have an effect on the conceptual layer.

in
●● The use of a new storage tool like difficult pressure or magnetic tapes
●● Editing the document agency method within the database

nl
●● Switching to special facts structures.
●● Changing the get right of entry to method.

O
●● Enhancing indexes.
●● Adjustments to compression techniques or hashing algorithms.
●● Alternate of place of database from say c power to do drilogical records

ty
independence. Logical information independence is the potential to trade the
conceptual scheme without converting
1. Outside views

si
2. External API or applications
Any change made might be absorbed by the mapping between external and
r
conceptual tiers. While in comparison to bodily records independence, it’s far tough to
achieve logical statistics independence. Examples of modifications underneath logical
ve
statistics independence

Because of logical independence, any of the beneath alternate will now not affect
the external layer.
ni

●● Add/modify/delete a brand-new characteristic, entity or dating is feasible without a

rewrite of present application programs
●● Merging two statistics into one
U

●● Breaking an existing report into or extra records

Lgical Data Independence Physical data independence

ity

Mainly concerned with the structur or Mainly concerned with the storage of the
changing the data definition data
It is difficult as the retrieving of data is mainly It is easy to retrieve
dependent on the logical structure of data
m

Need to make changes in the Application A change in the Physical level usually does
program if new fields are added of deleted not need change at the application program
from the database level
)A

Modification at the logical levels is significant Modification made at the internal levels
whenever the logical structures of the may or may not be needed to inprove the
database are changed performance of the structure
Concerned with conceptual schema Concerned with internal schema
Example: Add, Modify, Delete a new E x a m p l e : c h a n g e i n c o m p r e s s i o n
(c

attribute techniques, nashing algorithms, storage

devices, etc.

Amity Directorate of Distance & Online Education

 36 Introduction to Data Base Management Systems

Importance of Data Independence

Notes

e
◌◌ helps you to improve the pleasant of the data
◌◌ Database machine protection becomes low cost

in
◌◌ Enforcement of requirements and development in database security
◌◌ You don’t want to adjust records shape in utility packages

nl
◌◌ allow builders to awareness on the overall structure of the database as a
substitute demanding about the internal implementation
◌◌ It allows you to improve state which is intact or undivided

O
◌◌ Database incongruity is vastly reduced.
◌◌ Without problems make modifications inside the physical stage is needed to
enhance the performance of the system...

ty
r si
ve
ni

1.2.4 DBMS Language and Interfaces

1. DBMS languages
U

As soon as the layout of a database is completed and a DBMS is chosen to put

into effect the database, the first step is to specify conceptual and internal schemas for
the database and any mappings between the
ity

In many DBMSs in which no strict separation of stages is maintained, one

language, referred to as the facts definition language (ddl), is utilized by the dba and by
database designers to define both schemas. The DBMS will have a ddl compiler whose
feature is to method ddl statements that allows you to become aware of descriptions of
m

the schema constructs and to shop the schema description within the DBMS catalog.
In DBMSs in which a clean separation is maintained among the conceptual and inner
tiers, the ddl is used to specify the conceptual schema most effective. Any other
)A

language, the garage definition language (sdl), is used to specify the inner schema. The
mappings between the 2 schemas may be laid out in either this kind of languages. In
maximum relational DBMSs today, there is no precise language that plays the function
of sdl. As a substitute, the inner schema is special by way of a mixture of functions,
parameters.
(c

In DBMS in which a clear separation is maintained among the conceptual and

inner tiers, the ddl is used to specify the conceptual schema handiest. Every other

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 37

language, the garage definition language (sdl), is used to specify the inner schema. The
Notes

e
mappings between the two schemas can be specified in both any such languages. In
most relational DBMS today, there’s no particular language that performs the function
of sdl. Instead, the internal schema is distinctive by an aggregate of capabilities,

in
parameters, and specifications associated with garage. Those permit the dba body
of workers to govern indexing choices and mapping of records to storage. For a real
3-schema architecture, we’d want a third language, the view definition language (vdl),

nl
to specify person perspectives and their mappings to the conceptual schema, however
in most DBMS the ddl is used to outline both conceptual and outside schemas. In
relational DBMS, sq. Is used within the position of vdl to outline person or application

O
perspectives as consequences of predefined queries (see chapters 4 and five).
Once the database schemas are compiled and the database is populated with data,
customers ought to have some manner to control the database. Regular manipulations
encompass retrieval, insertion, deletion, and amendment of the information. The DBMS

ty
provides a set of operations or a language called the facts manipulation language
(dml) for these purposes. In cutting-edge DBMSs, the preceding sorts of languages are
generally no longer taken into consideration wonderful languages; as an alternative, a

si
comprehensive incorporated language is used that includes constructs for conceptual
schema definition, view definition, and data manipulation. Garage definition is usually
saved separate, considering the fact that it is used for defining bodily storage structures

r
to excellent-tune the performance of the database machine, that’s commonly performed
by means of the dba team of workers. A normal instance of a complete database
ve
language is the sq. Relational database language which represents a mixture of ddl,
vdl, and dml, as well as statements for constraint specification, schema evolution, and
other functions. The sdl changed into an element in early variations of square but has
been removed from the language to maintain it on the conceptual and outside ranges
ni

handiest. There are important forms of dmls. An excessive-level or nonprocedural dml

can be used on its very own to specify complicated database operations concisely.
Many DBMSs allow high-level dml statements both to be entered interactively from
U

a show reveal or terminal or to be embedded in a preferred-motive programming

language. In the latter case, dml statements have to be identified in the program so
they can be extracted by a precompiler and processed through the DBMS. A low-level
or procedural dml have to be embedded in a general-purpose programming language.
ity

This kind of dml normally retrieves man or woman facts or objects from the database
and strategies each separately. Consequently, it wishes to use programming language
constructs, inclusive of looping, to retrieve and procedure each file from a hard and fast
of statistics. Low-degree dmls also are referred to as document-at-a-time dmls because
of this belongings. Dl/1, a dml designed for the hierarchical model, is a low-degree dml
m

that uses instructions together with get particular, get subsequent, or get next within
figure to navigate from file to record within a hierarchy of statistics within the database.
Excessive-level dmls, such as sq., can specify and retrieve many data in an unmarried
)A

dml statement; therefore, they may be referred to as set-at-a-time or set-oriented dmls.

A question in a high-stage dml regularly specifies which statistics to retrieve instead of
a way to retrieve it; therefore, such languages also are called declarative. Every time
dml commands, whether or not excessive degree or low stage, are embedded in a
fashionable-reason programming language, that language is called the host language
(c

and the dml is called the facts sublanguage. Then again, a high-stage dml utilized in a
standalone interactive way is known as a question language. In general, each retrieval

Amity Directorate of Distance & Online Education

 38 Introduction to Data Base Management Systems

and update instructions of a high-level dml may be used interactively and are therefore
Notes

e
taken into consideration part of the question language. Informal stop customers usually
use a high-level question language to specify their requests, whereas programmers
use the dml in its embedded shape. For naïve and parametric users, there generally

in
are person-friendly interfaces for interacting with the records-base; these also can be
used by casual customers or others who do now not need to examine the details of an
excessive-level question language. We talk these sorts of interfaces subsequent.

nl
Database Language
o A DBMS has appropriate languages and interfaces to express database

O
queries and updates.
o Database languages can be used to read, store and update the data in the
database.

ty
Types of Database Language

r si
ve
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 39

Notes

e
in
nl
O
ty
r si
ve
●● DBMS interfaces
Person-pleasant interfaces furnished by means of a DBMS may also encompass
the subsequent:

Menu-based interfaces for net clients or browsing. Those interfaces pre-sent

ni

the consumer with lists of alternatives (referred to as menus) that lead the user via
the components of a request. Menus remove the need to memorize the particular
commands and syntax of a query language; alternatively, the question consists step-
U

by using-step by means of selecting options from a menu this is displayed by way of

the system. Pull-down menus are a totally popular technique in net-based totally user
interfaces.’ They’re also frequently used in browsing interfaces, which permit a user to
look through the contents of a database in an exploratory and unstructured manner.
ity

Bureaucracy-based totally interfaces. A forms-primarily based interface shows a

shape to each person. Users can fill out all of the shape entries to insert new facts, or
they can fill out handiest sure entries, wherein case the DBMS will retrieve matching
facts for the last entries. Bureaucracy are generally designed and programmed naïve
m

naive customers as inter-faces to canned transactions. Many DBMSs have forms

specification languages, which are special languages that assist programmers specify
such paperwork. Sq.*forms is a shape-based totally language that specifies queries
)A

using a shape designed in conjunction with the relational database schema. Oracle
bureaucracy is part of the oracle product suite that offers an in depth set of functions to
layout and construct packages the use of forms. A few systems have utilities that outline
a shape through letting the quit consumer interactively assemble a pattern shape on
the display screen. Graphical consumer interfaces. A GUIS commonly presentations a
(c

schema to the consumer in diagrammatic form. The person then can specify a query
by means of manipulating the diagram. In many cases, GUISs utilize both menus
and paperwork. Maximum GUISs use a pointing tool, together with a mouse, to pick

Amity Directorate of Distance & Online Education

 40 Introduction to Data Base Management Systems

positive elements of the displayed schema diagram. Natural language interfaces.

Notes

e
These interfaces accept requests written in English or a few other language and try and
apprehend them. An herbal language interface usually has its own schema, which is
much like the database conceptual schema, as well as a dictionary of critical phrases.

in
The herbal language interface refers back to the words in its schema, in addition to
the set of popular words in its dictionary, to interpret the request. If the interpretation
is successful, the inter-face generates an excessive-stage query corresponding to

nl
the natural language request and submits it to the DBMS for processing; otherwise,
a communicate is started out with the consumer to make clear the request. The
capabilities of natural language interfaces have now not advanced unexpectedly. These

O
days, we see search engines like Google that receive strings of herbal language (like
English or Spanish) words and match them with files at specific sites (for local serps)
or net pages on the net at huge (for serps or ask). They use predefined indexes on
words and use rating functions to retrieve and present ensuing documents in a

ty
decreasing diploma of suit. Such “unfastened shape” textual question interfaces are
not but common in established relational or legacy model databases, despite the fact
that a studies place known as key-word-based querying has emerged these days for

si
relational databases. Speech input and output. Restricted use of speech as an input
question and speech as an answer to a query or result of a request is turning into not
unusual. Programs with limited vocabularies such as inquiries for cell phone directory,

r
flight arrival/departure, and credit score card account statistics are allowing speech for
enter and output to enable customers to get right of entry to this records. The speech
ve
input is detected the usage of a library of predefined phrases and used to installation
the parameters which can be supplied to the queries. For output, a similar conversion
from text or numbers into speech takes location. Interfaces for parametric customers.
Parametric users, which include financial institution tellers, frequently have a small set
ni

of operations that they have to carry out repeatedly. For instance, a teller is able to use
single function keys to invoke recurring and repetitive transactions inclusive of account
deposits or withdrawals, or stability inquiries. Systems analysts and programmers
U

design and put into effect a special interface for each acknowledgednaïves of naive
users. Commonly a small set of abbreviated commands is protected, with the purpose
of minimizing the variety of keystrokes required for each request. As an example,
function keys in a terminal can be programmed to initiate numerous instructions.
ity

This lets in the parametric person to proceed with a minimum quantity of keystrokes.
Interfaces for the dba. Most database structures incorporate privileged instructions that
may be used simplest by means of the dba workforce. Those encompass commands
for growing accounts, putting machine parameters, granting account authorization,
changing a schema, and reorganizing the storage structures of a database.
m

1.2.5 Data Models

)A

Records model offers us an idea that how the very last gadget will appear like after
its complete implementation. It defines the information factors and the relationships
between the facts elements. Statistics fashions are used to expose how records is
saved, related, accessed and updated in the database control device. Right here, we
use a set of symbols and textual content to symbolize the records so that contributors
(c

of the organisation can speak and apprehend it. Though there are numerous data
fashions being used these days however the relational version is the most extensively

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 41

used model. Other than the relational model, there are many other kinds of statistics
Notes

e
fashions about which we can examine in info on this blog. A number of the statistics
fashions in DBMS are:

in
◌◌ Hierarchical Model
◌◌ Network Model
◌◌ Entity-Relationship Model

nl
◌◌ Relational Model
◌◌ Object-Oriented Data Model

O
◌◌ Object-Relational Data Model
◌◌ Flat Data Model
◌◌ Semi-Structured Data Model

ty
◌◌ Associative Data Model
◌◌ Context Data Model

Hierarchical Model

si
Hierarchical Model was the first DBMS model. This model organises the data in the
hierarchical tree structure. The hierarchy starts from the root which has root data and
then it expands in the form of a tree adding child node to the parent node. This model
r
easily represents some of the real-world relationships like food recipes, sitemap of a
ve
website etc. Example: We can represent the relationship between the shoes present on
a shopping website in the following way:
ni
U
ity

Features of a Hierarchical Model

m

o One-to-many relationship: The data here is organised in a tree-like structure where

the one-to-many relationship is between the data types. Also, there can be only one
path from parent to any node. Example: In the above example, if we want to go to the
)A

node sneakers we only have one path to reach there i’e. through men’s shoes node.
o Parent-Child Relationship: Each child node has a parent node but a parent node can
have more than one child node. Multiple parents are not allowed.
o Deletion Problem: If a parent node is deleted then the child node is automatically
(c

deleted.

Amity Directorate of Distance & Online Education

 42 Introduction to Data Base Management Systems

o Pointers: Pointers are used to link the parent node with the child node and are used
Notes

e
to navigate between the stored data. Example: In the abo‘e exa’ple the ‘shoes’ node
points to the t‘o other nod’s ‘women s‘oes’ node a’d ‘men’s shoes’ node.

in
nl
O
ty
Advantages of Hierarchical Model
●● It is very simple and fast to traverse through a tree-like structure.

si
●● Any change in the parent node is automatically reflected in the child node so, the
integrity of data is maintained.
●● Disadvantages of Hierarchical Model
●● r
Complex relationships are not supported.
ve
●● As it does not support more than one parent of the child node so if we have some
complex relationship where a child node needs to have two parent nod’ then that
can’t be represented using this model.
ni

●● If a parent node is deleted then the child node is automatically deleted.

Network Model
U

This model is an extension of the hierarchical model. It was the most popular
model before the relational model. This model is the same as the hierarchical model,
the only difference is that a record can have more than one parent. It replaces the
hierarchical tree with a graph. Example: In the example below we can see that node
ity

student has two parents i.e. CSE Department and Library. This was earlier not possible
in the hierarchical model.

Features of Network Model

m

1. Capacity to merge greater relationships: on this version, as there are more

relationships so data is more related. This version has the potential to control one-
to-one relationships in addition to many-to-many relationships.
)A

2. Many paths: as there are more relationships so there can be more than one course
to the same record. This makes data get right of entry to speedy and easy.
3. Circular linked listing: the operations on the community version are completed with
the assist of the round related list. The modern-day function is maintained with the
assist of an application and this position navigates through the statistics in keeping
(c

with the connection. Benefits of network model

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 43

The data may be accessed faster compared to the hierarchical version. This is
Notes

e
due to the fact the facts is greater related within the community model and there can
be a couple of course to attain a particular node. So, the records may be accessed
in lots of method’. As there’s a parent-infant relationship so facts integrity is gift. Any

in
trade in discern document is pondered in the toddler file. Dangers of community model.
As increasingly relationships want to be dealt with the system may get complex. So, a
consumer ought to be having distinctive knowledge of the model to paintings with the

nl
model. Any trade like updation, deletion, insertion could be very complicated.

Entity-Relationship Model

O
Entity-relationship model or honestly er version is a high-degree records version
diagram. In this model, we represent the real-world hassle within the pictorial form to
make it clean for the stakeholders to understand. It is also very smooth for the builders
to apprehend the gadget with the aid of just searching on the er diagram. We use

ty
the ER diagram as a visible device to represent an er version. Er diagram has the
subsequent 3 additives:

◌◌ Entities: entity is an actual-world issue. It is able to be a person, place, or

si
even a concept. Example: teachers, college students, direction, constructing,
branch, and so forth are a number of the entities of a faculty management
machine.
◌◌
r
Attributes: an entity contains a real-world assets called characteristic. This is
ve
the traits of that characteristic. Instance: the entity teacher has the assets like
trainer identity, income, age, and many others.
◌◌ Relationship: relationship tells how attributes are associated. Example:
instructor works for a branch.
ni

E.g.:
U
ity
m
)A

Within the above diagram, the entities are teacher and department. The attributes
of instructor entity are teacher_name, teacher_id, age, revenue, mobile_number.
The attributes of entity branch entity are dept_id, dept_name. The two entities are
connected using the connection. Here, each teacher works for a branch.
(c

Amity Directorate of Distance & Online Education

 44 Introduction to Data Base Management Systems

Capabilities of er version
Notes

e
◌◌ Graphical representation for better information: its miles very smooth and
simple to recognize so it is able to be used by the builders to speak with the

in
stakeholders.
◌◌ Er diagram: er diagram is used as a visible tool for representing the model.
◌◌ Database layout: this version allows the database designers to construct the

nl
database and is broadly used in database layout. Benefits of er model
◌◌ Simple: conceptually er version could be very clean to construct. If we realize
the connection among the attributes and the entities we are able to effortlessly

O
construct the er diagram for the version.
◌◌ Powerful verbal exchange device: this version is used extensively via the
database designers for communicating their ideas.

ty
◌◌ Easy conversion to any model: this version maps well to the relational version
and can be easily transformed relational model by using changing the er
version to the table. This model can also be converted to any other model like
community model, hierarchical version and many others.

si
◌◌ Conceptually it is quite simple: er model is quite simple due to the fact if we
realize courting between entities and attributes, then we are able to without

◌◌
r
problems draw an er diagram.
Better visible representation: er model is a diagrammatic representation of any
ve
logical structure of database. By way of seeing er diagram, we are able to
effortlessly understand dating amongst entities and courting.
◌◌ Enormously included with relational model: er model may be effortlessly
transformed into relational model through actually converting er model into
ni

tables.
◌◌ Easy conversion to any information model: er version may be effortlessly
converted into some other data version like hierarchical records version,
U

network information version and so forth. Disadvantages of er version

◌◌ No industry fashionable for notation: there may be no enterprise well known
for growing an er model. So one developer might use notations which aren’t
ity

understood by using other developers.

◌◌ Hidden information: a few records might be misplaced or hidden within the er
model. As it is an excessive-degree view so there are possibilities that a few
information of facts might be hidden.
m

◌◌ Restrained constraints and specification

◌◌ constrained relationship representation: er version represents limited
relationship as compared to any other statistics models like relational model
)A

and so on.
◌◌ No illustration of statistics manipulation: its miles difficult to show information
manipulation in er model.
◌◌ Popular for high level layout: er model could be very famous for designing
(c

excessive level layout

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 45

Relational Model
Notes

e
Relational model is the maximum broadly used version. In this version, the records
is maintained in the shape of a -dimensional desk. All the facts is stored in the shape of

in
row and columns. The basic shape of a relational version is tables. So, the tables are
also called members of the family inside the relational version. Example: in this case,
we have a worker table. Functions of relational version

nl
●● Tuples: every row inside the table is referred to as tuple. A row includes all of
the statistics about any instance of the item. Inside the above example, each
row has all of the data about any unique person just like the first row has facts

O
approximately john.
●● Attribute or subject: attributes are the belongings which defines the table or
relation. The values of the attribute should be from the equal domain. Within the
abo’e instance, we’ve different attributes of the worker like revenue, mobile no,

ty
and many others. Advantages of relational version
●● Easy: this version is extra easy as compared to the community and hierarchical
model.

si
●● Scalable: this model may be without difficulty scaled as we will add as many rows
and columns we need.
●● r
Structural independence: we are able to make adjustments in database shape
ve
without changing the manner to get right of entry to the information. Whilst we will
make changes to the database shape without affecting the capability to DBMS to
get admission to the statistics we are able to say that structural independence has
been accomplished. Disadvantages of relational model
ni

●● Hardware overheads: for hiding the complexities and making things less
complicated for the person this version calls for more effective hardware
computers and records storage devices.
U

●● Terrible layout: because the relational model could be very smooth to design
and use. So the customers don’t need to realize how the records is saved
with the intention to get right of entry to it. This ease of design can result in the
improvement of a poor database which could sluggish down if the database
ity

grows. But all these negative aspects are minor as compared to the advantages
of the relational model. Those problems can be averted with the assist of right
implementation and company.
m

Object-Oriented Data Model

The real-world troubles are extra closely represented through the item-oriented
information version. In this model, each the facts and courting are present in an
)A

unmarried structure referred to as an object. We can shop audio, video, photos, etc.
within the database which turned into now not viable inside the relational model (even
though you can shop audio and video in relational database, it’s far advised now not to
shop inside the relational database).

In this version, two are more gadgets are linked through links. We use this
(c

link to relate one item to other objects. This could be understood by means of the
instancnaïveiven under...

Amity Directorate of Distance & Online Education

 46 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Inside the ’bove instance, we’ve got objects worker and department. All the

ty
statistics and relationships of each object are contained as an unmarried unit. The
attributes like call, job title of the worker and the strategies for you to be completed with
the aid of that object are stored as an unmarried object. The two gadgets are linked via

si
a commonplace attribute I. E the department_id and the communication between these
two might be done with the assist of this commonplace identification.

r
1.2.6 Data Definitions Language, DML
ve
DDL or Data Definitions Language
A data definition language (DDL) is a computer language used to create and
modify the structure of database objects in a database. These database objects include
views, schemas, tables, indexes, etc. This term is also known as data description
ni

language in some contexts, as it describes the fields and records in a database table.

The present database industry incorporates DDL into any formal language
U

describing data. However, it is considered to be a subset of SQL (Structured Query

Language). SQL often uses imperative verbs with normal English such as sentences
to implement database modifications. Hence, DDL does not show up as a different
language in an SQL database, but does define changes in the database schema.
ity

It is used to establish and modify the structure of objects in a database by dealing

with descriptions of the database schema. Unlike data manipulation language (DML)
commands that are used for data modification purposes, DDL commands are used for
altering the database structure such as creating new tables or objects along with all
m

their attributes (data type, table name, etc.).

Commonly used DDL in SQL querying are CREATE, ALTER, DROP, and
TRUNCATE.
)A

Create
This command builds a new table and has a predefined syntax. The CREATE
statement syntax is:
(c

CREATE TABLE [table name] ([column definitions]) [table parameters];

For example:
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 47

CREATE TABLE Employee (Employee Id INTEGER PRIMARY KEY, First name

Notes

e
CHAR (50) NULL, Last name CHAR (75) NOT NULL);

The mandatory semi-colon at the end of the statement is used to process every

in
command before it. In this example, the string CHAR is used to specify the data type.
Other data types can be DATE, NUMBER, or INTEGER.

nl
Alter
An alter command modifies an existing database table. This command can add up
additional column, drop existing columns and even change the data type of columns

O
involved in a database table.

An alter command syntax is:

ALTER object type object name parameters;

ty
For example:
ALTER TABLE Employee ADD PRIMARY KEY (employee_pk);

si
In this example, we added a unique primary key to the table to add a constraint and
enforce a unique value. The constraint “employee_pk” is a primary key and is on the
Employee table.
r
ve
Drop
A drop command is used to delete objects such as a table, index or view. A DROP
statement cannot be rolled back, so once an object is destroyed, there’s no way to
recover it.
ni

Drop statement syntax is:

DROP object type object name;

U

For example:
DROP TABLE Employee;
ity

In this example, we’re deleting the Employee table.

Truncate
Similar to DROP, the TRUNCATE statement is used to quickly remove all records
m

from a table. However, unlike DROP that completely destroys a table, TRUNCATE
preserves its full structure to be reused later.

Truncate statement syntax is:

)A

TRUNCATE TABLE table_name;

For example:
TRUNCATE TABLE Employee;
(c

In this example, we’re marking all the extents of the Employee table for
deallocation, so they’re considered empty for reuse.

Amity Directorate of Distance & Online Education

 48 Introduction to Data Base Management Systems

Other statements
Notes

e
Other commonly used commands include RENAME and COMMENT. The first
one is used with the ALTER TABLE statement to change the name of an object (table,

in
column, etc.). COMMENT is used to add single line, multi-line and in-line comments.

DML (Data Manipulation Language)

nl
A data manipulation language (DML) is a computer programming language
used for adding (inserting), deleting, and modifying (updating) data in a database. A
DML is often a sublanguage of a broader database language such as SQL, with the

O
DML comprising some of the operators in the language. Read-only selecting of data
is sometimes distinguished as being part of a separate data query language (DQL),
but it is closely related and sometimes also considered a component of a DML; some
operators may perform both selecting (reading) and writing.

ty
A popular data manipulation language is that of Structured Query Language (SQL),
which is used to retrieve and manipulate data in a relational database. Other forms of
DML are those used by IMS/DLI, CODASYL databases, such as IDMS and others.

si
List of DML commands:
◌◌ INSERT: It is used to insert data into a table.
◌◌
r
UPDATE: It is used to update existing data within a table.
ve
◌◌ DELETE: It is used to delete records from a database table.
◌◌ LOCK: Table control concurrency,
◌◌ CALL: Call a PL/SQL or JAVA subprogram.
ni

◌◌ EXPLAIN PLAN: It describes the access path to data.

Difference between DML and DDL

U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 49

Check Your Understanding

Notes

e
A. Fill in the Blanks
1. _________ is also called view level.

in
2. ________ are the collection of information stored at a particular moment.
3. ________________is used by the DBA and by database designers to define both

nl
schemas.
4. ________ are used to link the parent node with the child node and are used to
navigate between the stored data.

O
5. An entity contains a real-world property called _________.
B. State True or False
1. A database schema defines its entities and the relationship among them

ty
2. A GUIS typically displays a schema to the user in diagrammatic form
3. In Hierarchical Model the data is maintained in the form of a two-dimensional table.

si
4. TCL is used to run the changes made by the DML statement.
5. Truncate is used to Rename all records from a table.

Summary r
ve
DBMS three level architecture has three degrees:

●● Outside level
Conceptual degree
ni

Internal level

External degree: it is also called view level. The motive this stage is referred to
U

as “view” is due to the fact numerous users can view their preferred records from this
stage that is internally fetched from database with the assist of conceptual and internal
degree mapping. The person doesn’t want to know the database schema info such as
statistics structure, desk definition and so on. User is handiest concerned approximately
ity

statistics that is what back again to the v’ew stage after it’s been fetched from database.
External degree is the “top stage” of the 3 stage DBMS structure.

Conceptual stage: it is also referred to as logical stage. The complete layout of

the database inclusive of courting amongst information, schema of statistics etc. Are
m

described in this stage. Database constraints and protection also are carried out in this
degree of structure. This level is maintained by way of dba (database administrator).

Inner stage: this level is likewise known as bodily stage. This degree describes
)A

how the facts is definitely stored inside the storage devices. This level is likewise
chargeable for allocating area to the data. This is the bottom stage of the architecture.
A database schema defines its entities and the connection amongst them. It consists
of a descriptive element of the database, which may be depicted by way of schema
diagrams. It’s the database designers who design the schema to assist programmers
(c

apprehend the database and make it useful.

Schema is of 3 types: logical schema, physical schema and think about schema.

Amity Directorate of Distance & Online Education

 50 Introduction to Data Base Management Systems

1. Logical schema – it describes the database designed at logical level.

Notes

e
2. Physical schema – it describes the database designed at bodily stage.

3. View schema – it defines the layout of the database at the view level.

in
A database example is a state of operational database with statistics at any given
time. It carries a picture of the database. Database instances have a tendency to

nl
change with time. A DBMS guarantees that it’s each instance (kingdom) is in a valid
state, by way of diligently following all the validations, constraints, and conditions that
the database designers have imposed. Statistics independence is defined as a property
of DBMS that helps you to alternate the database schema at one degree of a database

O
gadget without requiring to change the schema at the following higher level. Information
independence helps you to preserve statistics separated from all applications that
employ it.

ty
Varieties of facts independence

◌◌ In DBMS there are sorts of facts independence

si
Physical facts independence
Logical data independence.

●● r
With bodily independence, you could effortlessly exchange the physical storage
structures or devices with an effect at the conceptual schema. Any exchange
ve
achieved might be absorbed by using the mapping among the conceptual and
inner degrees. Physical records independence is achieved by way of the presence
of the internal degree of the database after which the transformation from the
conceptual stage of the database to the internal stage.
ni

Importance of facts independence

◌◌ Lets you improve the excellent of the information
U

◌◌ Database machine preservation becomes low-cost

◌◌ Enforcement of requirements and development in database safety
◌◌ You don’t need to alter statistics shape in utility programs
ity

◌◌ allow developers to awareness on the overall structure of the database

alternatively hand demanding approximately the internal implementation
◌◌ It allows you to enhance kingdom which is undamaged or undivided
m

◌◌ Database incongruity is hugely decreased.

Without problems make modifications inside the physical stage is wanted to
improve the performance of the system. In lots of DBMSs wherein no strict separation
)A

of degrees is maintained, one language, called the records definition language (ddl), is
used by the dba and via database designers to define each schemas. The DBMS will
have a ddl compiler whose feature is to procedure ddl statements as a way to discover
descriptions of the schema constructs and to shop the schema description within the
DBMS catalogue
(c

In DBMSs in which a clean separation is maintained between the conceptual

and inner stages, the ddl is used to specify the conceptual schema best. Another

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 51

language, the storage definition language (sdl), is used to specify the inner schema.
Notes

e
The mappings between the 2 schemas can be specified in both this sort of languages.
The view definition language (vdl), to specify consumer views and their mappings to
the conceptual schema, but in most DBMSs the ddl is used to outline both conceptual

in
and outside schemas. Once the database schemas are compiled and the database
is populated with information, customers should have a few means to manipulate the
database. Typical manipulations include retrieval, insertion, deletion, and modification

nl
of the facts. The DBMS seasoned-vides a set of operations or a language called the
information manipulation language (dml) for those functions. There are two major
forms of dmls. An excessive-level or nonprocedural dml can be used on its personal

O
to specify complex database operations concisely. Many DBMSs permit excessive-
degree dml statements both to be entered interactively from a show reveal or terminal
or to be embedded in a popular-reason programming language. In the latter case, dml
statements ought to be diagnosed in the software in order that they can be extracted

ty
via a precompiler and processed with the aid of the DBMS. A low-degree or procedural
dml have to be embedded in a popular-reason programming language. This form of
dml usually retrieves man or woman records or items from the database and tactics

si
each separately. Therefore, it wishes to use programming language constructs,
which includes looping, to retrieve and manner every file from a set of statistics. Low-
level dmls are also called document-at-a-time dmls because of this belongings. A

r
dml designed for the hierarchical version, is a low-degree dml that uses commands
inclusive of get specific, get subsequent, or get next within parent to navigate from
ve
document to document within a hierarchy of information in the database. High-stage
dmls, along with sq., can specify and retrieve many information in an unmarried dml
statemen’; therefore, they’re referred to as set-at-a-time or set-orientated dmls. A
question in an excessive-stage dml often specifies which statistics to retrieve as
ni

opposed to the way to retrieve it; therefore, such languages also are known as
declarative. Whenever dml instructions, whether or not high degree or low degree, are
embedded in a well-known-reason programming language, that language is known as
U

the host language and the dml is known as the records sublanguage. Then again, an
excessive-level dml utilized in a standalone interactive way is called a query language.
In popular, both retrieval and replace instructions of a high-degree dml may be used
interactively and are therefore considered part of the query language. Casual quit
ity

users commonly use a high-stage query language to specify their requests, whereas
programmers use the dml in its enaïveed form. For naive and parametric users, there
commonly are consumer-pleasant interfaces for interacting with the data-base; these
can also be used by casual users or others who do not need to analyse the information
of a higher query language.
m

Records definition language

)A

◌◌ Ddl stands for data definition language.

◌◌ Its miles used to define database shape or pattern.
◌◌ Its miles used to create schema, tables, indexes, constraints, and many
others. Inside the database.
◌◌ the use of the ddl statements, you could create the skeleton of the database.
(c

◌◌ information definition language is used to shop the statistics of metadata just

like the quantity of tables and schemas, their names, indexes, columns in

Amity Directorate of Distance & Online Education

 52 Introduction to Data Base Management Systems

each table, constraints, and so on. Here are some responsibilities that come
Notes

e
underneath ddl:
◌◌ create: its miles used to create items inside the database.

in
◌◌ Modify: it is used to adjust the structure of the database.
◌◌ drop: its miles used to delete gadgets from the data’ase.
◌◌ truncate: it’s far used to put off all information from a desk. O rename: it is

nl
used to rename an ’bject.
◌◌ remark: it’s far used to comment on the statistics dictionary. Those instructions
are used to replace the database schema it is why they come beneath

O
information definition language

Facts manipulation language

ty
●● Dml stands for information ma’ipulation language. It’s far used for gaining access
to and manipulating data in a database. It handles person requests. Right here are
a few tasks that come u’der dml:

si
●● select: it’s far used to retrieve data from a database.
●● insert: its miles used to insert information right into a table.
●● replace: it is used to update current statistics within a table.
●●
r
delete: its miles used to delete all information from a desk.
ve
●● merge: it plays upset
●● Operation, I. E., insert or replace operations.
●● call: it’s far used to call a structured question language or a java subprogram.
ni

●● explain plan: it has the parameter of explaining facts

●● lock table: it controls concurrency.
U

Facts manipulate language

◌◌ dcl stands for statistics manage language. It’s far used to retrieve the or
saved data.
ity

◌◌ O the dcl execution is transactional. It also has rollback parameters.

(But in oracle database, the execution of data manipulate language does no longer
have the function of rolling again.)
m

Right here are some responsibilities that come below dcl:

O supply: its miles used to provide person get entry to privto a database.
)A

●● O revoke: it’s far used to take returned permissions from the consumer. There are
the following operations which have the authorization of revoke:
Join, insert, usage, execute, delete, replace and select. d) Transaction manage
language
(c

Tell is used to run the changes made by means of the dml statement. Tell can be
grouped into a logical transaction. Here are a few obligations that come under ’tell:

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 53

●● O dedicate: it’s far used to shop the transaction at the database. O rollback: it
Notes

e
is used to repair the database to original since the closing dedicate. 2. DBMS
interfaces

in
Consumer-pleasant interfaces furnished by a DBMS may encompass the
following:
Menu-primarily based interfaces for internet customers or browsing. These

nl
interfaces pre-sent the consumer with lists of alternatives (referred to as menus) that
lead the user through the formulation of a request. Menus cast off the want to memorize
the specific commands and syntax of a question language; as an alternative, the

O
query consists step-with the aid of-step by way of picking options from a menu that
is displayed with the aid of the gadget. Pull-down menus are a completely famous
technique in web-based totally consumer interfaces. They’re also regularly utilized in
surfing interfaces, which permit a user to leaf through the contents of a database in

ty
an exploratory and unstructured way. Bureaucracy-based interfaces. A bureaucracy-
based totally interface presentations a shape to every person. Users can fill out all the
form entries to insert new information, or they can fill out only positive entries, wherein

si
case the DBMS will retrieve matching information for the last entries. Bureaucracy
are normally designaïvend programmed for naive users as inter-faces to canned
transactions. Many DBMSs have bureaucracy specification languages that are special
r
languages that assist programmers specify such forms. Square*paperwork is a shape-
primarily based language that specifies queries the use of a shape designed along
ve
with the relational database schema. Oracle forms is part of the oracle product suite
that offers an in depth set of functions to layout and construct applications the use of
forms. A few structures have utilities that define a form by way of letting the stop user
interactively construct a pattern shape at the display. Graphical consumer interfaces. A
ni

GUIS normally shows a schema to the person in diagrammatic shape. The consumer
then can specify a query by using manipulating the diagram. In many instances, GUISs
make use of each menus and paperwork. Most GUISs use a pointing device, which
U

includes a mouse, to pick out certain parts of the displayed schema diagram. Natural
language interfaces. These interfaces take delivery of requests written in English or
some other language and try to recognize them. An herbal language interface generally
has its personal schema, which is just like the database conceptual schema, as well
ity

as a dictionary of vital phrases. The natural language interface refers to the phrases
in its schema, as well as to the set of standard words in its dictionary, to interpret
the request. If the translation is a hit, the inter-face generates a high-stage question
similar to the herbal language request and submits it to the DBMS for processing; in
m

any other case, a communicate is started out with the user to clarify the request. The
skills of herbal language interfaces have now not superior swiftly. These days, we see
engines like Google that be given strings of herbal langu panishke English or spanish)
words and in shape them with files at precise websites (for neighbourhood serps) or
)A

net pages at the web at massive (for search engines like Google and yahoo or ask).
They use predefined indexes on words and use ranking functions to retrieve and
gift resulting documents in a decreasing diploma of match. Such “unfastened form”
textual query interfaces are not but common in structured relational or legacy model
(c

databases, even though a studies vicinity known as key-word-primarily based querying

has emerged currently for relational databases. Speech enter and output. Restrained
use of speech as an input query and speech as an answer to a query or result of a

Amity Directorate of Distance & Online Education

 54 Introduction to Data Base Management Systems

request is becoming common. Applications with limited vocabularies including

Notes

e
inquiries for cell phone directory, flight arrival/departure, and credit score card account
statistics are allowing speech for enter and output to permit clients to get admission
to this information. The speech input is detected the usage of a library of predefined

in
phrases and used to set up the parameters which are furnished to the queries. For
output, a similar conversion from textual content or numbers into speech takes location.
Interfaces for parametric users. Parametric customers, including bank tellers, often

nl
have a small set of operations that they must carry out repeatedly. For instance, a teller
is able to use unmarried function keys to invoke ordinary and repetitive transactions
including account deposits or withdrawals, or balance inquiries. Structures analysts and

O
programmers design and put in force a special interface fornaïve known magnificence
of naive users. Usually a small set of abbreviated instructions is protected, with the
purpose of minimizing the range of keystrokes required for every request. For instance,
function keys in a terminal can be programmed to initiate numerous instructions. This

ty
allows the parametric consumer to proceed with a minimum quantity of keystrokes.
Interfaces for the dba. Most database structures contain privileged commands that can
be used only by means of the dba workforce. Those consist of commands for growing

si
debts, putting machine parameters, granting account authorization, changing a schema,
and reorganizing the garage systems of a database. Information model gives us an idea
that how the final machine will appear like after its complete implementation. It defines

r
the information elements and the relationships among the statistics elements. Statistics
models in DBMS are:
ve
1. Hierarchical model
2. Community version
3. Entity-dating model
ni

4. Relational model
5. Item-orientated statistics version
U

6. sixteen. Object-relational records version

7. Flat information version
8. Semi-established records model
ity

9. Associative facts version

Context statistics version

Hierarchical version
m

●● Hierarchical version was the primary DBMS model. This model organises the
records within the hierarchical tree structure. The hierarchy starts from the
)A

foundation which has root statistics and then it expands in the shape of a tree
including toddler node to the parent node. This version without difficulty represents
some of the actual-world relationships like food recipes, sitemap of a website and
many others.
(c

tures of a hierarchical model

●● One-to-many relationship: the data here is organised in a tree-like shape in which

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 55

the only-to-many relationship is among the data types. Also, there may be best
Notes

e
one route from determine to any node. Example: inside the above example, if we
need to go to the node sneakers we handiest have one route to reach there I. E
through guys’ shoes node.

in
●● Parent-child dating: each child node has a discern node but a parent node could
have multiple infant n’de. A couple of dad and mom aren’t allowed.

nl
●● Deletion hassle: if a parent node is deleted then the child node is robotically
deleted.
●● Pointers: recommendations are used to link the determine node with the kid node

O
and are used to navigate between the stored facts.
●● Instanc‘: ins’de the above instance the ‘shoes’ node‘points to the 2’different ‘odes
‘ladies foot’ear’ node and ‘guys’ shoes’ node. Benefits of hierarchical model

ty
◌◌ It’s far quite simple and fast to traverse thru a tree-like shape.
◌◌ Any change inside the determine node is automatically meditated inside the
infant node so, the integrity of statistics is maintained. Risks of hierarchical

si
model
◌◌ ’Complex relationships aren’t supported.
◌◌ because it does no longer support more than one det’rmine of the child node
r
so if we’ve got a few complicated dating wherein a baby node needs to have
ve
two discern node then that cannot be represented the use of this version.
◌◌ If a determine node is deleted then the child node is routinely deleted.
Community model
This version is an extension of the hierarchical version. It was the maximum
ni

famous model earlier than the relational version. This version is the same as the
hierarchical model, the simplest distinction is that a record can have more than one
figure. It replaces the hierarchical tree with a graph.
U

Abilities of a network version:

◌◌ Ability to merge extra relationships: on this model, as there are more
ity

relationships so information is more related. This model has the capacity to

manage one-to-one relationships in addition to many-to-many relationships.
◌◌ Many paths: as there are greater relationships so there can be multiple
direction to the identical file. This makes facts access speedy and simple.
m

◌◌ Circular connected list: the operations on the community version are

accomplished with the help of the round linked listing. The modern position is
maintained with the help of a program and this function navigates through the
information in keeping with the connection. Benefits of community version
)A

◌◌ The statistics may be accessed quicker as compared to the hierarchical

model. That is due to the fact the records is more related in the network
version and there can be more than one path to attain a specific node. So the
facts can be accessed in’lots of approaches.
(c

◌◌ As there’s a figure-baby courting so statistics integrity is present. Any trade in

figure file is meditated inside the child report. Dangers of community model

Amity Directorate of Distance & Online Education

 56 Introduction to Data Base Management Systems

◌◌ As increasingly relationships want to be treated the device may get

Notes complicated. So, a user ought to be having distinctive information of the

e
version to work with the model.

in
◌◌ Any trade like updation, deletion, insertion could be very complicated. Entity-
courting model
Entity-courting model or definitely er model is a high-degree information version

nl
diagram. On this model, we constitute the real-international trouble within the pictorial
form to make it clean for t’e stakeholders to understand. It’s also very clean for the
developers to apprehend the system by using just searching on the er diagram. We
use the er diagram as a visible device to represent an er model. Er diagram has the

O
subsequent 3 components:

◌◌ Entities: entity is an actual-world aspect. It may be a person, location, or even

a concept. Instance: instructors, students, course, constructing, branch, and

ty
so forth are a number of the entities of a school management machine.
◌◌ Attributes: an entity carries a real-global property called attribute. This is
the characteristics of that characteristic. Example: the entity trainer has the

si
belongings like teacher identification, earnings, age, etc.
◌◌ dating: relationship tells how two attributes are related. Example: trainer works
for a branch. Capabilities of er model
◌◌ r
Graphical illustration for higher knowledge: its miles very easy and easy to
ve
recognize so it may be used by the builders to talk with the stakeholders.
◌◌ Er diagram: er diagram is used as a visible device for representing the
version.
◌◌ Database design: this version allows the database designers to construct the
ni

database and is broadly utilized in database design. Advantages of er version

◌◌ Easy: conceptually er model could be very clean to construct. If we
understand the relationship among the attributes and the entities we will
U

effortlessly build the er diagram for the version.

◌◌ Powerful communique tool: this version is used extensively by way of the
database designers for communicating their thoughts.
ity

◌◌ Easy conversion to any model: this model maps well to the relational model
and may be effortlessly converted relational version with the aid of converting
the er version to the desk. This model can also be transformed to every other
version like network model, hierarchical version etc.
m

◌◌ Conceptually its miles very simple: er version is very simple because if we

realize dating among entities and attributes, then we can without problems
draw an er diagram.
)A

◌◌ Higher visible representation: er model is a diagrammatic representation

of any logical structure of database. By using seeing er diagram, we can
effortlessly understand courting amongst entities and dating.
◌◌ Rather included with relational version: er version may be effortlessly
converted into relational model through virtually changing er model into tables.
(c

◌◌ Clean conversion to any statistics model: er version may be without difficulty

transformed into another statistics version like hierarchical data model,

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 57

network information model and so forth. Disadvantages of er model

Notes

e
◌◌ No enterprise popular for notation: there’s no enterprise standard for
developing an er version. So one developer may use notations which aren’t

in
understood by other builders.
◌◌ Hidden information: a few statistics is probably misplaced or hidden in the er
model. As it’s far an excessive-level view so there are probabilities that some

nl
info of information is probably hidden.
◌◌ onfined constraints and specification
◌◌ Restrained dating representation: er version represents restricted courting

O
compared to some other facts fashions like relational version etc.
◌◌ No illustration of information manipulation: its miles difficult to expose facts
manipulation in er version.

ty
◌◌ Popular for high degree layout: er version could be very popular for designing
high degree design

Relational version

si
Relational model is the maximum extensively used version. On this version, the
information is maintained within the form of a two-dimensional table. All the statistics is
stored within the form of row and columns. The simple structure of a relational version
r
is tables. So, the tables are also known as members of the family within the relational
ve
version. Functions of relational version

●● Tuples: each row within the table is referred to as tuple. A row carries all the
information approximately any example of the object. Within the above instance,
every row has all of the statistics approximately any precise person just like the
ni

first row has statistics approximately john.

●● Attribute or discipline: attributes are the property which defines the table or
relation. The values of the attribute should be from the same area. In the above
U

instance, we’ve distinct attributes of the employee like salary, mobile no, and many
others. Benefits of relational model
●● Simple: this version is simpler compared to the network and hierarchical model.
ity

●● Scalable: this model can be without difficulty scaled as we can add as many rows
and columns we want.
●● Structural independence: we will make changes in database shape without
changing the way to get right of entry to the records. While we will make
m

adjustments to the database shape without affecting the functionality to DBMS to

get admission to the statistics we can say that structural independence has been
carried out. Disadvantages of relational version
)A

●● Hardware overheads: for hiding the complexities and making things less difficult
for the person this version calls for greater effective hardware computer systems
and information garage devices.
●● Horrific layout: because the relational model may be very clean to layout and
(c

use. So the users don’t want to recognize how the data is saved to be able to
get admission to it. This ease of design can result in the development of a terrible

Amity Directorate of Distance & Online Education

 58 Introduction to Data Base Management Systems

database which could sluggish down if the database grows. However, some of
Notes

e
these hazards are minor compared to the blessings of the relational model. These
troubles may be avoided with the help of right implementation and company. Item-
oriented records model

in
The real-global troubles are more closely represented through the item-oriented
statistics version. In this model, both the records and dating are found in a single

nl
structure known as an object. We are able to save audio, video, snap shots, and so
on in the database which was now not feasible inside the relational version (although
you may shop audio and’video in relational database, it’s far advised not to save in
the relational database). In this model, two are greater items are connected via links.

O
We use this hyperlink to relate one item to different items. Dml (records manipulation
language):
The square instructions that offers with the manipulation of facts present in the

ty
database belong to dml or records manipulation language and this consists of
maximum of the sq. Statements. It’s far the element of the sq. Assertion that controls
access to statistics and to the database. Basically, dcl statements are grouped with
dml statements. Listing of dml’commands:

si
◌◌ Insert: it’s far used to insert data into a desk.
◌◌ replace: it is used to replace existing statistics inside ’ table.
◌◌ r
Delete: it’s far used to delete facts from a database desk.
ve
◌◌ Lock: desk control concurrency.
◌◌ Call: name a pl/square or java subprogram.
◌◌ provide an explanation for plan: it describes the get admission to path to
records.
ni

Activity
1. Formulate a case study describing the importance of Database Three Level
U

Architecture.

Questions & Exercises

ity

1. What are DML Commands and write its commands.

2. What are Data Models? Describe its types.
3. What is the importance of Data Independence?
4. What is the difference between Logical Data Independence and Physical Data
m

Independence?
5. What is the difference between schema and Instances?
)A

Glossary
Schema: a database schema is the skeleton shape that represents the logical view
of the complete database. It defines how the information is prepared and the way the
relations amongst them are related. It formulates all of the constraints that are to be
(c

implemented on the records.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 59

Instances: Instances are the collection of information saved at a particular moment.

Notes

e
The times can be modified by way of certain crud operations as like addition, deletion
of statistics. It can be noted that any search question will now not make any kind of
changes within the times.

in
Data independence: statistics independence is defined as an assets of DBMS
that helps you to alternate the database schema at one stage of a database machine

nl
without requiring to exchange the schema at the following higher stage. Records
independence lets you hold data separated from all programs that make use of it.
Information definition language: ddl stands for statistics definition language. It’s far used
to outline database structure or pattern. It’s far used to create schema, tables, indexes,

O
constraints, and so forth. Inside the database.

The use of the ddl statements, you could create the skeleton of the database.
Facts definition language is used to save the data of metadata like the variety of tables

ty
and schemas, their names, indexes, columns in each desk, constraints, and so on.
Right here are a few duties that come underneath ddl:

◌◌ Create: it’s far used to create items inside the database.

si
◌◌ Regulate: it’s far used to modify the structure of the database.
◌◌ Drop: its miles used to delete objects from the database.
◌◌ r
Truncate: it’s far used to eliminate all records from a table.
ve
◌◌ Rename: it is used to rename an item.
◌◌ Remark: it is used to touch upon the facts dictionary. These instructions are
used to replace the database schema it is why they come beneath statistics
definition language. Records manipulation language
ni

Dml stands for data manipulation language. It’s far used for gaining access to and
manipulating records in a database. It handles person requests. Right here are a few
tasks that come underneath dml:
U

◌◌ Select: its miles used to retrieve data from a database.

◌◌ Insert: it is used to insert statistics right into a table.
◌◌ Update: it is used to update present information inside’a desk.
ity

◌◌ Delete: it’s far used to delete all records from a desk.

◌◌ Merge: it plays upset operation, I. E., insert or replace’operations.
◌◌ Name: it’s far used to name a based query language or a java subprogram.
m

◌◌ Give an explanation for plan: it has the parameter of explaining data.

◌◌ Lock table: it controls concurrency. Statistics manage language: dcl
stands’for records control language. It’s far used to retrieve the stored or
)A

saved information. The dcl execution is transactional. It also has rollback

parameters. Here are a few obligations that come below dcl:
◌◌ Grant: it is used to present user get entry to privileges to a database.
◌◌ Revoke: it’s far used to take back permissions from the user.
◌◌ There are the subsequent operations which have the authorization of revoke:
(c

connect, insert, usage, execute, delete, update and choose. Transaction

Amity Directorate of Distance & Online Education

 60 Introduction to Data Base Management Systems

control language: tcl is used to run the adjustments made through the dml
Notes

e
declaration. Tcl may be grouped into a logical transaction. Right here are a
few obligations that come below tcl:

in
◌◌ Devote: its miles used to shop the transaction at the datab’se.
◌◌ Rollback: it’s far used to repair the database to unique for the reason that
closing dedicate.

nl
◌◌ Graphical user interfaces. A GUIS normally shows a schema to the person in
diagrammatic shape. The person then can specify a query via manipulating
the diagram. In lots of cases, GUISs utilize both menus and bureaucracy.

O
◌◌ Most GUISs use a pointing tool, which include a mouse, to pick certain
components of the displayed schema diagram.
◌◌ Forms-based totally interfaces. A forms-based interface shows a shape
to each user. Customers can fill out all of the form entries to insert new

ty
information, or they could fill out handiest certain entries, wherein case the
DBMS will retrieve matching statistics for the remaining entries.
◌◌ Paperwork are norm naïve designed and programmed for naive users as

si
inter-faces to canned transactions.
◌◌ Menu-based interfaces for web customers or surfing. Those interfaces pre-
sent the consumer with lists of options (referred to as menus) that lead the
r
consumer via the method of a request. Menus cast off the need to memorize
ve
the unique commands and syntax of a query language; alternatively, the
query consists step-with the aid of-step via selecting alternatives from a
menu this is displayed via the machine. Natural language interfaces. These
interfaces accept requests written in English or a few other language and
ni

try to understand them. An herbal language interface generally has its very
own schema that is similar to the database conceptual schema, as well as
a dictionary of important phrases. The natural language interface refers to
U

the phrases in its schema, as well as to the set of widespread words in its
dictionary, to interpret the request.

Further Readings
ity

1. Carlos Coronel and Steven Morris. (2018). Database Systems: Design,

Implementation and Management. Thirteenth Edition. Course Technology, Inc.
2. Hectar Garcia-Molina, Jeff Ullman and Jennifer Widom. (2009). Database
Systems: The Complete Book. Second Edition. Pearson Education Inc.
m

3. Thomas Connolly and Carolyn Begg. (2014). Database Systems: A Practical

Approach to Design, Implementation and Management. Sixth Edition. Pearson
Education, Inc.
)A

Check your Understanding-Answers

A. Fill in the Blanks
1. External level
(c

2. Instances

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 61

3. Data Definition Language

Notes

e
4. Pointers
5. Attribute

in
B. True / False
1. True

nl
2. True
3. False

O
4. True
5. False

ty
si
r
ve
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 62 Introduction to Data Base Management Systems

Unit - 1.3: Database Users

Notes

e
Objectives

in
At the end of this unit, you will be able to:

●● Know about types of database users

nl
●● Understand the responsibility of database administrator

1.3.1 Types of Database Users

O
There are some of users who can access or retrieve the information on demand
the use of the utility and the interfaces supplied through the database management
machine (DBMS). Database administrator (dba):

ty
Database administrator (dba) is a person/crew who defines the schema and also
controls the 3 degrees of database. The dba will then create a brand new account id
and password for the user if he/she need to get right of entry to the information base.

si
Dba is likewise chargeable for providing security to the facts base and he lets in best
the legal customers to get admission to/alter the records base.

●● Dba also monitors the restoration and lower back up and offer technical help.
●●
r
The dba has a dba account inside the DBMS which called a system or super user
ve
account.
●● Dba repairs harm caused because of hardware and/or software disasters.
a. Naive / parametric give up users: Parametric stop customers are the
ni

unsophisticated who don’t have any DBMS information but they frequently
use the statistics base applications in their daily lifestyles to get the preferred
outcomes. For examples, railway’s ticket reserving users are naive users.
U

Clerks in any financial institution is a naive consumer due to the fact they don’t
have any DBMS information however they nevertheless use the database and
perform their given mission.
b. Device analyst: Gadget analyst is a user who analyses the requirements of
ity

parametric cease customers. They check whether all of the requirements of

end customers are happy.
c. State-of-the-art users: State-of-the-art customers may be engineers,
scientists, business analyst, who are familiar with the database. They
m

could broaden their personal records base applications in step with their
requirement. They don’t write this system code but they have interaction the
statistics base with the aid of writing sq. Queries at once through the question
)A

processor.
d. Records base designers: Facts base designers are the customers who layout
the structure of records base which incorporates tables, indexes, views,
constraints, triggers, saved tactics. He/she controls what records must be
stored and how the statistics gadgets to be related.
(c

e. Utility program: Application software are the returned quit programmers who
writes the code for the application packages. They’re the computer experts.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 63

These applications might be written in programming languages which includes

Notes

e
visual primary, developer, c, FORTRAN, COBOL etc.
f. Casual users / temporary users: Informal customers are the customers who

in
once in a while use/access the information base however every time once
they access the information base they require the new statistics, for example,
centre or better degree supervisor.

nl
1.3.2 Responsibility of Database Administrator
A database administrator’s (dba) primary task is to make certain that statistics is to

O
be had, blanketed from loss and corruption, and without problems available as needed.
Underneath are some of the chief obligations that make up the everyday paintings of a
dba. Dsp deliver an outsourced dba provider within the UK, providing oracle help and
square server support; even as mind set and toolset may be exceptional, whether a

ty
database is living on-premise or in a public / private cloud, the position of the dba isn’t
always that extraordinary.

1. Software program installation and maintenance: A dba regularly collaborates on the

si
initial set up and configuration of a new oracle, sq. Server and so forth database.
The device administrator sets up hardware and deploys the operating machine for
the database server, then the dba installs the database software and configures it
r
to be used. As updates and patches are required, the dba handles this on-going
ve
preservation. And if a new server is wanted, the dba handles the switch of records
from the existing machine to the brand new platform.
2. Data extraction, transformation, and loading: Known as etl, facts extraction,
transformation, and loading refers to correctly importing big volumes of records
ni

which have been extracted from multiple structures right into a statistics warehouse
surroundings. This outside statistics is cleaned up and converted to healthy the
favoured format in order that it could be imported into a relevant repository.
U

3. Specialised data managing: These days’s databases can be massive and might
comprise unstructured statistics sorts which include pics, files, or sound and video
documents. Managing a completely huge database (vldb) may additionally require
higher-stage talents and extra monitoring and tuning to preserve performance.
ity

4. Database backup and recovery: Dbas create backup and recuperation plans and
tactics based totally on enterprise quality practices, then make certain that the
essential steps are accompanied. Backups value money and time, so the dba may
have to steer management to take necessary precautions to maintain records.
m

System admins or other personnel may additionally definitely create the backups,
but its miles the dba’s responsibility to make certain that the entirety is executed on
agenda. Inside the case of a server failure or different shape of facts loss, the dba
)A

will use current backups to restore lost information to the machine. Extraordinary
types of disasters may also require different recuperation techniques, and the dba
must be organized for any eventuality. With technology exchange, it’s far becoming
ever greater ordinary for a dba to backup databases to the cloud, oracle cloud for
oracle databases and MS azure for square server.
(c

5. Safety: A dba wishes to understand capacity weaknesses of the database software

and the agency’s general system and work to minimise dangers. No machine is

Amity Directorate of Distance & Online Education

 64 Introduction to Data Base Management Systems

100 per cent proof against attacks, but implementing fine practices can minimise
Notes

e
dangers. Within the case of a safety breach or irregularity, the dba can seek advice
from audit logs to look who has carried out what to the data. Audit trails are also
important while working with regulated records.

in
6. Authentication: Putting in place worker get right of entry to is a vital issue of database
safety. Dbas control who has get entry to and what kind of get entry to they’re allowed.

nl
As an instance, a person can also have permission to peer only certain portions of
statistics, or they may be denied the capacity to make changes to the machine.
7. Ability making plans: The dba needs to know how huge the database currently is and

O
how rapid it is developing with a purpose to make predictions about future wishes.
Garage refers to how a lot room the database takes up in server and backup area.
Capacity refers to usage level. If the company is growing quickly and adding many
new users, the dba ought to create the capacity to address the more workload.

ty
8. Overall performance monitoring: Tracking databases for performance issues is part
of the on-going machine protection a dba plays. If a few part of the gadget is slowing
down processing, the dba may additionally need to make configuration changes

si
to the software program or upload additional hardware potential. Many styles of
monitoring tools are available, and a part of the dba’s process is to apprehend what
they need to tune to enhance the machine. Third birthday party organisations may
r
be ideal for outsourcing this thing, however ensure they offer cutting-edge dba help.
ve
9. Database tuning: Performance monitoring suggests wherein the database must be
tweaked to function as efficaciously as feasible. The physical configuration, the way
the database is indexed, and how queries are treated can all have a dramatic impact
on database performance. With effective monitoring, it’s far viable to proactively
ni

track a gadget based on application and utilization in place of ready till a problem
develops.
10. Troubleshooting: Dbas are on call for troubleshooting in case of any issues. Whether
U

or not they want to quickly restore lost data or accurate a difficulty to minimise harm,
a dba desires to speedy recognize and respond to troubles when they arise.

Check your Understanding

ity

A. Fill in the Blanks

1. ___________ is a person/team who defines the schema.
2. ___________ are the unsophisticated Users.
m

3. ___________ is a user who analyses the requirements of parametric end users.

4. DBA primary job is to ensure that data is _________, __________ from loss and
corruption
)A

5. ________ is also responsible for providing security to the data base and he allows
only the authorized users to access/modify the data base.
B. State True or False
1. DBA is also responsible for providing security to the data base and he allows only the
(c

authorized users to access/modify the data base.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 65

2. Parametric End Users are the sophisticated Users.

Notes

e
3. Sophisticated users can beClerks, Railway’s ticket booking users.
4. Application Program are the back end programmers who writes the code for the

in
application programs.
5. Casual Users are the users who design the structure of data base which includes
tables, indexes, views, constraints, triggers, stored procedures.

nl
Summary
There are a number of users who can get admission to or retrieve the records on

O
call for the use of the application and the interfaces supplied by way of the database
management gadget (DBMS).

1. Database administrator (dba): Database administrator (dba) is a person/team who

ty
defines the schema and also controls the three levels of database. The dba will then
create a new account identity and password for the person if he/she need to access
the records base. Dba is likewise accountable for imparting protection to the facts

si
base and he permits handiest the legal users to access/alter the information base.
2. Naive / parametric give up users: Parametric quit users are the unsophisticated
who don’t have any DBMS expertise but they frequently use the information base
r
packages of their everyday life to get the desired effects. For examples, railway’s
price ticket reserving users are naive customers.
ve
3. System analyst: Gadget analyst is a person who analyses the necessities of
parametric quit customers. They check whether all the necessities of cease users
are happy.
ni

4. Sophisticated users: Sophisticated customers can be engineers, scientists,

enterprise analyst, who’re familiar with the database. They are able to develop their
own statistics base packages in keeping with their requirement. They don’t write this
U

system code however they have interaction the records base by means of writing
square queries directly through the question processor.
5. Statistics base designers: Facts base designers are the users who design the shape
ity

of records base which includes tables, indexes, perspectives, and constraints,

triggers, saved tactics. He/she controls what facts ought to be saved and the way
the statistics gadgets to be related.
6. Utility application: Utility software are the again end programmers who writes
the code for the software programs. They may be the computer professionals.
m

Those programs could be written in programming languages consisting of visual

fundamental, developer, c, FORTRAN, COBOL and so on.
)A

7. Informal users / temporary customers: Casual users are the customers who once
in a while use/get entry to the records base however on every occasion after they
get right of entry to the records base they require the new information, for example,
centre or higher level manager. A database administrator’s (dba) primary activity is to
make sure that statistics is available, blanketed from loss and corruption, and without
(c

problems available as wished.

Amity Directorate of Distance & Online Education

 66 Introduction to Data Base Management Systems

Underneath are some of the leader duties that make up the daily paintings of a dba.
Notes

e
1. Software set up and maintenance: a dba regularly collaborates at the preliminary
installation and configuration of a new oracle, square server and so forth database.

in
The device administrator units up hardware and deploys the operating system for
the database server, then the dba installs the database software and configures
it to be used. As updates and patches are required, the dba handles this on-going

nl
protection.
2. Statistics extraction, transformation, and loading: referred to as etl, records extraction,
transformation, and loading refers to correctly uploading big volumes of statistics

O
which have been extracted from a couple of systems into an information warehouse
environment.
3. Specialized records managing: nowadays databases can be large and might contain
unstructured facts sorts which include pics, documents, or sound and video files.

ty
Dealing with a completely massive database (vldb) may additionally require higher-
degree capabilities and further monitoring and tuning to hold performance.
4. Database backup and recuperation: dbas create backup and restoration plans and

si
techniques primarily based on enterprise first-rate practices, then ensure that the vital
steps are accompanied. Backups fee money and time, so the dba may additionally
have to steer control to take necessary precautions to keep facts.
5.
r
Safety: a dba desires to realize capability weaknesses of the database software and
ve
the organization’s standard machine and paintings to minimise dangers. No gadget
is 100 per cent proof against attacks, but implementing nice practices can minimise
dangers.
6. Authentication: setting up worker get entry to is an important component of database
ni

security. Dbas manipulate who has get admission to and what kind of get admission
to they may be allowed. As an instance, a person can also have permission to peer
only positive pieces of facts, or they will be denied the ability to make modifications
U

to the machine.
7. Potential planning: the dba needs to recognize how huge the database presently is
and the way speedy it is growing that allows you to make predictions about destiny
ity

desires. Storage refers to how plenty room the database takes up in server and
backup space. Potential refers to utilization level.
8. Overall performance monitoring: monitoring databases for overall performance
problems is part of the on-going machine renovation a dba plays. If some part of the
m

system is slowing down processing, the dba may also want to make configuration
changes to the software program or add additional hardware ability.
9. Database tuning: overall performance monitoring indicates in which the database
)A

ought to be tweaked to perform as correctly as feasible. The bodily configuration,

the way the database is listed, and how queries are handled can all have a dramatic
effect on database overall performance.
10. Troubleshooting: dbas are on call for troubleshooting in case of any problems.
Whether they need to quick repair misplaced statistics or accurate a trouble to
(c

minimise damage, a dba wishes to fast recognize and reply to troubles after they
occur.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 67

Activity
Notes

e
1. Read about the Types of Database Users and prepare a report on it.

in
Questions & Exercise
1. Define database administrator.
2. Who’re unsophisticated users?

nl
3. Give an explanation for database backup and healing.
4. Provide an explanation for the sorts of database users.

O
5. Provide an explanation for the duties of database administrator.

Glossary

ty
●● Database administrator (dba): database administrator (dba) is a person/crew
who defines the schema and additionally controls the 3 levels of database. Dba
is likewise accountable for offering security to the records base and he lets in
simplest the authorized users to get right of entry to/alter the data base.

si
●● Naive / parametric stop users: parametric stop users are the unsophisticated who
don’t have any DBMS know-how however they regularly use the information base

r
applications of their everyday existence to get the favoured results. For examples,
railway’s price ticket booking customers are naive users. Clerks in any bank is a
ve
naive consumer due to the fact they don’t have any DBMS knowledge however
they nevertheless use the database and carry out their given task.
●● Device analyst: gadget analyst is a user who analyses the requirements of
parametric give up customers. They check whether all of the requirements of stop
ni

users are happy.

●● Sophisticated users: state-of-the-art users may be engineers, scientists,
U

commercial enterprise analyst, who are familiar with the database. They can
develop their own records base programs according to their requirement. They
don’t write the program code but they interact the records base with the aid of
writing square queries at once through the query processor.
ity

●● Information base designers: facts base designers are the users who layout the
structure of statistics base which incorporates tables, indexes, views, constraints,
triggers, saved processes. He/she controls what records need to be saved and the
way the information items to be related.
m

●● Utility application: utility software are the back end programmers who writes the
code for the software applications. They’re the computer professionals. These
applications will be written in programming languages which includes visual
)A

simple, developer, c, FORTRAN, COBOL and so forth.

●● Informal customers / brief customers: casual users are the customers who
occasionally use/get admission to the facts base however whenever when they
get entry to the statistics base they require the brand new statistics, for example,
middle or better degree supervisor.
(c

Amity Directorate of Distance & Online Education

 68 Introduction to Data Base Management Systems

Further Readings
Notes

e
1. Thomas Connolly and Carolyn Begg. (2014). Database Systems: A Practical
Approach to Design, Implementation and Management. Sixth Edition. Pearson

in
Education, Inc.
2. Raghu Ramakrishnan and Johannes Gehrke. (2003). Database Management
Systems. Third Edition. McGraw Hill

nl
Check your Understanding-Answers
A. Fill in the Blanks

O
1. Database Administrator
2. Parametric End Users
3. System Analyst

ty
4. available, protected
5. DBA

si
B. True or False
1. True
2. False r
ve
3. False
4. True
5. False
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 69

Module - 2: Relational Database & ER Model

Notes

e
Structure:

in
2.1 Relational Model
2.1.1 Introduction

nl
2.1.2 Relational System Concept
2.1.3 Properties of Relation

O
2.1.4 CODD’s Rules
2.1.5 Relational Model
2.1.6 Data Modeling using Entity Relationship Model

ty
2.1.7 Elements of E-R Model
2.1.8 Entity and Entity Types, Entity Set
2.1.9 Attribute Concept and Types of Attributes

si
2.1.10 Relationship Types
2.1.11 Role and Structural Constraints
2.2 Enhanced Entity -Relationship Model r
ve
2.2.1 Enhanced E-R Model Concepts
2.2.2 Subclasses, Superclasses and Inheritance
2..2.3 Generalization
ni

2.2.4 Specialization
2.2.5 Constraints and Characteristics of of EER Model
U

2.2.6 Modeling of UNION types using Categories

Case Study
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 70 Introduction to Data Base Management Systems

Unit -2.1: Relational Model

Notes

e
Objectives

in
At the end of this unit, you will be able to:

●● Know about introduction to Relational system and its concept.

nl
●● Learn about properties of the relation.
●● Understand CODD’s Rules.

O
●● Know about relational model.
●● Understand data modeling using entity relationship model.
●● Know about elements of E-R model.

ty
●● Understand about Entity and Entity Types an Entity set.
●● Learn about Attribute concept and types of attributes.
●● Know about Relationship types.

si
●● Understand about Role and structural constraints.

2.1.1: Introduction r
ve
Relational database is a sort of database that stores and gives access to data
points that are identified with each other.

Relational Model addresses how data is put away in Relational Databases. A

relational database stores data as relations (tables).
ni

This model arranges data into at least one tables (or “relations”) of columns and
rows, with a unique key recognizing each line. Rows are additionally called records or
U

tuples. Columns are additionally called attributes.

Consider a relation STUDENT with attributes ROLL_NO, NAME, ADDRESS,

PHONE and AGE shown in table.
ity

Student

Roll_ Name Address Phone Age

no
m

1 R Delhi 9455123451 18
2 Resh Gurgaon 9652431543 18
3 Sujit Rohtak 9156253131 20
)A

4 Suresh Delhi 18

What is RDBMS?
(c

RDBMS represents Relational Database Management System. RDBMS data

is organized in database tables, fields and records. Each RDBMS table comprises of

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 71

database table rows. Every database table line comprises of at least one database
Notes

e
table fields. RDBMS store the data into assortment of tables, which may be
connected by normal fields (database table columns). RDBMS likewise give relational
administrators to control the data put away into the database tables. Most RDBMS

in
use SQL as database query language. The most famous RDBMS are MS SQL Server,
DB2, Oracle and MySQL. The relational model is an illustration of record-based
model. Record based models are so named in light of the fact that the database is

nl
organized in fixed arrangement records of a few kinds. Each table contains records of
a specific kind. Each record type characterizes a proper number of fields, or qualities.
The columns of the table relate to the traits of the record types. The relational data

O
model is the most generally utilized data model, and a larger part of current database
systems depend on the relational model. The relational model was planned by the
IBM research researcher and mathematician, Dr. E.F.Codd. Numerous advanced
DBMS don’t adjust to the Codd’s meaning of an RDBMS, yet in any case they are

ty
as yet viewed as RDBMS. Two of Dr.Codd’s super focal points when designing the
relational model were to additionally diminish data redundancy and to further develop
data integrity inside database systems. Prophet/2 the relational model started from a

si
paper created by Dr.codd named “A Relational Model of Data for Large Shared Data
Banks”, written in 1970. This paper incorporated the accompanying ideas that apply to
database management systems for relational databases. The connection is the main

r
data structure utilized in the relational data model to address the two elements and
connections between them.
ve
Difference between DBMS & RDBMS
A DBMS has to be persistent, that is it should be accessible when the progR
created the data ceases to exist or even the application that created the data restarted.
ni

A DBMS also has to provide some uniform methods independent of a specific

application for accessing the information that is stored. RDBMS is a Relational Data
Base Management System Relational DBMS. This adds the additional condition
U

that the system supports a tabular structure for the data, with enforced relationships
between the tables. This excludes the databases that don’t support a tabular structure
or don’t enforce relationships between tables. You can say DBMS does not impose any
constraints or security with regard to data manipulation it is user or the programmer
ity

responsibility to ensure the ACID PROPERTY of the database whereas the RDBMS is
more with this regard because RDBMS define the integrity constraint for the purpose of
holding ACID PROPERTY.
m

2.1.2 Relational System Concept

Important Terminologies
)A

●● Attributes is the property that define a relation. e.g.; ROLL_NO, NAME

●● A relation schema represents name of the relation with its attributes. E.g.:
STUDENT - ROLL_NO, NAME, ADDRESS, PHONE and AGE is relation schema
for STUDENT. If a schema has more than 1 relation, it is called Relational
(c

Schema.
●● Each row in the relation is known as Tuple. The above relation contains 4 tuples,
one of which is shown as:
Amity Directorate of Distance & Online Education
 72 Introduction to Data Base Management Systems

◌◌ The set of tuples of a relation at a particular instance of time is called as

Notes

e
relation instance.
◌◌ The number of attributes in the relation is known as degree of the relation.

in
The STUDENT relation defined above has degree 5.
◌◌ The number of tuples in a relation is known as cardinality. The STUDENT
relation defined above has cardinality 4.

nl
◌◌ Column represents the set of values for a particular attribute. The column
ROLL_NO is extracted from relation STUDENT.
Roll_no

O
1 R DELHI 9455123451 18
2
3

ty
4
◌◌ The value which is not known or unavailable is called NULL value. It is

si
represented by blank space. e.g.; PHONE of STUDENT having ROLL_NO 4
is NULL.

r
Constraints in Relational Model:
While designing Relational Model, we define some conditions which must hold for
ve
data present in database are called Constraints.

These constraints are checked before performing any operation in database. If

there is a violation in any of constrains, operation will fail.
ni

Domain Constraints: An attribute can only take values which lie inside the domain
range. e.g., If a constrains AGE>0 is applied on STUDENT relation, inserting negative
value of AGE will result in failure.
U

Key Integrity: Every relation in the database should have at least one set of
attributes which defines a tuple uniquely. Those set of attributes is called key. e.g.;
ROLL_NO in STUDENT is a key. No two students can have same roll number. So a key
ity

has two properties:

◌◌ It should be unique for all tuples.

◌◌ It can’t have NULL values.
Referential Integrity: When one attribute of a relation can only take values from
m

other attribute of same relation or any other relation, it is called referential integrity. Let
us suppose we have 2 relations
)A

Student

Roll_ Name Address Phone Age B_code

no
1 R Delhi 9455123451 18 Cs
(c

2 Resh Gurgaon 9652431543 18 Cs

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 73

3 Sujit Rohtak 9156253131 20 Ece

Notes

e
4 Suresh Delhi 18 It

in
Branch

B_code B_name

nl
Cs Computer Science
It Information Technology
Ece Electronics And Communication Engineering

O
B_CODE of STUDENT can only take the values which are present in B_CODE of
BRANCH which is called referential integrity constraint.

ty
The relation which is referencing to other relation is called REFERENCING
RELATION (STUDENT in this case) and the relation to which other relations refer is
called REFERENCED RELATION (BRANCH in this case).

si
2.1.3 Properties of Relation
Relational Database Management Systems (RDBMS) are efficient and a common
r
choice for storing financial records, logistical information, personnel data, and other
ve
information in new databases. Because they are easier to understand and use than
NoSQL databases, relational databases also frequently replace legacy hierarchical
databases and network databases.

Relational databases have the following properties:

ni

◌◌ Values are atomic.

◌◌ All of the values in a column have the same data type.
U

◌◌ Each row is unique.

◌◌ The sequence of columns is insignificant.
◌◌ The sequence of rows is insignificant.
ity

◌◌ Each column has a unique name.

◌◌ Integrity constraints maintain data consistency across multiple tables.

2.1.4 CODD’s Rules

m

CODD’s Rules proposed by E.F. Codd are given below:

1. Foundation Rule: For any system that is advertised as, or claimed to be, a relational
data base management system, that system must be able to manage data bases
)A

entirely through its relational capabilities.

2. Information Rule: Data stored in Relational model must be a value of some cell of a
table.
3. Guaranteed Access Rule: Every data element must be accessible by table name, its
(c

primary key and name of attribute whose value is to be determined.

Amity Directorate of Distance & Online Education

 74 Introduction to Data Base Management Systems

4. Systematic Treatment of NULL values: NULL value in database must only correspond
Notes

e
to missing, unknown or not applicable values.
5. Active Online Catalog: Structure of database must be stored in an online catalog

in
which can be queried by authorized users.
6. Comprehensive Data Sub-language Rule: A database should be accessible by a
language supported for definition, manipulation and transaction management

nl
operation.
7. View Updating Rule: Different views created for various purposes should be
automatically updatable by the system.

O
8. High level insert, update and delete rule: Relational Model should support insert,
delete, update etc. operations at each level of relations. Also, set operations like
Union, Intersection and minus should be supported.

ty
9. Physical data independence: Any modification in the physical location of a table
should not enforce modification at application level.
10. Logical data independence: Any modification in logical or conceptual schema of a

si
table should not enforce modification at application level. For E g, merging of two
tables into one should not affect application accessing it which is difficult to achieve.
11. Integrity Independence: Integrity constraints modified at database level should not
r
enforce modification at application level.
ve
12. Distribution Independence: Distribution of data over various locations should not be
visible to end-users.
13. Non-Subversion Rule: Low level access to data should not be able to bypass integrity
ni

rule to change data.

2.1.5 Relational Model

U

The database is represented as a set of relations in the Relational Model (RM).

A table of values is all that a relation is. Each table row indicates a grouping of linked
data values. The table’s rows represent a real-world entity or relationship. The table
and column names aid in deciphering the significance of each row’s values. A set of
ity

relations is used to represent the data. Tables are used to hold data in the relational
model. The physical storage of data, on the other hand, is unrelated to the logical
organization of the data.

The relational model (RM) for database management is an approach to data

m

management that uses a structure and language consistent with first-order predicate
logic, which was first described in 1969 by English computer scientist Edgar F. Codd,
and in which all data is represented in terms of tuples, which are then grouped into
)A

relations. A relational database is one that is organized using the relational paradigm.

The relational model’s purpose is to provide a declarative method for specifying

data and queries: users state directly what information the database contains and what
information they want from it, and the database management system software handles
(c

describing data structures for storing the data and retrieval procedures for answering
queries.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 75

The SQL data definition and query language is used by most relational databases;
Notes

e
these systems implement what can be considered an engineering approximation to
the relational paradigm. A predicate variable corresponds to a table in a SQL database
schema; a table’s contents to a relation; key constraints, other constraints, and SQL

in
queries to predicates. SQL databases, on the other hand, deviate from the relational
model in a number of ways, and Codd was adamant about not compromising the basic
ideas.

nl
2.1.6: Data Modeling using Entity Relationship Model

O
Data modeling is a method to report a product device making use of diagrams and
symbols. It is utilized to address verbal exchange of facts. The maximum big stage of
deliberation for the statistics model is called the entity dating diagram (ERD). Its miles
a graphical portrayal of facts requirements for a database. It is a giant level information

ty
model. This model is applied to signify the data factors and relationship for a predefined
machine.

2.1.7 Elements of E-R Model

si
There are 3 additives in ERD.

●● Entities: number of tables you need in your database.

●●
r
Attributes: facts along with property, information you want to explain every desk.
ve
●● Relationships: how tables are related together.

2.1.8 Entity and Entity Types, Entity Set

ni

Entity: An entity is an aspect in a true international with impartial life. An entity can
exist independently and is distinguishable from exceptional gadgets. It has a tendency
to be prominent uniquely. An entity might be enormous like a pupil, an e book, or
U

theoretical like an event or a selected concept. An entity is addressed by means of a

set of attributes. In a specific connection in RDBMS, a particular record is known as an
entity.
ity

Attributes: Attribute is a belongings or feature of an entity. An entity may

incorporate quite some of attributes. One of the attributes is taken into consideration
because the number one key. Like, table student has attributes like name, age, roll
wide variety and some extra. To uniquely understand the student, we utilize the primary
key as roll wide variety because it isn’t always repeated. Attributes can likewise be
m

partitioned into one more set of attributes.

Relationships: It’s far frequently the manner or greater records units are related.
)A

Relationships allow the datasets to share and shop the facts in separate tables. In
addition, they assist hyperlink disparate facts with each different.

Entity and Entity Kinds and Entity SET

Entity: Ann entity is a component in a real-global with independent existence. An
(c

entity can exist independently and is distinguishable from different items. It could be
diagnosed uniquely.

Amity Directorate of Distance & Online Education

 76 Introduction to Data Base Management Systems

An entity can be of two types:

Notes

e
◌◌ Tangible entity: entities that exist in the real world bodily. Instance: person,
vehicle, etc.

in
◌◌ Intangible entity: entities that exist best logically and haven’t any physical
existence. Instance: bank account, etc.

nl
Example:
◌◌ A student with a selected roll variety is an entity.
◌◌ An organization with a selected registration variety is an entity.

O
Entity type: It refers to the category that a particular entity belongs to. It is
represented by way of the call of the desk and its schema.

Instance:

ty
◌◌ A desk named pupil in a university database.
◌◌ A table named employee in an employer database.

si
Entity set: an entity set is a group or set of all entities of a selected entity type at
any factor in time. The form of all of the entities must be the identical.

Example: r
ve
◌◌ The collection of all the college students from the student table at a specific
immediate of time is an instance of an entity set.
◌◌ The collection of all the employees from the worker desk at a particular
immediate of time is an instance of an entity set.
ni

Difference Table:

Entity Entity Type Entity Set

U

A thing in the real world with A category of a particular Set of all entities of a
independent existence entity particular entity type.
Any particular row (a record) The name of a relation All rows of a relation (table)
ity

in a relation (table) is known (table) in RDBMS is an in RDBMS is entity set

as an entity. entity type

2.1.9 Attribute Concept and Types of Attributes

Attribute: It is an assets or characteristic of an entity. An entity may also include
m

any quantity of attributes. Student has attributes like name, age, roll variety and many
greater. To uniquely discover the pupil, we use the primary key as roll variety because it
isn’t repeated. There are 5 such varieties of attributes:
)A

1. Easy
2. Composite
3. Single-valued
(c

4. Multi-valued
5. Derived attribute.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 77

Types of Attributes
Notes

e
1. Simple attribute: A characteristic which cannot be in addition subdivided into
components is an easy attribute. Example: the roll variety of a scholar, the identity

in
variety of an employee.
2. Composite attribute: An attribute which can be splitted into components is a composite
attribute. Example: the deal with may be in addition splitted into residence quantity,

nl
street range, town, nation, usa and pin code, the call also can be splitted into first call
middle name and ultimate name.
3. Single-valued characteristic: The attribute which takes up simplest a single cost for

O
every entity instance is unmarried-valued attribute. Instance: the age of a pupil.
4. Multi-valued attribute: The attribute which takes up extra than a single value for every
entity example is multi-valued characteristic. Instance: telephone wide variety of a

ty
scholar: landline and mobile.
5. Derived attribute: An attribute that can be derived from other attributes is derived
attribute. Instance: total and common marks of a scholar. Relationship: it’s far in the

si
main the manner or extra data units are connected. Relationships allow the datasets
to proportion and save the records in separate tables.

2.1.10 Relationship Types r

ve
In addition, they assist hyperlink disparate statistics with each different. Courting
(relationship) types:

◌◌ One to 1 courting (1:1): it’s miles utilized to make a courting between tables
where a single row of the primary table need to be associated with one and
ni

only one statistics of a second desk. Moreover, the row of a 2nd table can
likewise be associated with each person row of the first desk. Following is the
example to expose a relational database, as proven beneath.
U

Customer (Table A) Order (Table B)

ity
m
)A

◌◌ One to many courting: it’s far utilized to make a relationship between two
tables. Any single rows of the primary table may be related to one or more
rows of the second one tables, but the rows of 2nd tables can most effective
relate to the only row within the first table. It is in any other case known as a
many to 1 courting. Illustration of 1 to many relational databases:
(c

Representation of One to Many relational databases:

Amity Directorate of Distance & Online Education

 78 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Representation of many to one relational database

ty
r si
ve
◌◌ Many too many Courtship: it is many to many relationships that make a
relationship between two tables. Every file of the first table can relate to
ni

any information (or no records) within the second desk. Likewise, each file
of the second one desk can likewise relate to more than one record of the
primary table. There are numbers (represented through m and n) written
U

above the traces which connect relationships and entities. Those are known
as cardinality ratios. These constitute the most range of entities that can be
related to every other thru relationship, r.
It is also represented an N: N relationship.
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 79

For example, there are many people involved in each project, and every person
Notes

e
can involve more than one project.

2.1.11 Role and Structural Constraints

in
Cardinality Ratios of relationships:

The entities are denoted by rectangle and relationships by diamond.

nl
Types of cardinality:

O
ty
There may be 4 kinds of cardinality –

1. One-to-one (1:1) –

si
When one entity in each entity set takes element at maximum once within the
dating, the cardinality is one-to-one.

2. One-to-many (1: n) –
r
ve
if entities in the first entity set participate in the relationship set at maximum
once and entities within the 2nd entity set participate normally (at the least twice), the
cardinality is said to be one-to-many.
ni

3. Many-to-one (n: 1) –
if entities in the first entity set participate in the relationship set regularly (at the
U

least two times), whilst entities within the second entity set take part at most as soon as,
the cardinality is said to be many-to-one.

4. Many-to-many (n: n) –
ity

The cardinality is stated to be many to many if entities in each the entity sets
participate frequently (at least two times) in the relationship set.
m
)A
(c

Amity Directorate of Distance & Online Education

 80 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
Participation Constraints:
r si
Participation constraints allow us to understand that that the participation in a
ve
relationship can both be general or partial. On the point when every entity in an entity
set participates in a relation, it is called overall participation. Be that as it is able to,
while all entities in the given entity set do keep away from a relation, it’s far called
partial participation.
ni
U
ity
m
)A

Structural constraints also are known as structural residences of a database

management system (DBMS). Cardinality ratios and participation constraints taken
together are known as structural constraints. The call constraints allude to the way
that such barriers have to be imposed on the records, for the DBMS framework to be
consistent with the necessities.
(c

The structural constraints are represented with the aid of min-max notation. This
is a pair of numbers (m, n) that seem on the connecting line between the entities and

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 81

their relationships. The minimum quantity of instances an entity can appear in a relation
Notes

e
is represented by way of m whereas, the most time it’s miles available is denoted by
way of n. If m is zero it signifies that the entity is participating within the relation partly,
whereas, if m is both more than or equal to 1, it denotes total participation of the entity.

in
nl
O
Check your understanding
A. Fill in the blanks:
1. In RDBMS a database is considered as a collection of___________

ty
2. In the RDBMS terminology, a record is called a ________.
3. Degree of a table means the number of ________ in a table.
4. A ________ is set of all possible data values.

si
5. A domain is a set of ________ values.
B. State true or false:
1.
r
The relational database model was created by E.F. Codd.
ve
2. In a database, data is stored in spreadsheets which have rows and columns.
3. A database has data and relationships
4. In the relational modes, cardinality is number of tuples...
ni

5. An entity set that does not have sufficient attributes to form a primary key is a strong
entity set.
U

Summary
●● In a relational model genuine objects are addressed in tables. Each table is made
out of rows and columns. Each row, otherwise called tuple or record, is made out
ity

of fields, otherwise called attributes. Each Attribute represents a specific element

of this present reality object. An attribute is characterized by a name and its value.
●● Relations between tuples address existing relationships between objects (tables).
Besides key attributes must be characterized (ordinarily showed underlined in a
m

connection). They are vital for the assignment (connection) of objects (tables) and
permit unique gets to tables.
●● Integrity or consistency represents the quality and reliability of data of a database
)A

system. A database is steady if the data mirrors the referred to objects effectively.
It is conflicting if there exist questionable or inconsistent tuples, relations or tables
in the database.
●● A connection model (conspire, entity) ought to reflect relationships that likewise
logically (in reality) have a place together. To stay away from anomalies various
(c

sorts of normalisations help keeping the database reliable.

Amity Directorate of Distance & Online Education

 82 Introduction to Data Base Management Systems

Activity
Notes

e
1. Prepare a script on the pros and cons of Relational System Concept.

in
Questions and exercises
1. What is Relational Model?
2. Define entity and attributes

nl
3. List the different types of entities and define them.
4. List the different types of attributes and define them

O
5. What are different types of model?

Glossary

ty
●● Attribute: A column of a table represents an attribute. It can also be described as
and a domain D has in a relation scheme R.
●● Candidate key: Each attribute or minimal combination of attributes that uniquely

si
identifies any tuple in a relation is called a candidate key. Minimal means that
removing an attribute leaves the key without the ability to uniquely identify any
tuple and therefore not being a candidate key anymore.
●● r
Domain: A domain D is a set of atomic values that defines the value range of
attributes.
ve
●● First normal form: A relation is in first normal form if every attribute in every row
can contain only one single (atomic) value.
●● Foreign key: An attribute in a relational scheme R1 is a foreign key if it is in
ni

relationship with a primary key from R2 and if:

1. The domain (value range) of the foreign key in R1 is the same as the domain
of the primary key in R2.
U

2. The set of values of the foreign key in R1 is a subset of all primary key values
in R2.
ity

Foreign keys usually are marked dotted underlined.

◌◌ Full functional dependency: We talk about full functional dependency if
attribute B is functional dependent on A, if A is a composite primary key and B
is not already functional dependent on parts of A.
m

◌◌ Functional dependency: If A and B are attributes of relation R, B is functionally

dependent on A (denoted A --> B), if each value of An in R is associated with
exactly one value of B in R.
)A

◌◌ Identification key: If every attribute B of R is functionally dependent of A, than

attribute A is a primary key.
◌◌ Primary key: The primary key is one chosen key candidate that acts as
the identification key for a relation. Usually this is a short attribute like an
ID-number (identification key) or username, Attributes of the primary are
(c

commonly underlined.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 83

◌◌ Relation: A relation is one instance of the relation scheme R (A1, A2, An)
Notes

e
containing a set of n-tuples r =({1, t2, ... , tn}.
◌◌ Relational database scheme: A relational database scheme is a set of relation

in
schemes S = {R(, ..., Rn} together with a set of integrity conditions.
◌◌ A relational database is a relational database scheme together with a
database instance.

nl
◌◌ Relation scheme: A relation scheme R (A1, A2, ..., An) is made up of a relation
name R and a list of attributes {A1, A2, ..., An}.
◌◌ Second normal form: A relation is in second normal form if it is in 1NF and

O
every non key attribute is fully functionally dependent on the primary key.
◌◌ Third normal form: A relation is in third normal form if it is in 2NF and no non
key attribute is transitively dependent on the primary key.

ty
◌◌ Transitive dependency: If A determines B and B determines C then C is
determined by (dependent on) A. We write A--> B and B --> C but not B >A.
◌◌ Tupel: A tuple t is a list with n values t = <d1, d2, ..., dn where each value di
is either an element of the domain Di or NULL. A tuple is a record in a relation

si
(row in a table).

Further Readings
1. r
Hectar Garcia-Molina, Jeff Ullman and Jennifer Widom. (2009). Database
ve
Systems: The Complete Book. Second Edition. Pearson Education Inc.
2. Ramez Elmasri and Shamkant B. Navathe. (2016). Fundamentals of Database
Systems. Seventh Edition. Pearson Education, Inc.
ni

Check your understanding (answers)

A. Fill in the blanks
U

1. Interrelated data
2. Tuple
3. the number of columns
ity

4. Domain
5. input values
B. State true or false
m

1. True
2. True
3. True
)A

4. True
5. False
(c

Amity Directorate of Distance & Online Education

 84 Introduction to Data Base Management Systems

Unit - 2.2: Enhanced Entity -Relationship Model

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Know about Enhanced E-R Model concepts.

nl
●● Understand subclasses, superclasses and Inheritance.
●● Learn about Generalization and Specialization.

O
●● Understand Constraints and Characteristics of EER Model.
●● Learn about Modeling of UNION types using Categories.

2.2.1 Enhanced E-R Model Concepts

ty
Enhances entity-relationship diagrams are advanced database diagrams very
similar to regular ER diagrams which represent requirements and complexities of
complex databases.

si
It is a diagrammatic technique for displaying the Sub class and super class;
specialization and generalization; union or category; aggregation etc.
r
ve
2.2.2 Subclasses, Superclasses and Inheritance
Sub class and super class relationship leads the concept of Inheritance.

The relationship between sub class and super class is denoted with symbol the
ni

given symbol.
U

The Enhanced Entity Relationship Model contains all the features of the Entity
Relationship model. In addition to all that, it also contains features of Subclasses,
Superclasses and Inheritance.

All of these in detail are as follows −

ity

Subclasses
A subclass is a class derived from the superclass. It inherits the properties of the
superclass and also contains attributes of its own.
m

Superclasses
A superclass is the class from which many subclasses can be created. The
)A

subclasses inherit the characteristics of a superclass. The superclass is also known as

the parent class or base class.
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 85

Notes

e
in
nl
O
Example

ty
Car, Truck and Motorcycle are all subclasses of the superclass Vehicle. They all
inherit common attributes from vehicle such as speed, colour etc. while they have
different attributes also i.e Number of wheels in Car is 4 while in Motorcycle is 2.

si
In the above example, Vehicle is the Superclass and its subclasses are Car, Truck
and Motorcycle.

r
ve
ni
U

Inheritance
ity

We make use of all of the above functions of er-version to make training of items
in item-oriented programming. The subtleties of entities are for the maximum element
hidden from the person, this cycle known as abstraction. Inheritance is a significant
element of generalization and specialization. It allows decrease-level entities to inherit
the attributes of better-degree entities.
m
)A
(c

Amity Directorate of Distance & Online Education

 86 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
For example, the attributes of a Person class such as name, age, and gender can
be inherited by lower-level entities such as Student or Teacher.

si
2.2.3 Generalization
Generalization is the process of generalizing the entities which include the homes
of all the generalized entities.

●●
r
It’s far a bottom technique, wherein two decrease degree entities integrate to
ve
shape a higher level entity.
●● Generalization is the opposite manner of specialization.
●● It defines a popular entity kind from a hard and fast of specialised entity type.
ni

●● It minimizes the difference between the entities by means of identifying the

common functions.
U

For example:
ity
m
)A

In the above example, Tiger, Lion, Elephant can all be generalized as Animals.

2.2.4 Specialization
(c

Specialization is a method that defines a set entities which is divided into sub
agencies based on their function.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 87

●● It’s miles a top down approach, in which one higher entity can be damaged down
Notes

e
into decrease stage entity.
●● It maximizes the difference between the individuals of an entity by figuring out the

in
precise feature or attributes of each member.
●● It defines one or greater sub class for the remarkable elegance and additionally
paperwork the superclass/subclass relationship.

nl
For example:

O
ty
si
In the above example, Employee can be specialized as Developer or Tester,
based on what role they playin an Organization. r
ve
Category or Union
Class represents an unmarried great magnificence or sub class relationship with
more than one super relationship.
ni

It may be a complete or partial participation.

As an instance vehicle reserving, vehicle owner may be a person, a financial

U

institution (holds an ownership on an automobile) or a business enterprise. Class

(sub elegance) → proprietor is a subset of the union of the 3 first-rate classes →
organization, financial institution, and individual. A class member should exist in at least
one in all its notable classes.
ity
m
)A

Aggregation
◌◌ Aggregation is a system that constitute a courting between a whole object and
(c

its component elements.

◌◌ It abstracts a relationship among gadgets and viewing the connection as an
object.
Amity Directorate of Distance & Online Education
 88 Introduction to Data Base Management Systems

◌◌ It’s miles a system while two entity is dealt with as a single entity.
Notes

e
in
nl
O
ty
In the above example, the relation between College and Course is acting as an
Entity in Relation with Student.

si
2.2.5 Constraints and Characteristics of EER Model
Following are the constraints on specialization/generalization:
r
There are three constraints that can follow to a specialization/generalization: club
ve
constraints, disjoint constraints and completeness constraints.

●● Membership constraints Condition described: club of a specialization/

generalization courting can be defined as a situation in the necessities User
ni

defined: every so often the dressmaker can outline the superclass-subclass

relationship. This could be executed to simplify the layout version or represent a
complicated relationship that exists among entities.
U

●● Disjoint constraint: The disjoint constraint best applies while a superclass has
multiple subclass. If the subclasses are disjoint, then an entity incidence can be
a member of simplest one of the subclasses, e. G. Postgrads or undergrads –
you can’t be both. To symbolize a disjoint superclass/subclass courting, oar€™ is
ity

used.
m
)A

Overlapping: this applies when an entity occurrence may be a member of multiple

subclass, e. G. Scholar and personnel a few human beings are each. Its miles used to
represent the overlapping
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 89

Notes

e
in
nl
Completeness constraints: Specialization/generalization relationship within the er
diagram. Every superclass (better-level entity) ought to belong to subclasses (lower-

O
level entity sets), e. G. A scholar have to be postgrad or undergrad. To represent
completeness in the specialization/generalization courting, the keyword Mandatory is
used.

ty
r si
ve
Partial: A few superclasses won’t belong to subclasses (decrease-level entity
sets), e. G. A few humans at uct are neither student nor team of workers. The key-word
optional is used to represent a partial specialization/generalization courting.
ni
U
ity

We can show both disjoint and completeness constraints in the ER diagram.

Following our examples, we can combine disjoint and completeness constraints.
m
)A

Some members of a university are both students and staff. Not all members of the
(c

university are staff and students

Amity Directorate of Distance & Online Education

 90 Introduction to Data Base Management Systems

Notes

e
in
nl
A student in the university must be either an undergraduate or postgraduate, but
not both.

O
ty
si
Mapping specialization/generalization to relational tables:

Specialization/generalization relationship can be mapped to relational tables in

r
three methods. To illustrate the methods, we can take the student, postgraduate and
ve
undergraduate courting. A student within the college has a registration range and a call.
Only postgraduate college students have supervisors. Undergraduates accumulates
factors via their coursework.

Approach 1: all the entities within the dating are mapped to person tables. Student
ni

(regno, name) posgrad (regno, supervisor) undergrad (regno, points)

Approach 2: best subclasses are mapped to tables. The attributes inside the
superclass are duplicated in all subclasses. Posgrad (regno, call, and supervisor)
U

undergrad (regno, call, points) this method is maximum desired when inheritance is
disjoint and complete, e. G. Every pupil is either posgrad or undergrad and no one is
both.
ity

Anapproach 3: handiest the superclass is mapped to a table. The attributes inside

the subclasses are taken to the superclass. 10 pupil (regno, call, manager, factors) this
approach will introduce null values. While we insert an undergraduate report inside the
desk, the supervisor column fee may be null. Within the equal way, whilst we insert a
postgraduate document in the table, the point’s value can be null.
m

ERD entity symbols:

)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 91

Notes

e
in
nl
O
ty
si
ERD relationship symbols
Inside entity-relationship diagrams, relationships are used to record the interplay
between entities. r
ve
Relationships are typically verbs which include assign, partner, or track and provide
useful facts that could not be discerned with simply the entity types...
ni
U
ity

ERD attribute symbols:

m

ERD attributes are characteristics of the entity that assist users to better apprehend
the database. Attributes are included to include information of the various entities which
might be highlighted in a conceptual er diagram.
)A
(c

Amity Directorate of Distance & Online Education

 92 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
r si
ve
Creating an Effective EERD
ni

A nicely-designed EERD will help you build storage structures which can be
long-lasting and beneficial. Do not forget the following whilst evaluating your entity
relationship diagram to be sure that you’re modeling a gadget design with the intention
to meet the necessities of your commercial enterprise:
U

◌◌ Stability: will the diagram support changing commercial enterprise desires?

◌◌ Breadth: can all the facts that we want to save be prepared within the
version?
ity

◌◌ Flexibility: can facts on this version be re-prepared to Guide new statistics

necessities?
◌◌ Efficiency: is that this model the simplest answer viable? Is the data modelled
with the proper symbols?
m

◌◌ Accessibility: can both creators and give up customers without difficulty

recognize your ear?
)A

◌◌ Conformity: will the version combine effortlessly along with your present
database structure?

Characteristics of EER Model

◌◌ EER creates a layout extra accurate to database schemas.
(c

◌◌ It displays the data homes and constraints more precisely.

◌◌ It consists of all modeling standards of the er model.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 93

◌◌ Diagrammatic method enables for showing the eer schema.

Notes

e
◌◌ It consists of the idea of specialization and generalization.
◌◌ It is used to represent a collection of gadgets this is union of items of different

in
of various entity types.

2.2.6 Modeling of UNION Types using categories

nl
Every now and then important to symbolize a single superclass/subclass courting
with multiple superclass, wherein the superclasses represent exclusive entity kinds.
In this case, the subclass will represent a set of objects that could be a subset of the

O
union of awesome entity sorts; we name such a subclass a union type or a class. As
an instance, suppose that we have 3 entity types: character, financial institution, and
agency. In a database for motor automobile registration, a proprietor of a car may be
someone, a bank (holding a lien on an automobile), or a corporation. We need to create

ty
a category (series of entities) that includes entities of all 3 types to play the function
of vehicle proprietor. A category (union kind) owner that is a subclass of the union of
the three entity units of company, financial institution, and individual can be created

si
for this motive. We display categories in an eer diagram as proven in determine eight.
The superclasses business enterprise, bank, and character are related to the circle
with the image, which stands for the set union operation. An arc with the subset symbol

r
connects the circle to the (subclass) owner class. If a defining predicate is needed,
its miles displayed subsequent to the road from the superclass to which the predicate
ve
applies.

In figure 8. we’ve got classes: owner, that’s a subclass of the union of man or
woman, bank, and business enterprise; and registered_vehicle, that is a subclass of the
union of car and truck.
ni

Check Your Understanding

U

A. Fill in the blanks:

1. The E-R model that is supported with the additional semantic concepts is called the
________.
ity

2. The relationship between a superclass and a subclasses is ________.

3. A total participation specifies that every member in the ________ must participate.
4. A partial participation constraint specifies that a member of a ________ need not
belong to any of its ________.
m

5. Attribute inheritance is the property by which attribute inherit values of all attributes
of the ________.
)A

B. State true or false:

1. Subclasses are the sub-grouping of occurrences of entities in an entity type that
shares common attributes or relationships distinct from other sub-groupings.
2. In case of super type, objects in one set are grouped or subdivided into one or more
(c

classes in many systems.

Amity Directorate of Distance & Online Education

 94 Introduction to Data Base Management Systems

3. The relationship between a superclass and a subclasses is a one-to-many (1: N)

Notes

e
relationship.
4. Attribute inheritance is the property by which super type entities inherit values of all

in
attributes of the subtype.
5. Each member of the subclass is also a member of the superclass.

nl
Summary
●● Draw the ER diagram for a small database for a bookstore. The database will
store information about books for sale. Each book has an ISBN, title, price and

O
short description. Each book is published by a publisher in a certain publishing
year. For each publisher, the database maintains the name, address and phone
number. Each book is written by one or more authors. For each author, the
database maintains his/her ID, name and a short introduction. Each book is

ty
stored in exactly one warehouse with a particular quantity. For each warehouse,
the database maintains the warehouse name, the location and the phone number.
Each book has one or more sellers, which may be either companies (corporate

si
vendors) or individuals (individual vendors). For each company, the database
maintains a name of the company, its address, its phone numbers (there could
be more than one phone number, each with a number and a description) and its

r
contact person. For each individual vendor, the database keeps a name, a phone
number and an email address. A contact person whose company sells a book
ve
cannot be selling the same book as an individual vendor at the same time (he/she
may sell other books as an individual seller).

Activity
ni

1. Prepare a concept map on the topic Constraints and Characteristics of EER Model.

Glossary
U

1. Super types and Subtypes

◌◌ Super type - an entity type that relates to one or more subtypes.
◌◌ Subtype - a subgroup of entities with unique attributes.
ity

◌◌ Inheritance - the concept that subtype entities inherit the values of all super
type attributes.
Note: subtype instances are also classified as super type instances.
2. Generalization & Specialization
m

◌◌ Generalization - the process of defining a general entity type from a collection

of specialized entity types.
)A

◌◌ Specialization - the opposite of generalization, since it defines subtypes of the

super type and determines the relationship between the two.
3. Constraints
◌◌ Disjointness constraints - You will need to decide whether a supertype
(c

instance may simultaneously be a member of two or more subtypes. The

disjoint rule forces subclasses to have disjoint sets of entities. The overlap rule

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 95

forces a subclass (also known as a supertype instance) to have overlapping

Notes

e
sets of entities.
◌◌ Completeness constraints - decide whether a supertype instance must also be

in
a member of at least one subtype. The total specialization rule demands that
every entity in the superclass belong to some subclass. Just as with a regular
ERD, total specialization is symbolized with a double line connection between
entities. The partial specialization rule allows an entity to not belong to any of

nl
the subclasses. It is represented with a single line connection.
4. Subtype Discriminators

O
A subtype discriminator is an attribute of the supertype that indicates an entity’s
subtype. The attribute’s values are what determine the target subtype.

◌◌ Disjoint subtypes - simple attributes that must have alternative values to

ty
indicate any possible subtypes.
◌◌ Overlapping subtypes - composite attributes whose subparts pertain to
various subtypes. Each subpart has a Boolean value that indicates whether or
not the instance belongs to the associated subtype.

si
5. ERD
◌◌ Conceptual Data Models establish a broad view of what should be included in
the model set. r
ve
◌◌ Entities are objects or concepts that represent important data. Entities are
typically nouns such as product, customer, location, or promotion.
◌◌ There are three types of entities commonly used in entity relationship
diagrams.
ni

◌◌ Strong entity,
◌◌ Weak entity and Associative entity
◌◌ Within entity-relationship diagrams, relationships are used to document the
U

interaction between two entities.

◌◌ Types of relationships: weak and strong
◌◌ Relationships are usually verbs such as assign, associate, or track and
ity

provide useful information that could not be discerned with just the entity
types.
◌◌ ERD attributes are characteristics of the entity that help users to better
understand the database. Attributes are included to include details of the
m

various entities that are highlighted in a conceptual ER diagram.

◌◌ EER creates a design more accurate to database schemas.
◌◌ It reflects the data properties and constraints more precisely.
)A

◌◌ It includes all modeling concepts of the ER model.

◌◌ Diagrammatic technique helps for displaying the EER schema.
◌◌ It includes the concept of specialization and generalization.
◌◌ It is used to represent a collection of objects that is union of objects of different
(c

of different entity types.

◌◌ Sometimes necessary to represent a single superclass/subclass relationship

Amity Directorate of Distance & Online Education

 96 Introduction to Data Base Management Systems

with more than one superclass, where the superclasses represent different
Notes

e
entity types. In this case, the subclass will represent a collection of objects
that is a subset of the UNION of distinct entity types; we call such a subclass
a union type or a category.

in
Further Readings
1. C.J. Date. (2004). An Introduction to Database Systems. Pearson Education,

nl
Inc.
2. Carlos Coronel and Steven Morris. (2018). Database Systems: Design,
Implementation and Management. Thirteenth Edition. Course Technology, Inc.

O
Check your understanding (answers):
Fill in the blanks:

ty
1. Enhanced ER Model
2. Property Inheritance

si
3. Entity set, relationship set
4. Entity set, relationship set
5. Parent product class
State true or false:
r
ve
1. True
2. True
3. True
ni

4. True
5. True
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 97

Case Study
Notes

e
Construct an ER Diagram for Company having following details: Company

in
organized into DEPARTMENT.

Each department has unique name and a particular employee who manages the
department. Start date for the manager is recorded. Department may have several

nl
locations. A department controls a number of PROJECT. Projects have a unique name,
number and a single location.

Company’s EMPLOYEE name, ssno, address, salary, sex and birth date are

O
recorded. An employee is assigned to one department, but may work for several
projects (not necessarily controlled by her dept). Number of hours/week an employee
works on each project is recorded; the immediate supervisor for the employee.

ty
Employee’s DEPENDENT are tracked for health insurance purposes (dependent
name, birthdate, relationship to employee).

r si
ve
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 98 Introduction to Data Base Management Systems

Module - 3: Relational Model Objects

Notes

e
Structure:

in
3.1 Understanding Relational Model
1.1.1 Domains and Relation

nl
1.1.2 Keys Concepts and Its Type
1.1.3 Relational Model Constraints

O
1.1.4 Domain Constraints
1.1.5 Key Constraints
1.1.6 Referential Key Constraints

ty
1.1.7 Update Operations and Dealing with Constraints Violations
3.2 Relational Model Operations and SQL Language

si
2.1.1 Relational Operators
2.1.2 Set Operations on Relation
2.1.3 Join Operations
2.1.4
r
SQL Language and Its Characteristics
ve
2.1.5 Data Definition Language and its Features
2.1.6 Data Manipulation Language and its Features
2.1.7 Data Definitions Commands-Create, Alter, Truncate, Drop
ni

2.1.8 Data Manipulation Commands-Select, update, delete

2.1.9 SQL String Functions
U

2.1.10 Views
2..1.11 SQL Joins
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 99

Unit - 3.1: Understanding Relational Model

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Understand about Domain and Relation

nl
●● Know about key concepts and its types and constraints
●● Understand Domain and Referential Key Constraints

O
●● Know about Relational Model Constraints
●● Understand about Update Operations and Dealing with Constraints Violations

3.1.1: Domains and Relation

ty
DOMAINS:
A domain is a unique set of values allowed for a trait in a table. As an example,

si
a website of month-of-year can acknowledge January, February…. December as
possible values, a website of integers can renowned entire numbers which can be
negative, fantastic and 0. Area constraints are person described records kind and we
can represent them like this: r
ve
Domain Constraint = data type + Constraints (NOT NULL / UNIQUE / PRIMARY
KEY / FOREIGN KEY / CHECK / DEFAULT)

Example:
ni

For example I want to create a table “student_infor” with “stu_id” field having vale
greater than 100, I can create a domain and table like this:
U
ity
m
)A

Another example:
(c

I want to create a tabe “bank_account” with “account_type” field having value either
“checkin’ or ‘saving’:

Amity Directorate of Distance & Online Education

 100 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Relationship:
Connection is at times used to allude to a table in a relational database however

ty
is all the more commonly used to painting the relationships that can be made among
the ones tables in a relational database. In relational databases, a dating exists among
two tables when considered one of them has a foreign key that references the primary

si
key of the opposite table. This unmarried truth allows relational databases to cut up
and save facts in diverse tables, yet nonetheless hyperlink the disparate information
gadgets together. It is one of the elements that makes relational databases such

r
excellent and effective stores of records. The ability to characterize relationships is
entirely fundamental hence good sized, to the point that that is the component that
ve
separates relational databases from other forms of databases, like flat-file databases.
Connection, consequently, is the characterizing detail of relational databases. Relation
will also be known as relationship.
ni

3.1.2: Keys Concepts and Its Type

What are Keys in DBMS?

U

Keys in DBMS is an attribute or set of attributes which assists you with identifying
a row (tuple) in a relation (table). They can help you discover the relation among tables.
Keys assist to uniquely perceive a row in a desk through a blend of at the least one
columns in that desk. Key’s additionally useful for locating novel file or row from the
ity

desk. Database secrets additionally beneficial for finding special file or row from
the table. A secret is applied inside the meanings of different varieties of integrity
constraints. A table in a database addresses a collection of records or events for a
specific relation. Currently there may be a great many such records, some of which
m

might be copied. There have to be a manner of figuring out every report independently
and uniquely, for example no copies. Keys permit us to be liberated from this problem.
Permit us to take a proper illustration of the database of every understudy considering
)A

in a designing school. What characteristic of the understudy do you observed will

uniquely identify each one in every of them? You could allude to an understudy by
utilizing their call, workplace, year and segment. Or however, you could specify just the
university roll range of the understudy, and you may get the huge variety of diverse
subtleties from that.
(c

Varieties of keys in DBMS

There are broadly seven types of keys in DBMS:
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 101

◌◌ Primary Key
Notes

e
◌◌ Candidate Key
◌◌ Super Key

in
◌◌ Foreign Key
◌◌ Composite Key
◌◌ Alternate Key

nl
◌◌ Unique Key

1. Primary Key

O
A primary key is a column of a desk or a hard and fast of columns that assists with
spotting each file present in that desk uniquely. There may be handiest one number one
key in a table. Moreover, the primary key cannot have the same values rehashing for

ty
any row. Every well worth of the primary key must be exclusive without any repetitions.
The primary key (pk) constraint placed on a column or set of columns may not allow
them to have any null values or any duplicates. One desk will have most effective one
primary key constraint. Any well worth in the primary key cannot be changed via any

si
overseas keys (clarified below) which allude to it.

Rules for defining Primary key:

◌◌ Two rows can’t have the same primary key feer

ve
◌◌ It should for every row to have a number one key fee.
◌◌ The primary key field can’t be null.
◌◌ The fee in a primary key column can never be changed or updated if any
foreign key refers to that number one key.
ni

Example:
In the following example, <code>studio</code.is a Prilmary key.
U

Studio Roll No First Name Last Name Email

1 11 Tom Price [email protected]
ity

2 12 Nick Wright [email protected]

3 13 Dans Natan [email protected]

2. Super Key
Super key’s the set of all of the keys which assist to distinguish rows in a table
m

uniquely. This means that that multitude of columns of a desk than equipped for spotting
the alternative columns of that desk uniquely can be typically considered as terrific
keys. Outstanding key is the superset of a candidate key (explained underneath). The
)A

primary key of a desk is picked from the super key set to be made the table’s identity
attribute.
(c

Amity Directorate of Distance & Online Education

 102 Introduction to Data Base Management Systems

Example:
Notes

e
EmpSSN EmpNum Empname
9812345098 AB05 Shown

in
9876512345 AB06 Roslyn
1999937890 AB07 James

nl
In the above-given example, EmpSSN and EmpNum name are superkeys.

3. Candidate Key

O
Candidate keys are those attributes that uniquely distinguish rows of a desk. The
number one key of a desk is selected from one of the candidate keys. On this way,
candidate keys have the equal residences because the vital keys clarified previously.

ty
There may be multiple candidate keys in a table. Candidate key Example: In the given
table Stud ID, Roll No, and email are candidate keys which help us to uniquely identify
the student record in the table.

si
Studio Roll No First Name Last Name Email
1 11 Tom Price [email protected]
2 12 Nick Wright [email protected]
3 13
r Dans Natan [email protected]
ve
ni
U
ity

4. Alternate Key

As stated above, a table can have a couple of picks for a primary key; however, it
is able to pick out best one. So, all the keys which did not become the number one key
m

are referred to as change keys.

Example:
)A

In this table, studID, Roll No, Email are qualified to become a primary key. But
since StudID is the primary key, Roll No, email becomes the alternative key.

Studio Roll No First Name Last Name Email

1 11 Tom Price [email protected]
(c

2 12 Nick Wright [email protected]

3 13 Dans Natan [email protected]

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 103

5. Foreign Key
Notes

e
Foreign Key is applied to accumulate relationships among two tables. A foreign
key would require every worth in a column or set of columns to coordinate with the

in
number one key of the referential desk. Overseas keys assist to keep records and
referential integrity.

Example:

nl
DeptCode DeptName
001 Science
002 English

O
005 Computer

Teacher ID Fname Lname

ty
8002 David Warner
8017 Sara Joseph
8009 Mike Drunton

si
In this key in dbms example, we have two table, teach and department in a school.
However, there is no way to see which search work in which department.
r
In this table, adding the foreign key in deptcode to the Teacher name, we can
ve
create a relationship between the two tables.

6. Composite Key
ni

Composite Key is a set of two or more attributes that help identify each tuple in
a table uniquely. The attr in the set may not be unique when considered separately.
However, when combined, they will ensure uniqueness.
U

7. Unique Key
Unique Key is a column or set of columns that uniquely identify each record in
a table. All values will have to be unique in this Key. A unique Key is not considered
ity

same as a primary key because it can have only one null value, whereas a primary Key
cannot have any null values.

3.1.3: Relational Model Constraints

m

Relational model became proposed through e. F. Codd to model statistics as family

members or tables. Subsequent to making plans the conceptual model of database
making use of er diagram, we need to change over the conceptual model in the
)A

relational model which can be accomplished utilising any RDBMS dialects like oracle
square, MySQL and so forth so we will perceive what relational model is.

On modeling the plan of the relational database we can placed some restrictions
like what values are allowed to be inserted within the relation, what type of changes and
(c

deletions are allowed inside the relation. Those are the restrictions we pressure at the
relational database. Constraints in relational model

Amity Directorate of Distance & Online Education

 104 Introduction to Data Base Management Systems

At the same time as designing relational model, we define some situations which
Notes

e
have to preserve for facts present in database are called constraints. Those constraints
are checked earlier than performing any operation (insertion, deletion and updation) in
database. If there’s a contravention in any of constrains, operation will fail.

in
Constraints in the databases can be categorized into 3 main categories:

◌◌ Constraints which can be carried out inside the information version is referred

nl
to as implicit constraints.
◌◌ Constraints which might be immediately carried out inside the schemas of
the data model, via specifying them in the ddl (statistics definition language).

O
Those are called as schema-based constraints or explicit constraints.
◌◌ Constraints that cannot be immediately applied inside the schemas of the
statistics version. We name those application based totally or semantic

ty
constraints. So here we will deal with implicit constraints.

Mainly Constraints on the relational database are of 4 types:

◌◌ Domain constraints

si
◌◌ Key constraints
◌◌ Entity Integrity constraints
◌◌ r
Referential integrity constraints
ve
Domain Constraints:
◌◌ Each domain need to comprise atomic values (smallest indivisible units) it way
composite and multi-valued attributes are not allowed.
ni

◌◌ We perform data type take a look at here, this means that whilst we assign an
information type to a column we limit the values that it could include. E.g.. If
we assign the datatype of characteristic age as int, we cant deliver it values
U

different then int datatype.

Example:

EID Name Phone

ity

01 Vikas Dutta 123456789

234456678

Explanation:
m

In the above relation, name is a composite attribute and Phone is a multi-vales

attribute, so it is violating domain constraint.

Key Integrity Constraints :

)A

◌◌ these are referred to as forte constraints because it ensures that each tuple
inside the relation must be precise.
◌◌ a relation could have more than one keys or candidate keys(minimum
superkey), out of which we pick one of the keys as number one key, we don’t
(c

have any restrict on deciding on the primary key out of candidate keys, but it
is counseled to go with the candidate key with much less variety of attributes.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 105

◌◌ Null values are not allowed inside the number one key, therefore no longer
Notes

e
null constraint is likewise a part of key constraint.
Example:

in
EID Name Phone
01 Vikas 6000000009

nl
02 Paul 9000000009
01 Tuhin 9234567892

O
Explanation:
In the above table, EID is the primary key, and first and the last tuple has the same
value in EID ie 01, so it is violating the key constraint.

ty
Entity Integrity Constraints:
◌◌ Entity integrity constraints says that no number one key can take null fee, due
to the fact that using primary key we perceive each tuple uniquely in a relation.

si
Example:

EID Name Phone

01 Vikas r 8000900099
ve
02 Paul 6000000009
Null Sony 9234567890

Explanation:
ni

In the above table, EID is primary key, and the rimary key cant take Null values but
in the third tuple, the primary key is null, so it is a violating entity integrity constraints..
U

Referential Integrity Constraints:

◌◌ The referential integrity constraints is exact between relations or tables and
used to hold the consistency the various tuples in two relations.
ity

◌◌ This constraint is enforced thru overseas key, whilst an characteristic in the

foreign key of relation r1 have the same domain(s) as the number one key of
relation r2, then the foreign key of r1 is said to reference or seek advice from
the primary key of relation r2.
m

◌◌ the values of the overseas key in a tuple of relation r1 can both take the
values of the number one key for a few tuple in relation r2, or can take null
values, however can’t be empty.
)A

EID Name Phone

01 Divine 12
02 Dino 22
(c

04 Vivian 14

Amity Directorate of Distance & Online Education

 106 Introduction to Data Base Management Systems

DNO Place
Notes

e
12 Jaipur
13 Mumbai

in
14 Delhi

Explanation:

nl
In the above, DNO of the first relation is the foreign key, and DNO in the second
relation is the primary key. DNO = 22 in the foreign key of the first table is not allowed
since DNO = 22

O
Is not defined in the primary key of the second relation. Therefore, referential
integrity constraints is voiolated here

ty
3.1. 4: Domain Constraints
Domain Constraints are user-defined columns that assist the user with entering
the value as indicated by the data type. Furthermore, if it experiences an off-base info

si
it gives the message to the user that the column isn’t satisfied as expected. Or then
again at the end of the day, it is an attribute that determines every one of the potential
values that the attribute can hold like number, character, date, time, string, and so forth
r
It characterizes the domain or the set of values for an attribute and guarantees that
ve
the value taken by the attribute should be an atomic value (Can’t be divided) from its
domain.

Domain Constraint = data type (integer / character/date / time / string / etc.)

+ Constraints (NOT NULL / UNIQUE / PRIMARY KEY /FOREIGN KEY / CHECK /
ni

DEFAULT

Type of domain constraints:

U

There are two types of constraints that come under domain constraint and they
are:

1. Domain Constraints – Not Null: Null values are the values that are unassigned or
ity

we are able to likewise say what are unknown or the missing characteristic values
and as a matter of direction, a column can maintain the null values. Currently as
we realise that the now not null quandary restricts a column to now not renowned
the null values which means it simply restricts a field to consistently include a worth
m

which suggests you can’t embed a brand new record or replace a document without
including a well worth into the field.
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 107

2. Domain Constraints – Check: It characterizes a situation that every row must fulfil
Notes

e
which suggests it restricts the cost of a column between stages or we will say that
its miles very just like a situation or channel checking previous to saving information
right into a column. It ensures that after a tuple is embedded within the relation need

in
to satisfy the predicate given inside the take a look at clause.

nl
O
ty
The abolve exmaple creates CHECK constraints on the employee_id column and
specifies that the column employee_id must only include integers that 0.

si
Example 1:
Creating a table “student” with the “RCLL” field having a value greater than 0.

r
ve
ni
U
ity
m
)A

The above example will only accept the roll no. which is greater than 0.
(c

Amity Directorate of Distance & Online Education

 108 Introduction to Data Base Management Systems

3.1.5: Key Constraints

Notes

e
Those are called specialty constraints because it guarantees that every tuple within
the relation ought to be particular.

in
A relation may have multiple keys or candidate keys(minimal super key), out of
which we pick one of the keys as number one key, we don’t have any limit on selecting
the number one key out of candidate keys, however it is cautioned to go along with the

nl
candidate key with much less number of attributes.

Null values aren’t allowed inside the number one key, for this reason no longer null
constraint is also part of key constraint.

O
Example:
NOT NULL: This constraint tells that we cannot save a null price in a column. This

ty
is, assuming a column is determined as not null, we cannot store null on this particular
column any more.

If we determine a subject in a desk to be no longer null. Then, at that factor, the

si
sphere won’t ever acknowledge null fee. That is, you may be not accredited to embed a
brand new row in the table without indicating any price to this subject.

r
For instance, the under question creates a table pupil with the fields identification
and name as no longer null. That is, we will absolutely indicate values for those fields
ve
each time we wish to embed a brand new row.
ni
U

UNIQUE: This constraint whilst indicated with a column, tells that each one of the
values in the column have to be specific. That is, the values in any row of a column
ity

need to now not be repeated.

This constraint serves to uniquely distinguish every row inside the table. For
example for a particular column, each one of the rows ought to have precise values.
We will have more than one precise columns in a desk. For instance, the below query
m

creates a table pupil where the field identification is special as specific. I. E, no college
students could have the identical id.
)A
(c

Primary
Key: A

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 109

Primary key’s a discipline which could uniquely perceive each row in a desk.
Notes

e
Additionally, this constraint is utilized to signify a subject in a desk as

Number one key.

in
Primary key’s an area which uniquely recognizes every row within the table.
Assuming an area in a table as primary key, the field can’t incorporate null values just

nl
as each one of the rows should have precise values for this subject. Alongside these
lines, all in all we will say that this is combo of now not null and specific constraints. A
table will have only one subject as number one key. Beneath question will create a desk
named scholar and specifies the field id as primary key.

O
ty
si
FOREIGN KEY: A foreign secret is a field which can uniquely understand every row
in a some other table. Also, this constraint is applied to determine a subject as foreign
key. r
ve
As we can see clearly that the field C_ID in Orders table is the primary key in
Customers table, i.e. it uniquely identifies each row in the Customers table. Therefore, it
is a foreign key in orders table.
ni

Syntax:
U
ity

Check:
This constraint assists with approving the values of a column to fulfil a selected
m

situation. That is, it assists with making certain that the really worth put away in a
column meets a specific condition.
)A

Using the check constraint we can determine a condition for an area, which need to
be fulfilled on the hour of coming into values for this subject. As an example, the below
query creates a desk pupil and specifies the circumstance for the sector age as (age >=
18). This is, the user will not be allowed to go into any file inside the desk with age < 18.
(c

Amity Directorate of Distance & Online Education

 110 Introduction to Data Base Management Systems

Notes

e
in
nl
Default:
This constraint determines a default value for the column whilst no cost is indicated
via the purchaser. This constraint is utilized to present a default fee to the fields. This

O
is, if at the hour of entering new records in the desk assuming the customer does not
determine any value for these fields, the default cost can be doled out to them. As an
example, the below query will create a desk named pupil and specify the default cost
for the sector age as 18.

ty
r
3.1.6: Referential Key Constraints
si
ve
Tables in a database are frequently no longer unbiased entities, in light of the truth
that usually there are hyperlinks among tables. The facts in a desk that isn’t always
connected is maximum probable of no worth to one-of-a-kind tables. To maintain up
with referential integrity of facts that is related throughout tables, the idea of referential
ni

constraints is offered. A referential constraint is described for a specific column (known

as a foreign key) while a table is defined. A table wherein a referential constraint and
an overseas key are defined is referred to as a referencing table, whilst a table this is
U

referenced from a referencing desk with a foreign secret’s known as a referenced table.
In a referenced desk, a primary key that is referenced through the overseas key need to
be pre-described.

The foreign key in the referencing table references the primary key to decide the
ity

call of the manufacturer...

Figure 5-7 Example of referenced table and referencing table

m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 111

3.1. 7: Update Operations and Dealing with Constraints Violations

Notes

e
The operations of the relational version may be ordered into retrievals and
updates. A relational algebra expression frames any other relation in the wake of

in
making use of diverse algebraic operators to a current association of relations; its
essential use is for querying a database to retrieve facts. The client paperwork a query
that indicates the statistics of hobby, and some other relation is framed by means of
applying relational operators to retrieve this records. That final results relation will

nl
become the response to (or consequence of) the purchaser’s query.

In this phase, we concentrate on database amendment or update operations.

There are three simple operations that can trade the states of relations inside the

O
records-base: insert, delete, and replace (or adjust). They insert new information, delete
vintage statistics, or adjust present information statistics. Insert is used to insert one or
extra new tuples in a relation, delete is used to delete tuples, and replace (or modify)
is used to exchange the values of some attributes in current tuples. On every occasion

ty
those operations are carried out, the integrity constraints precise at the relational
database schema have to no longer be violated. On this segment we discuss the forms
of constraints that may be violated by way of every of these operations and the types
of movements that may be taken if an operation causes a contravention. We use the

si
database proven in figure three. 6 for examples and talk simplest key constraints, entity
integrity constraints, and the referential integrity constraints shown in discern three. 7.
For every kind of operation, we supply a few examples and discuss any constraints that
every operation can also violate. r
ve
1. The Insert Operation
The insert operation gives a list of attribute values for a new tuple t that is to be
inserted right into a relation r. Insert can violate any of the 4 varieties of constraints
ni

dis-stubborn inside the past segment. Domain constraints may be violated if a

characteristic cost is given that doesn’t display up in the comparing domain or isn’t
of the appropriate data type. Key constraints may be violated if a key price inside the
new tuple t as of now exists in some other tuple in the relation r(r). Entity integrity
U

can be violated if any a part of the primary key of the new tuple t is null. Referential
integrity may be violated if the price of any overseas key in t alludes to a tuple that
does not exist inside the referenced relation. Here are some Guides to illustrate this
conversation.
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 112 Introduction to Data Base Management Systems

Notes

e
in
nl
O
If an insertion violates one or extra constraints, the default option is to reject the

ty
insertion. In this example, it would be beneficial if the DBMS may want to offer a cause
to the user as to why the insertion was rejected. Another alternative is to try to accurate
the motive for rejecting the insertion, but this is commonly not used for violations
because of insert; as a substitute, it is used greater frequently in correcting violations

si
for delete and update. Within the first operation, the DBMS could ask the person to
provide a cost for ssn, and could then accept the insertion if a legitimate ssn cost is
furnished. In operation three, the DBMS ought to both ask the consumer to trade the
r
value of dno to a few valid price (or set it to null), or it may ask the user to insert a
ve
department tuple with dnumber = 7 and will be given the original insertion simplest
after such an operation turned into commonplace. Observe that within the latter case
the insertion violation can cascade returned to the employee relation if the consumer
attempts to insert a tuple for branch 7 with a fee for mgr_ssn that doesn’t exist in the
ni

worker relation.

2. The Delete Operation

U

The delete operation can violate only referential integrity. This occurs if the tuple
being deleted is referenced by foreign keys from different tuples in the database. To
specify deletion, a circumstance on the attributes of the relation selects the tuple (or
tuples) to be deleted.
ity
m
)A
(c

Right here are some examples. A few options are available if a deletion operation
reasons an infringement. The number one choice, known as restrict, is to dismiss the

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 113

deletion. The following desire, known as cascade, is to undertaking to cascade (or

Notes

e
unfold) the deletion via deleting tuples that reference the tuple that is being deleted.
As an instance, in operation 2, the DBMS should robotically delete the offending tuples
from works_on with essn = ‘999887777’. A third alternative, called set null or set default,

in
is to adjust the referencing characteristic values that cause the violation; every such fee
is both set to null or changed to reference another default legitimate tuple. Be aware
that if a referencing characteristic that causes a viola-tion is part of the primary key, it

nl
cannot be set to null; otherwise, it would violate entity integrity..

Combinations of these three options are also possible. For example, to avoid
having operation 3 cause a violation, the DBMS amy automatically delete all tuples

O
from Works_ON and EPENDENT with Essn = ‘333445555’. tuples im EMPLOYEE with
Super_ssn = ‘33445555’ and the tuple in DEPARTMENT with Mgr_ssn = ‘33445555’
can have their super_ssn and Mgr_ssn values changed to other valid values or to

ty
NULL. Althorugh it may make sense to delete automatically the WORKS_ON and
EPENDENT tuples that refer to an EMPLOYEE tuple, it may not make sense to delete
other EMPLOYEE tuples or a DEPARTMENT tuple.

si
In general, whilst a referential integrity constraint is exact in the ddl, the DBMS
will allow the database dressmaker to specify which of the options applies in case of a
violation of the constraint.

3. The Update Operation

r
ve
The update (or modify) operation is used to trade the values of 1 or more attributes
in a tuple (or tuples) of some relation r. Its miles important to specify a condition on the
attributes of the relation to select the tuple (or tuples) to be changed. Here are a few
examples. Updating an attribute this is neither a part of a number one key nor of an
ni

overseas key commonly reasons no issues; the DBMS want handiest check to affirm
that the brand new fee is of the right records type and domain.
U
ity
m
)A

Editing a primary key cost is just like deleting one tuple and putting another in
its area due to the fact we use the primary key to identify tuples. Subsequently, the
(c

issues mentioned in advance in each sections three. 3. 1 (insert) and three. 3. 2

(delete) come into play. If a foreign key attribute is modified, the DBMS have to make

Amity Directorate of Distance & Online Education

 114 Introduction to Data Base Management Systems

sure that the brand new cost refers to a present tuple within the referenced relation (or
Notes

e
is ready to null). Comparable options exist to cope with referential integrity violations
caused by update as the ones options mentioned for the delete operation. In reality,
when a referential integrity constraint is exact in the ddl, the DBMS will allow the person

in
to choose separate alternatives to cope with a contravention caused by delete and a
violation resulting from replace

nl
4. The Transaction Concept
A database software program strolling in opposition to a relational database
generally executes at the least one transactions. A transaction is an executing

O
application that includes some database operations, like perusing from the database,
or applying inclusions, cancellations, or updates to the database. Towards the end of
the transaction, it must depart the database in a valid or constant specific that fulfils
each one of the constraints decided on the database schema. A single transaction may

ty
consist of pretty some retrieval operations and quite a few update operations. These
retrievals and updates will collectively body an atomic unit of neutralize the database.
As an instance, a transaction to apply a bank withdrawal will typically examine the

si
consumer account document, check if there’s an enough balance, and then update the
file with the aid of the withdrawal quantity.

A massive wide variety of business applications running towards relational

r
databases in online transaction processing (OLTP) structures are executing
ve
transactions at costs that reach several hundred per second.

Solutions which can be possible to accurate the violation to the referential integrity
due to deletion are listed under:
ni

◌◌ Restrict –
Right here we reject the deletion.

◌◌ Cascade –
U

Here if a record in the parent table (referencing relation) is deleted, then the
corresponding facts in the infant desk (referenced relation) will automatically be deleted.

◌◌ Set null or set default –

ity

Here we alter the referencing characteristic values that cause violation and we
either set null or alternate to another valid cost

Check Your Understanding:

m

A. Fill in the blanks:

1. Relational calculus is a___________.
)A

2. Cartesian product in relational algebra is__________.

3. A domain is a ___________ set of values permitted for an attribute in a table.
4. __________is sometimes used to refer to a table in a relational database but is more
commonly used to describe the relationships that can be created between those
(c

tables in a relational database.

5. A domain is the original sets of _____values used to model data.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 115

B. State True or False:

Notes

e
1. The term attribute refers to a column of a table (table)
2. A domain is atomic if elements of the domain are considered to be divisible units.

in
(False)
3. Some relations may not have a key. (False)

nl
4. A key of a relation may have more than one attribute. (True).
5. The elements of the domain of an attribute may be ordered pair of values. (False)

O
Activity
1. Formulate a presentation on Referential Key Constraints.

Summary

ty
●● A domain is a unique set of values allowed for a trait in a table. For instance, a
domain of month-of-year can acknowledge January, February… .December as
possible values, a domain of integers can acknowledge whole numbers that are

si
negative, positive and zero.
●● Connection is at times used to allude to a table in a relational database however
is all the more usually used to portray the relationships that can be made between
those tables in a relational database. r
ve
Further Readings
1. Raghu Ramakrishnan and Johannes Gehrke. (2003). Database Management
Systems. Third Edition. McGraw Hill
ni

2. Ramez Elmasri and Shamkant B. Navathe. (2016). Fundamentals of Database

Systems. Seventh Edition. Pearson Education, Inc.
U

Check your Understanding (answers):

A. Fill in the blanks:
1. Non- Procedural language
ity

2. a Binary operator.
3. Unique
4. relation
m

5. atomic
B. State True or False:
)A

1. True
2. False
3. False
4. True
(c

5. False

Amity Directorate of Distance & Online Education

 116 Introduction to Data Base Management Systems

Unit - 3.2: Relational Model Operations and SQL

Notes

e
Language

in
Objectives:
At the end of this unit, you will be able to:

nl
●● Understand Relational operators
●● Know about Set operations in relation and join operations

O
●● Understand SQL Language and its features
●● Understand Data definition language and Data manipulation language and their
features

ty
●● Know about Data Definitions Commands-Create, Alter, Truncate, Drop
●● Know about Data Manipulation Commands-Select, update, delete
●● Learn about SQL String Functions and SQL Joins

si
●● Know about views

3.2.1 Relational Operators

r
ve
Selection Operation
●● Selection Operator (σ) is a unary operator in relational algebra that performs a
selection operation.
●● It selects those rows or tuples from the relation that satisfies the selection
ni

condition.

E.g. :
U

LOAN Relation

Branch_name Loan_no Amount

ity

Downtown L-17 000

Redwood L-23 2000
Ferry L-15 1500
Downtown L-14 1500
m

Mian L-13 500

Round L-11 900
Perry L-16 1300
)A

Input:
◌◌ σ BRANCH_NAME=”Ferry” (LOAN)

Output:
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 117

Branch_name Loan_no Amount

Notes

e
Ferry L-15 1500
Perry L-16 1300

in
Project Operation:
◌◌ This operation shows the list of those attributes those have to appear in the

nl
result.
◌◌ It is denoted by ∏.

O
Notation: ∏ A1, A2, An (r)

Here,

A1, A2, A3 are used as an attribute name of relation r.

ty
E.g. :

Name Street City

si
J M Harrison
S N Rye
H M Harrison
C N
r Rye
ve
J A Brooklyn
B S Brooklyn
ni

Input:
◌◌ ∏ NAME, CITY (CUSTOMER)
U

Output:

Name City
J Harrison
ity

S Rye
H Harrison
C Rye
J Brooklyn
m

B Brooklyn
)A

3.2.2 Set Operations on Relation

Union Operation
●● Let there be two tuples R and S. The union operation contains all the tuples that
are either in R or S or both in R & S.
(c

●● It removes all the duplicate tuples.

●● And is denoted by U.
Amity Directorate of Distance & Online Education
 118 Introduction to Data Base Management Systems

Notation: R U S
Notes

e
Union operation must hold the following condition:

o Both R and S must have the attribute of the same number.

in
o Duplicate tuples are removed automatically.

E.g.:

nl
Depositor Relation

O
C_NAME A_NO
J A-101
S A-121
M A-321

ty
T A-176
J A-273

si
J A-427
L A-284

Borrow Relation :
r
ve
C_NAME L_NO
J L-17
S L-23
ni

H L-15
J L-14
C L-93
U

S L-11
W L-17
ity

Input:

◌◌ ∏ C_NAME (BORROW) U ∏ C_NAME (DEPOSITOR)

Output:
m

C_NAME
J
)A

S
H
T
J
(c

L
J

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 119

C
Notes

e
W
M

in
Set Intersection:
◌◌ Let there be two tuples R and S. The set intersection operation contains all

nl
tuples that are in both R & S.
◌◌ Denoted by intersection ∩.

O
Notation: R ∩ S
E.g.:

Depositor Relation

ty
C_NAME A_NO
J A-101

si
S A-121
M A-321
T A-176
J A-273r
ve
J A-427
L A-284

Borrow Relation:
ni

C_NAME L
J L-17
U

S L-23
H L-15
J L-14
ity

C L-93
S L-11
W L-17
m

Input:
◌◌ ∏ C_NAME (BORROW) ∩ ∏ C_NAME (DEPOSITOR)
)A

Output:
C_NAME

S
(c

Amity Directorate of Distance & Online Education

 120 Introduction to Data Base Management Systems

Set Difference:
Notes

e
◌◌ Let there be R and S. The set Difference operation contains all tuples that are
in R but not in S.

in
◌◌ Denoted by intersection minus (-).

Notation: R - S

nl
E.g.:

Depositor Relation

O
C_NAME A_NO
J A-101
S A-121

ty
M A-321
T A-176
J A-273

si
J A-427
L A-284

Borrow Relation:
r
ve
C_NAME L_NO
J L-17
S L-23
ni

H L-15
J L-14
U

C L-93
S L-11
W L-17
ity

Input:
◌◌ ∏ C_NAME (BORROW) - ∏ C_NAME (DEPOSITOR)
Output:
m

C_NAME

J
)A

C
(c

Cartesian product
◌◌ This used to combine each row in one table with each row in the other table. It
is also known as a cross product.
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 121

◌◌ Denoted by X.
Notes

e
Notation: E X D

E.g.:

in
EMP

E_ID E_NAME E_DEPT

nl
1 S A
2 H C

O
3 J B

DEPT

D_NO D_NAME

ty
A M
B S

si
C L
Input:

◌◌ EMP X DEPT
r
ve
Output:

E_ID E_NAME E_DEPT D_NO D_

NAME
1 S A A M
ni

1 S A B S
1 S A C L
U

2 H C A M
2 H C B S
2 H C C L
ity

3 J B A M
3 J B B S
3 J B C L
m

Rename Operation:
This operation is used to rename the output relation and is denoted by rho (ρ).

E.g.: Lets use rename operator to rename STUDENT relation to STUDENT1.

)A

◌◌ ρ(STUDENT1, STUDENT)

3.2.3 Join Operations

(c

It combines related tuples from different relations, if and only if a given join
condition is satisfied and it is denoted by U.

E.g.:
Amity Directorate of Distance & Online Education
 122 Introduction to Data Base Management Systems

EMP
Notes

e
E_CODE E_NAME
101 S

in
102 J
103 H

nl
SAL

E_CODE S

O
101 50000
102 30000
103 25000

ty
Operation: (EMP x SAL)

si
Result:

E_CODE E_NAME S
101
r S 50000
ve
102 J 30000
103 H 25000
ni

Types of Join operations:

1. Natural Join:
U

◌◌ The set of tuples of all combinations in R and S that are equal on their
common attribute names.
◌◌ Denoted by .
ity

E.g.: EMP

E_CODE E_NAME
m

101 S
102 J
103 H
)A

SAL

E_CODE S
101 50000
(c

102 30000
103 25000

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 123

Operation: (EMP x SAL)

Notes

e
Input:

◌◌ ∏E_NAME, SAL (EMP SAL)

in
Output :

E_NAME SAL

nl
S 50000
J 30000

O
H 25000

2. Outer Join:

ty
It is used to deal with missing information.

E.g.:

si
EMP

E_NAME STREET CITY

R
S
C
Park street
r Mumbai
Kolkata
ve
R M.G. Street Delhi
H Nehru nagar Hyderabad
ni

FACT_WORKERS

E_NAME BRANCH SAL

U

R Infosys 10000
S Wipro 20000
K HCL 30000
H TCS 50000
ity

Input:

◌◌ (EMP ⋈ FACT_WORKERS)
m

Output:

E_NAME STREET CITY BRANCH SAL

)A

R C Mumbai Infosys 10000

S P Kolkata Wipro 20000
H N Hyderabad TCS 50000

Outer join is of three types:

(c

◌◌ Left outer join

◌◌ Right outer join

Amity Directorate of Distance & Online Education

 124 Introduction to Data Base Management Systems

◌◌ Full outer join

Notes

e
Left outer join:

in
◌◌ It contains the set of tuples of all combinations in R and S that are equal on
their common attribute names.
◌◌ In this, tuples in R have no matching tuples in S.

nl
◌◌ And is denoted by .
E.g.:

EMP

O
E_NAME STREET CITY

R C Mumbai

ty
S P Kolkata

R M Delhi

si
H N Hyderabad

FACT_WORKERS

E_NAME BRANCH
r SAL
ve
R Infosys 10000

S Wipro 20000

K HCL 30000
ni

H TCS 50000

Input:
U

◌◌ EMP ⟕ FACT_WORKERS

Output

E_NAME STREET CITY BRANCH SAL

ity

R C Mumbai Infosys 10000

S P Kolkata Wipro 20000
H N Hyderabad TCS 50000
m

Ravi M Delhi NULL NULL

Right outer join:

)A

o It contains the set of tuples of all combinations in R and S that are equal on
their common attribute names.
o In this , tuples in S have no matching tuples in R.
(c

o And is denoted by .

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 125

E.g.:
Notes

e
EMP

in
E_NAME STREET CITY
R C Mumbai
S P Kolkata

nl
R M Delhi
H N Hyderabad

O
FACT_WORKERS

E_NAME BRANCH SAL

ty
R Infosys 10000
S Wipro 20000
K HCL 30000

si
H TCS 50000

Input:

◌◌ EMP FACT_WORKERS
r
ve
Output:

E_NAME BRANCH SAL STREET CITY

R Infosys 10000 C Mumbai
ni

S Wipro 20000 P Kolkata

H TCS 50000 N Hyderabad
U

K HCL 30000 NULL NULL

Full outer join:

ity

o It is like a left or right join except that it contains all rows from both tables.
o In this , tuples in R that have no matching tuples in S and tuples in S that have
no matching tuples in R in their common attribute name.
o And is denoted by .
m

E.g. :

EMP
)A

E_NAME STREET CITY

R C Mumbai
S P Kolkata
Ravi M Delhi
(c

H N Hyderabad

Amity Directorate of Distance & Online Education

 126 Introduction to Data Base Management Systems

FACT_WORKERS
Notes

e
E_NAME BRANCH SAL
R Infosys 10000

in
S Wipro 20000
K HCL 30000

nl
H TCS 50000

Input:

O
◌◌ EMP FACT_WORKERS
Output:

E_NAME STREET CITY BRANCH SAL

ty
R C Mumbai Infosys 10000
S P Kolkata Wipro 20000
H N Hyderabad TCS 50000

si
Ravi M Delhi NULL NULL
K NULL NULL HCL 30000

r
ve
Equi join:
◌◌ Another name for Equi Join is an inner join.
◌◌ It is one of the most common join.
◌◌ And is based on matched data as per the equality condition.
ni

◌◌ It uses the comparison operator(=).

E.g.:
U

Customer Relation
CLASS_ID NAME
ity

1 J

2 H

3 J
m

Product
PRODUCT_ID CITY
)A

1 Delhi

2 Mumbai

3 Noida
(c

Input:

◌◌ CUSTOMER PRODUCT

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 127

Output :
Notes

e
CLASS_ID NAME PRODUCT_ CITY
ID

in
1 J 1 Delhi
2 H 2 Mumbai

nl
3 H 3 Noida

3.2.4 SQL Language and Its Characteristics

O
Structured Query Language (SQL) is the standard language used for writing
queries in a databases. It was approved by ISO (International Standard Organization)
and ANSI (American National Standards Institute).

ty
SQL contains of some important features and they are:

◌◌ Data Definition language (DDL):

It contains of commands which defines the data. The commands are:

si
◌◌ Create : It is used to create a table.
Syntax:
r
Create table table name(attribute1 datatype......attributen datatype);
ve
◌◌ Drop: It is used to delete the table including all the attributes.
Syntax:

drop table tablename;

ni

◌◌ Alter: alter is a reserve word which modifies the structure of the table.
Syntax:
U

alter table tablename add(new column1 datatype......new columnxdatatype);

◌◌ Rename: A table name can be changed using the ‘rename’ operator.

Syntax:
ity

Rename (old table name) to (new table name);

Data Manipulation Language (DML):

It contains Cmds used to manipulate the data.

m

The Cmds used in this language are:

◌◌ Insert: It is generally used after the create Cmd to insert a set of values into
)A

the table.
Syntax:

insert into (t_name) values(attribute_1 datatype);

(c

insert into (t_name) values (attributes datatype);

◌◌ Delete: This Cmd isused to delete particular tuples or rows from the table.

Amity Directorate of Distance & Online Education

 128 Introduction to Data Base Management Systems

Syntax:
Notes

e
delete from (t_name)where condition;

◌◌ update: It updates the tupples in a table.

in
Syntax:

Updatet_name set tupplename=’attributename’;

nl
Triggers:
These are the actions performed when certain conditions are met in the data.

O
◌◌ It contains of three parts.
1. (i) event – The change in the database that activates the trigger is event.
2. (ii) condition – A query or test that is run when the trigger is activated.

ty
3. (iii) action – A procedure that is executed when trigger is activated and the
condition met is true.

si
Client server execution and remote database access:
Client server technology maintains a many to one relationship of clients.Where

r
clients are many and server is one . There are Cmds in SQL that control how a client
application can access the database over a network.
ve
Security and authentication:
Structured Query Language provides a mechanism to control the database that is
ni

it makes sure that only the particular details of the database is shown to the user and
the original database is secured by DBMS.

Embedded SQL:
U

Structured Query Language provides the feature of embedding host languages

such as C, COBOL, Java for query from their language at runtime.
ity

Transaction Control Language:

Transactions are an important element of DBMS and to control the transactions,
Transaction Control Language is used. It has Cmds like commit, rollback and savepoint.

◌◌ Commit: Saves the data in the database at any point whenever database is
m

consistent.
Syntax:
)A

commit;

◌◌ rollback: The Cmdrollbacks/undo to the previous point of the transaction.

Syntax:

rollback;
(c

◌◌ savepoint: This Cmdgoes back to the previous transaction without going back
to the entire transaction.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 129

Syntax:
Notes

e
savepoint;

Advanced SQL:

in
The current features in SQL include OOP ones like recursive queries, decision
supporting queries and also query supporting areas like data mining, spatial data and

nl
XML(XtensibleMarkup Language).

3.2.5 Data Definition Language and its Features

O
A Data Definition Language (DDL) is a list of Cmds which is used to create and
modify the structure/Schema of database.

It is also called Data Description Language as it portrays the fields and records in a

ty
database table.

Nowadays database industry joins DDL into any proper language describing data
and it is viewed as a subset of the Structured Query Language. SQL regularly utilizes

si
the basic action words with ordinary English like sentences to implement database
modifications.

DDL doesn’t appear as an alternate language in a SQL database, then also it

characterizes changes in the database composition. r
ve
It is utilized to set up and modify the structure of objects in a database or the
Schema of the database by managing depictions of the database construction. With
respect to Data Manipulation Language (DML) Cmds that are utilized for modification of
data , the DDL Cmds are used for altering the database structure, for Eg, creating new
ni

tables or objects alongside the datatype of their attributes (data type, table name, etc.).

DDLCmds are commonly used in SQL query.

U

Those Cmds are CREATE, ALTER, DROP, and TRUNCATE.

Create

Cmdwhich is used to build a new table and has a predefined syntax. The CREATE
ity

statement syntax is:

CREATE TABLE [table name] ([column definitions]) [datatype];

For Eg:
m

CREATE TABLE EMP (EMP Id INTEGER PRIMARY KEY, First name CHAR (50)
NULL, Last name CHAR (75) NOT NULL);
)A

Semi-colon is must at the end of the statement . ‘;’ is used to process every Cmd
before it. In this Eg, the string CHAR is used to specify the data type of th attribute.
Other data types can be

DATE, NUMBER, or INTEGER.

(c

Alter

The Alter Cmd updates an existing database table. This Cmd can add up new

Amity Directorate of Distance & Online Education

 130 Introduction to Data Base Management Systems

column, drop the already existing columns and even change the data type of columns
Notes

e
present in a database table.

Syntax is:

in
ALTER object type object name paReters;

For Eg:

nl
ALTER TABLE EMP ADD PRIMARY KEY (EMP_pk);

In this Eg, we added a unique primary key to the table to add a constraint and
enforce a unique value. The constraint “EMP_pk” is a primary key and is on the EMP

O
table.

Drop

ty
The drop Cmd is used to remove objects such as a table, index or view. A DROP
statement cannot be rolled back, If once an object is destroyed, there’s no way to
recover it.

si
Drop statement syntax is:

DROP object type object name;

For Eg:
r
ve
DROP TABLE EMP;

In this Eg, we’re deleting the EMP table.

Truncate
ni

The TRUNCATE statement is used to remove all records from a table. This Cmds
preserves its full structure to be reused in the future.

Truncate statement syntax is:

U

TRUNCATE TABLE table_name;

For Eg:
ity

TRUNCATE TABLE EMP;

In this Eg, we’re marking all the extents of the EMP table for deallocation, so
they’re considered empty for reuse.
m

Other statements

Other Ofently used Cmds include RENAME and COMMENT. The first one is used
with the ALTER the name of an object (table, column, etc.). COMMENT is used to add
)A

single line, multi-line and in-line comments.

3.2.6 Data Manipulation Language and its Features

Data Manipulation Language (DML) is a list of Cmds including the Cmds which
(c

allow the clients to manipulate data in a database. This manipulation includes insertion
of data into database tables, retrieving existing data, deleting data from existing tables
and modification of existing data.
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 131

The functional capability of DML is organized in manipulating Cmds like SELECT,

Notes

e
UPDATE, INSERT INTO and DELETE FROM, as mentioned below:

◌◌ SELECT: cmd used to retrieve rows from a table.

in
The syntax is SELECT [column name(s)] from [table name] where [conditions].

◌◌ UPDATE: cmd modifies data of one or more records.

nl
syntax is UPDATE [table name] SET [column name = value] where [condition].

◌◌ INSERT: cmd adds one or more records to a database table.

syntax is INSERT INTO [table name] [column(s)] VALUES [value(s)].

O
◌◌ DELETE: cmd removes one or more records from a table according to
specified conditions.
Cmd syntax is DELETE FROM [table name] where [condition].

ty
3.2.7 Data Definitions Commands-Create, Alter, Truncate, Drop
The Create Cmd

si
The create table Cmd defines each entity of the database uniquely. Each column
has minimum of three attributes.

◌◌ Name r
ve
◌◌ Data type
◌◌ Size(column width).
Each table column definition is a single clause in the create table syntax. Each
ni

table column definition is separated from the other by a ‘,’ .The SQL statement is
terminated with a semicolon.

The Structure of Create Table Cmd

U

Table name is Student

C_name Data_type Size

Regno varchar2 10
ity

Name Char 30
DateOfBirth Date
Add varchar2 50
m

Eg:

CREATE TABLE Student (Regnovarchar2(10), Name char(30), DateOfBirth date,

)A

Add varchar2(50));

The DROP Cmd

Syntax:
(c

DROP TABLE <table_name>

Eg:

Amity Directorate of Distance & Online Education

 132 Introduction to Data Base Management Systems

DROP TABLE Student;

Notes

e
It will remove the table and all data which will be recorded in it.

The TRUNCATE Cmd

in
Syntax:

TRUNCATE TABLE <Table_name>

nl
Eg:

TRUNCATE TABLE Student;

O
The RENAME Cmd

ty
Syntax:

RENAME <OldT_name> TO <NewT_name>

si
Eg:

RENAME <Student> TO <Stu>

r
The old table name Student is now changed to Stu.
ve
The ALTER Table Cmd
ni

ALTER TABLE Cmds help us to modify our exiting table.

Adding New Columns

U

Syntax:

ALTER TABLE <table_name>

ity

ADD (<NewColumnName><Data_Type>(<size>),......n)

Eg:

ALTER TABLE Student ADD (Age number(2), Marks number(3));

m

The Student table is already created and then we added two more columns Age
and Marks, by the use of above Cmd.
)A

Dropping a Column from the Table

Syntax:

ALTER TABLE <table_name> DROP COLUMN <column_name>

Eg:
(c

ALTER TABLE Student DROP COLUMN Age;

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 133

This Cmd will drop particular column

Notes

e
Modifying Existing Table

Syntax:

in
ALTER TABLE <table_name> MODIFY (<column_
name><NewDataType>(<NewSize>))

nl
Eg:

ALTER TABLE Student MODIFY (Name Varchar2(40));

O
The Name column already exist in Student table, its datatype was char and size
30, now it is modified by Varchar(2) and size 40.

Restriction on the ALTER TABLE

ty
Using the ALTER TABLE clause the following tasks cannot be performed.

◌◌ Change the name of the table

si
◌◌ Change the name of the column
◌◌ Decrease the size of a column if table data exists

r
3.2.8 Data Manipulation Commands-Select, Update, Delete.
ve
Viewing Data in the Table (Select Cmd)

Once you have inserted the data into the table, then to view the inserted data .This
is where the SELECT cmd is used.
ni

To display All Rows and All Columns

Syntax: SELECT * FROM Table_name;

eg: Select * from Student; It shows all the table records.

U

SELECT First_name, DateOfBirth FROM STUDENT WHERE Regno = ‘S101’;

Cover it by single inverted comma if its datatype is varchar or char.
ity

Here, the output will have one row. Because the given condition for only one row
and particular records. If condition which is given in WHERE Clause is true then records
will be fetched otherwise it will show no records selected.

Eliminating Duplicates:
m

A table can hold duplicate rows. In such a case, we can eliminate duplicates.

Syntax: SELECT DISTINCT col, col, .., FROM table_name;

)A

eg : SELECT DISTINCT * FROM Student;

or : SELECT DISTINCT first_name, city, pincode FROM Student;

SQL scans through entire rows, and eliminates rows that have exactly the same
contents in each column.
(c

Sorting DATA:

Amity Directorate of Distance & Online Education

 134 Introduction to Data Base Management Systems

The Rows retrieved from the table can be displayed in sorted sorted order
Notes

e
either Ascending or Descending order depending on the condition specified in select
statement, the Keyword has used ORDER BY.

in
SELECT * FROM Student

ORDER BY First_Name;

nl
The output will show records as alphabetical order from A to Z ascending order. If
you want Descending order means Z to A then used DESC Keyword at last.

eg : SELECT first_name, city,pincode FROM Student

O
ORDER BY First_name DESC;

DELETE Operation:

ty
The DELETE Cmd will remove all the tuples from the table or a set of tuples from
the table.

eg: DELETE FROM student; It will DELETE all the rows from student table.

si
eg: DELETE FROM student WHERE reg_no=’A101’; If condition will be satisfied
then it will delete a row from the table Register number A101 will be deleted from the
table
r
If a column is to be deleted then use ALTER TABLE Cmd.
ve
eg : ALTER TABLE student DROP COLUMN dob; The DOB column will be
deleted from the student table.
ni

UPDATE Operation:
The UPDATE Cmd is used to update or modify or change instances in a table
.The UPDATE Cmd can Update all the rows from the table or a set of rows from the
U

table.

eg : UPDATE Student SET course=’MCA’;

Course is a column name, suppose ant time you want to update something like
ity

that in the student table course should be MCA for all students then you can use this
type of query. It will update all the rows in the table all rows will have MCA course.

Now, if you want update particular row then see below.

m

UPDATE Student SET course=’MCA’ where reg_no=’A101’; it will update only one
row that will have Register no. A101.

you can use different-different types of condition in WHERE clause, eg salary

)A

updation, if salary has increased someone’s then simply multiply, addition you can do in
salary column.

Suppose someone’s has increased salary by 10% then what should to do. See
example:
(c

UPDATE employee SET salary= salary+salary*0.1 WHERE employee_id=’E101’;

Salary will be automatically increased by 10% of salary.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 135

3.2.9 SQL String Functions

Notes

e
SQL string functions are used primarily for string manipulation.

in
S.No Function & Description of String Functions
ASCII()
Returns a numeric value of left-most character

nl
BIN()
Returns a string representation of the argumenl
BIT_LENGTH()

O
Returns the length of argument in bits
CHAR_LENGTH()
Returns a number of characters in argument
CHAR()

ty
Returns the character for each integer passed
CHARACTER_LENGTH()
A synonym for CHAR_LENGTH()

si
CONCAT_WS()
Returns concatenate with separator
CONCAT()
Returns concatenated string
r
ve
CONV()
Converts numbers between different number bases
ELT()
ni

Returns string at index number

EXPORT_SET()
Returns a string such that for every bit set in the value bits, you get
U

an on string and for every unset bit, you get an off string
FIELD()
Returns the index (position) of the first argument in the subsequent
arguments
ity

FIND_IN_SET()
Returns the index position of the first argument within the second
argument
FORMAT()
m

Returns a number formatted to specified number of decimal places

HEX()
Returns a string representation of a hex value
)A

INSERT()
Inserts a substring at the specified position up to the specified
number of characters
INSTR()
(c

Returns the index of the first occurrence of substring

Amity Directorate of Distance & Online Education

 136 Introduction to Data Base Management Systems

LCASE()
Notes

e
Synonym for LOWER()
LEFT()

in
Returns the leftmost number of characters as specified
LENGTH()
Returns the length of a string in bytes

nl
LOAD_FILE()
Loads the named file

LOCATE()

O
Returns the position of the first occurrence of substring
LOWER()
Returns the argument in lowercase

ty
LPAD()
Returns the string argument, left-padded with the specified string
LTRIM()

si
Removes leading spaces
MAKE_SET()
Returns a set of comma-separated strings that have the
r
corresponding bit in bits set
ve
MID()
Returns a substring starting from the specified position
OCT()
Returns a string representation of the octal argument
ni

OCTET_LENGTH()
A synonym for LENGTH()
ORD()
U

If the leftmost character of the argument is a multi-byte character,

returns the code for that character
POSITION()
ity

A synonym for LOCATE()

QUOTE()
Escapes the argument for use in an SQL statement
REGEXP
m

Pattern matching using regular expressions

REPEAT()
Repeats a string the specified number of times
)A

REPLACE()
Replaces occurrences of a specified string
REVERSE()
Reverses the characters in a string
(c

RIGHT()
Returns the specified rightmost number of characters

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 137

RPAD()
Notes

e
Appends string the specified number of times
RTRIM()

in
Removes trailing spaces
SOUNDEX()
Returns a soundex string

nl
SOUNDS LIKE
Compares sounds
SPACE()

O
Returns a string of the specified number of spaces
STRCMP()
Compares two strings

ty
SUBSTRING_INDEX()
Returns a substring from a string before the specified number of
occurrences of the delimiter

si
SUBSTRING(), SUBSTR()
Returns the substring as specified
TRIM()
r
Removes leading and trailing spaces
ve
UCASE()
Synonym for UPPER()
UNHEX()
Converts each pair of hexadecimal digits to a character
ni

UPPER()
Converts to uppercase
U

ASCII(str)

Returns the numeric value of the leftmost character of the string str. Returns 0 if str
is the empty string. Returns NULL if str is Null. ASCII() works forcharacters with numeric
ity

values from 0 to 255.

m
)A
(c

Amity Directorate of Distance & Online Education

 138 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
r si
ve
Returns a string representation of the binary value of N, where N is a longlong
(BIGINT) number. This is equivalent to CONV (V,10,2). Returns NULL if N is NULL.

SQL> SELECT BIG(12):

ni

._________________.

IBIN(12)
U

._________________.

|1100 |

1 row in set (0.00 sec)

ity

BIT_LENGTH(str)

Returns the length of the string str in bits.

SQL.SELECT BIT_LENGTH(Text);
m
)A

1 row in set (0.00 sect)

CHAR(N... [USING charset_name)]

(c

CHAR() interprets each argument N as an interger and returns a string consisting

of the characters given by the code values of those integers. NULL values are skipped.
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 139

Notes

e
in
1 row in set (0.00se)

nl
CHAR_LENGTH(str)

Returns the length of the string str measured in characters. A multi-byte character

O
counts ae a single character. This means that for a string containing five tow-byts
character LENGTH() returns 10, whereas CHAR_LENGTH() retuns 5.

ty
r si
ve
CHARACTER_ENGTH() is a synonym for CHAR_LENGTH().

CONCAT (str1, str2,..)

Returns the string that results from concalennating the arguments. May have
ni

one or more arguments. If all arguments are non-binary strings, the results a non-
binary string. If the arguments include any binary strings, the result is a binary string. A
numeric argument is converted to its equivalent binary string from; if you want to avoid
U

that, you can use an expilicit type cast, as in this exmple.

ity
m
)A

CONCT_WS() stands for concatenate with separator and is a special form

of CONCAT() The first argument is the separator for the rest of the arguments. The
separator is added between the strings to be concatenated. The separator can be a
string, as can the rest of the arguments. If the separator is NULL, the result is NULL.
(c

Amity Directorate of Distance & Online Education

 140 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Converts numbers between different number bases. Returns a string
representation of the number N, converted frombase from_base to to_base. Returns
NULL if any argument is NULL. The argument N is interpreted as an integer, but may

ty
be specified as an integer or a string. The minimum base is 2 and the maximum base is
36. If to_base is a negative number, N is regarded as a signed number. Otherwise, N is
treated as unsigned. Conv() works with 64-bit precision.

r si
ve
ni

returns str1 if N = 1, str2 if N = 2, and so on. Returns NULL if N is less than 1 or

U

greater than the number of arguments. ELT() is the complement of FIELD().

ity
m
)A

returns a string suck that for every bit set in the value bits, you et an on string and
for every bit not set in the value, you get an off string. Bits in bits are exmined from right
to left (from low-order to high-order bits). Strings are added to the result from left to
right, separated by the seprarator string (the default being the comma character...). The
(c

number of bits examined is given by number_of_bits (defaults to 64).

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 141

Notes

e
in
nl
O
ty
r si
ve
ni
U

Formats the number X to a format like ‘#,###,###,##’, rounded to D decimal places,

and retuns the result as a string. If D is 0, the result has no decimal point or fractional
ity

part.
m
)A
(c

If N_or_S is a number, returns a string representation of the hexadecimal value of

N, where N is a longlong (BIGINT) number. This is equivalent to CONV(N, 10, 16).

Amity Directorate of Distance & Online Education

 142 Introduction to Data Base Management Systems

If N_or_S is a string, returns a hexadecimal string representation of N-or_S where

Notes

e
each character in N_or_S is converted to two hexadecimal digits.

SQL>SELECT HEX(255);

in
nl
O
ty
r si
Returns the string str, with the substring beginning at position pos and len
characters long replaced by the string newstr. Returns the original string if pos is not
ve
within the length of the string. Replaces the rest of the sring from position pos if len is
not withing the length of the rest of the string. Returns NULL if any arguments is NULL.
ni
U
ity

Returns the position of the first occurence of substring substr in string str. This
is the same as the two-argument from of LOCATE(), except that the order of the
arguments is reversed.
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 143

Notes

e
in
nl
O
returns the length of the string str, measured in bytes. A multi-byte character counts
as multiple bytes. This means that for a string containing five two-byte characters,
LENGTH() returns 40, whereas CHAR-LENGTH() returns 5.

ty
r si
ve
Reads the file and returns the file contents as a string. To use this function, the file
must be located on the server host. you must specify the full pathname to the file, and
ni

you must have the FILE privilege. The file must be readable by all and its size less than
max_allowed_ packet bytes.

If the file does not exist or cannot be read because one of the preceding conditions
U

is not satisfied, the function retuns NULL.

As of SQL 5.0.19, the character_set_filesystem system variable controls

interpretation of file names that are given as literal strings.
ity
m

The first syntax returns the position of the first occurrence of substring substr in
strling str. The second syntax returns the position of the first occurrence of substring
)A

substr in string str, starting at position pos. Returns () if substr is not in str.
(c

Amity Directorate of Distance & Online Education

 144 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Returns the string str with all characters changed to lowercase according to the
current character set mapping.

ty
r si
ve
Returns the string str, left-padded with the string padstr to a length of len
characters. if str is longer than len, the return value is shortened to len character.
ni
U
ity
m
)A

Retuns a set value (a string containing substrings separated by. characters)

consisting of the strings that have the corresponding bit in bits set. str1 conrresponds to
(c

bit(), str2 to bit 1, and so on. NULL values in str1, str2, ... are not appnded to the result.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 145

Notes

e
in
nl
O
Retuns a string representaion of the octal value of N, where N is a longlong
(BIGINT) number. This is equivalent to CONV(N, 10, B) retuns NULL if N is NULL

ty
r si
ve
ni

If the leftmost character of the string str is a multibyte character, returns the code
for that character, calculated from the numeric values of its consitent bytes using theis
formula-
U

If the leftmost chacacter is not a multibyte character, ORD() returns the same value
as the ASC() function
ity
m
)A

Quotes a string to produce a resuit that can be used as a properly escaped data
(c

vfalue in an SQL statement. The string is returned endosed by single quotes and with
each instance of single quote (‘), backslash (/), ASCII NUL, and control Z preceded

Amity Directorate of Distance & Online Education

 146 Introduction to Data Base Management Systems

by a backslash, if the arguments is NULL, the return value is the word “NULL” without
Notes

e
endosing single quotes.

in
nl
O
Note: Please check if your installation has any bug with this functio then don’t use
this function.

ty
expr REGEXP pattern

this function performs a pattern match of expr against pattern. Returns 1 if expr
matches pat, otherwise it returns 0 if either expr or pat is NULL, the result is NULL

si
REGEXP is not case sensitive, except when used with binary strings.

r
ve
ni
U
ity
m
)A
(c

Returns a string consisting of the string atr repeated count times. If count is less
than 1, returns an emply string. Retuns NULL if str or count are NULL.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 147

Notes

e
in
nl
O
Returns the string str with all occurrences of the string from_str replaced by the
string to_str. REPLACE() performs a case-sensitive match when searching for from_str.

ty
r si
ve
ni
U
ity
m
)A

Returns the string str, right-padded with the string padstr to a length of len
characters. If str is longer than len, the return value is shorterned to len characters.
(c

Amity Directorate of Distance & Online Education

 148 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
si
Returns a soundex string from str. Two strings that sound almost the same should
have identical soundex strings. A standard soundexsting is four characters long, but the
r
SOUNDEX() function returns an arbitrarity long string. you can use SUBSTRING() on
ve
the result to get a standard soundard soundex string. All non alphabetic characaters in
str are ignored. All international alphabetic characters outside the A-Z range are treated
as vowels.
ni
U
ity
m
)A
(c

STRCMP(STR1, STR2)

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 149

Compares two strings and returns 0 if both strings are equal, if returns-1 if the first
Notes

e
argument is smaller than the second according to the current sort order otherwise if
returns 1.

in
nl
O
ty
r si
ve
ni
U
ity
m

at position pos. The forms that use FROM are standard SQL syntax. It is also
possible to use a negative value for pos. In this case, the beginning of the substring is
pos characters from the end of the string, rather than the beginning. A negative value
)A

any be used for pos in any of the forms of this function.

(c

Amity Directorate of Distance & Online Education

 150 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
r si
ve
ni
U

Returns the substring from string str before count occurrences of the delimiter
delim. If count is positive, everything to the left of the final delimiter (counting from the
ity

left) is retuned. If count is negative, everything to the right of the final delimiter (counting
from the right) is retunred. SUBSTRING_INDEX() performs a case senditive match
when searching for delim.
m
)A

Returns the string str with all remstr pretixes or suffixes removed. If none of the
(c

specifiers Both, LEADING, or TRALING is given BOTH is assumed. remstr is optional

and, if not specified, spaces are removed.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 151

Notes

e
in
nl
O
ty
r si
ve
ni
U
ity
m

Performs the inverse operation of HEX(str), That is, if interprets each pair of
hexadecimal digits in the argument as a number and converts it to the character
)A

represented by the number. The resulting characters are returned as a binary string.
(c

Amity Directorate of Distance & Online Education

 152 Introduction to Data Base Management Systems

Notes

e
in
nl
The characters in the argument string must e legal hexadecimal digits: ‘0’.. ‘9’,

O
‘A’.. ‘F’, ‘a’.. ‘I’. If UNHEX() ancounters any non-hexadecimal digits in the arguments, it
returns NULL.

returns the string str with all characters changed to uppercase according to the

ty
current characte set mapping.

r si
ve
3.2.10 Views
A database view is a subset of a database and is based totally on a query that
ni

runs on one or more database tables. Database views are stored within the database
as named queries and can be used to save often used, complicated queries. There are
styles of database perspectives:
U

◌◌ dynamic views
◌◌ static views
ity

Dynamic views :
Dynamic views can include information from one or two tables and automatically
include all of the columns from the specified desk or tables. Dynamic perspectives are
robotically refreshed while related gadgets or prolonged objects are made or modified.
m

At the point whilst you make a dynamic view with data from tables, you should make
sure that the two tables have the identical primarykeycolseq columns or comprise
specific indexes with the equal column call inside the equal request.
)A

Static views:
Static views can contain information from more than one tables and the critical
columns from these tables not actually settled in the choose and in which clauses of
the static view. Static views have to be manually refreshed whilst related gadgets or
(c

elevated items are made or modified.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 153

Database views are populated relying upon the factor on which they may be based
Notes

e
totally. As an instance, if you upload or dispense with an attribute from the workorder
object, the attribute is either added or taken out from the dynamic view that relies upon
the item. Right while you change an characteristic, no longer all progressions are

in
applied to the related database view. For instance, if you alternate the statistics sort
of an attribute, the change is carried out to the database view. Regardless, within the
occasion which you alternate or upload a site to the default worth of the workorder

nl
object, the change isn’t typically implemented to the database view. Taking everything
into account, you ought to observe this transformation to the database view.

O
3.2.11 SQL Joins
SQL join(be a part of )statements permit us to get admission to records from
or more tables right now. Additionally they preserve our database normalized.

ty
Normalization allows us to preserve records redundancy low so that we are able
to lower the quantity of facts anomalies in our software when we delete or update a
record. Different kinds of joins are:

si
◌◌ internal join
◌◌ left join
◌◌ proper be part of
◌◌ full be part of
r
ve
Student
ni
U
ity

StudentCourse
m

Course_ID Roll_No
1 1
2 2
)A

2 3
3 4
1 5
4 9
(c

5 10
4 11

Amity Directorate of Distance & Online Education

 154 Introduction to Data Base Management Systems

The simplest Join is INNER JOIN.

Notes

e
Inner join. Internal join: this cmd selects all rows from both the tables as long as
the circumstance satisfies. This may create the result-set by combining all rows from

in
each the tables in which the circumstance satisfies i. E cost of the commonplace field
might be identical.

nl
O
ty
We can also write JOIN instead of INNER JOIN. JOIN is same as INNER NOIN.

si
this query will show the names and age of students enrolled in different courses.

Select student course. Course_ID, studentNAME, StudentAGE FROM Stundet

INNER JOIN StudnetCourse ON StudentORLL_NO = studentCourse.ROLL_NO.
r
ve
Output:

Course_ID Name Age

1 harsh 18
ni

2 Pratik 16
2 Riyanka 20
3 Deep 18
U

1 Saptarhi 19

Left join: this cmd returns all of the rows of the desk at the left side of the join and
matching rows for the desk at the right side of be a part of. The rows for which there
ity

may be no matching row on right aspect, the resultant -set will contain null. Other name
for it’s miles left outer be part of.

Syntax:
Select Table1, column1,table1.column2,table2.column1,...
m

FROM table1
LEFT JOIN table2
ON table1.matching_column=table2.matching_column;
)A

table1:First table.
table2:second table
matching_column:Column common to both the tables.
(c

We can also use LEFT OUTER JOIN instead of LEFT JOIN, both are same.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 155

Notes

e
in
Name Course_ID

nl
Harsh 1
Pratik 2
Riyanka 2

O
Deep 3
Saptarhi 1
Dhanraj NULL

ty
Rohit Null
Niraj Null

Output :

si
Right join: this be part of is just like left join. It returns all of the rows of the table on
the right facet of the join and matching rows for the table at the left aspect of be a part

r
of. The rows for which there is no matching row on left facet, the consequent -set will
include null. Other call for it’s miles proper outer be a part of.
ve
Syntax:
Select Table1,column1,table1.column2,table2.column1,...
FROM table1
RIGHT JOIN table2
ni

ON table1.matching_column=table2.matching_column;

table1:First table.
U

table2:second table
matching_column:Column common to both the tables
ity

SELECT Student.NAME,StudentCourse.COURSE_ID
FROM Student
RIGHT JOIN StudentCourse
ON studentCourse.ROLL_NO=Student.RoLL_NO;
m

Output:

Name Course_ID
Harsh 1
)A

Pratik 2
Riyanka 2
Deep 3
Saptarhi 1
(c

NULL 4
NULL 5
NULL 4

Amity Directorate of Distance & Online Education

 156 Introduction to Data Base Management Systems

Notes Full join : full join creates the result-set by way of combining result of each left be

e
a part of and right join. The result-set will comprise all of the rows from both the tables.
The rows for which there may be no matching, the end result-set will comprise null

in
values.

Syntax:

nl
Select Table1,column1,table1.column2,table2.column1,...
FROM table1
FULL JOIN table2

O
ON table1.matching_column=table2.matching_column;

table1:First table.

ty
table2:second table
matching_column:Column common to both the tables

SELECT Student.NAME,StudentCourse.COURSE_ID

si
FROM Student
FULL JOIN StudentCourse
ON studentCourse.ROLL_NO=Student.RoLL_NO;
r
ve
Output:

Name Course_ID
Harsh 1
ni

Pratik 2
Riyanka 2
Deep 3
U

Saptarhi 1
Dhanraj NULL
Rohit NULL
ity

Niraj NUll
NULL 9
NULL 10
NULL 11
m

Check your understanding:

)A

A. Fill in the blanks:

1. A primary key ________identifies the record in the table.( uniquely)
2. A foreign key is used to establish ________between any two tables of RDBMS.(
relationship).
(c

3. A _____ in DBMS is an attribute or a set of attributes that help to uniquely identify a

tuple in a relation.(key)

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 157

4. A unique Key differs from a primary key because it can have only _____value.( one
Notes

e
null)
5. Super Key is the set of all the keys which help to identify rows in a table_______.

in
( uniquely)

B. State True or False:

nl
1. There can be only one candidate keys in a table.(false)
2. Primary keys cannot contain null values(true)
3. Foreign key establishes a relation between two tables(true)

O
4. Unique key is similar to primary key but can have only one null value(.true)
5. Super key is not a super set of a candidate key.(false).

ty
Activity
1. Prepare a power point presentation on ‘Data Definition Language and its Features’.

si
Questions and exercises
1. What is the significance of a primary key?
2. Write the significance of a foreign key.
r
ve
3. List the types of keys
4. Which key is used to establish relation between two tables. Explain.
5. Describe all the types of keys.
ni

Summary
●● KEY is used to uniquely identify any record or row of data from the table. It is also
used to establish and identify relationships between tables.
U

Types of Keys
◌◌ Super keys
ity

◌◌ Candidate keys
◌◌ Primary keys
◌◌ Alternate/ Secondary keys
◌◌ Foreign keys
m

◌◌ Composite keys

Glossary
)A

Primary key
●● It is the first key which is used to identify one and only one instance of an entity
uniquely. An entity can contain multiple keys as we saw in PERSON table. The key
which is most suitable from those lists become a primary key.
(c

Candidate key

Amity Directorate of Distance & Online Education

 158 Introduction to Data Base Management Systems

●● A candidate key is an attribute or set of an attribute which can uniquely identify

Notes

e
a tuple . The remaining attributes except for primary key are considered as a
candidate key. The candidate keys are as strong as the primary key.

in
Super Key
●● Super key is a set of an attribute which can uniquely identify a tuple. Super key is
a superset of a candidate key.

nl
Foreign key
●● Foreign keys are the column of the table which is used to point to the primary key

O
of another table.

Composite Key:

ty
●● Sometimes single attribute fails to uniquely identify the rows/tuples in a table.
Composite Key is a combination of two or more attributes that uniquely identify
rows in a table. (Eg: Name, DOB).Always have two or more attributes.

si
Further Readings
1. Hectar Garcia-Molina, Jeff Ullman and Jennifer Widom. (2009). Database

r
Systems: The Complete Book. Second Edition. Pearson Education Inc.
ve
2. Thomas Connolly and Carolyn Begg. (2014). Database Systems: A Practical
Approach to Design, Implementation and Management. Sixth Edition. Pearson
Education, Inc.
3. C.J. Date. (2004). An Introduction to Database Systems. Pearson Education,
ni

Inc.

Check your understanding(answers):

U

A. Fill in the blanks

1. Uniquely
2. Relationship
ity

3. Key
4. One null
5. Uniquely
m

B. State True or False:

1. False
)A

2. True
3. True
4. True
5. False
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 159

Module - 4 : Database Design

Notes

e
Structure:

in
4.1 Functional Dependencies
4.1.1 Introduction

nl
4.1.2 Rules of Functional Dependency
4.1.3 Types of Functional Dependencies

O
4.2 Normalization
4.2.1 Design of Database Schema Anomalies
4.2.2 Process of Normalization

ty
4.2.3 First Normal Form
4.2.4 Second Normal Form

si
4.2.5 Third Normal Formal
4.2.6 Boyce -CODD Normal Form
4.2.7 Multivalued Dependency
4.2.8 Forth Normal Form
r
ve
4.2.9 Join Dependency
4.2.10 Fifth Normal Form
CASE STUDY
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 160 Introduction to Data Base Management Systems

Unit - 4.1: Functional Dependencies

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Learn about Rules of Functional Dependency

nl
●● Know the types of Functional Dependencies

4.1.1 Introduction

O
Functional Dependencies
Functional dependency (FD) is a set of constraints between two attributes in a
relation. Functional dependency says that assuming two tuples have same values for

ty
attributes A1, A2,..., An, those two tuples should have to have same values for attributes
B1, B2, ..., Bn.

si
A functional dependency A->B in a relation holds if two tuples having same value of
attribute An also have same value for attribute B. A functional dependency is indicated
by an arrow “→”. The functional dependency of X on Y is represented by X → Y.

r
The following is an Egs would make it easier to understand functional dependency
ve
–

Student Table 1

Stud_No Stud_Name Stud_ Stud_State Stud_ Stud_AGE

ni

Phone country
1 Ram 976271721 Haryana India 20
2 Ram 9898291281 Pubjab India 19
U

3 Sujit 7898291981 Rajsthan India 18

4 Suresh Punjab India 21

STUD_NO->STUD_NAME, STUD_NO->STUD_PHONE hold

ity

but

STUD_NAME->STUD_ADDR do not hold

The functional dependency is a relationship that exists between two attributes. It

m

typically exists between the primary key and non-key attribute within a table.

X → Y
)A

The left side of FD is known as a determinant, the right side of the production is
known as a dependent.

For Eg:

Assume we have an EMP table with attributes: Emp_Id, E_NAME, Emp_Address.

(c

Here Emp_Id attribute can uniquely identify the E_NAME attribute of EMP table
because if we know the Emp_Id, we can tell that EMP name associated with it.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 161

Functional dependency can be written as:

Notes

e
Emp_Id → E_NAME

We can say that E_NAME is functionally dependent on Emp_Id.

in
We have a <Department> table with two attributes – DeptId and DeptName.

DeptId = Department ID

nl
DeptName = Department Name

The DeptId is our primary key. Here, DeptId uniquely identifies the DeptName

O
attribute. This is because if you want to know the department name, then at first you
need to have the DeptId.

DeptId DeptName

ty
001 Finance

002 Marketing

si
003 HR

Therefore, the above functional dependency between DeptId and DeptName can
be determined as DeptId is functionally dependent on DeptName –

DeptId -> DeptName

r
ve
Key Terms Description
Axiom Axioms is a set of inference rules used to infer all the functional
dependences on a relational database.
ni

Decomposition It is a rule that suggests if you have a table that appears to

contain two entities which are determined by the same primary
key then yhou should consider breaking them up into two
U

different tables.
Dependent It is desplayed on the righ side of the functional dependencey
diagR.
Determinant It is displayed on the left side of the functional dependency
ity

diagR.
Union It suggests that if two tables are separate, and the PK is the
same, you should consider putting them. together
m

Armstrong’s Axioms
On the off chance that F is a set of functional dependencies, the closure of
F, denoted as F+, is the set of all functional dependencies logically inferred by F.
)A

Armstrong’s Axioms are a set of rules, that when applied repeatedly, creates a closure
of functional dependencies.

●● Reflexive rule – If alpha is a set of attributes and beta is_subset_of alpha, then
alpha holds beta.
(c

●● Augmentation rule – If a → b holds and y is attribute set, then ay → by also

holds. That is adding attributes in dependencies, does not change the basic
dependencies.
Amity Directorate of Distance & Online Education
 162 Introduction to Data Base Management Systems

●● Transitivity rule – Same as transitive rule in algebra, if a → b holds and b → c

Notes

e
holds, then a → c also holds. a → b is called as a functionally that determines b.
How to find functional dependencies for a relation?

in
Functional Dependencies in a relation are dependent on the domain of the
relation. Consider the STUDENT relation given in Table 1.

nl
●● We know that STUD_NO is unique for each student. So STUD_NO->STUD_
NAME, STUD_NO->STUD_PHONE, STUD_NO->STUD_STATE, STUD_NO-
>STUD_COUNTRY and STUD_NO -> STUD_AGE all will be true.

O
●● Similarly, STUD_STATE->STUD_COUNTRY will be true as if two records have
same STUD_STATE, they will have same STUD_COUNTRY as well.
●● For relation STUDENT_COURSE, COURSE_NO->COURSE_NAME will be true
as two records with same COURSE_NO will have same COURSE_NAME.

ty
4.1.2 Rules of Functional Dependency
Below are the Three most important rules for Functional Dependency in Database:

si
●● Reflexive rule –. If X is a set of attributes and Y is_subset_of X, then X holds a
value of Y.
●● r
Augmentation rule: When x -> y holds, and c is attribute set, then ac -> bc also
ve
holds. That is adding attributes which do not change the basic dependencies.
●● Transitivity rule: This rule is very much similar to the transitive rule in algebra if x ->
y holds and y -> z holds, then x -> z also holds. X -> y is called as functionally that
determines y.
ni

4.1.3 Types of Functional Dependencies

There are mainly four types of Functional Dependency in DBMS. Following are the
U

types of Functional Dependencies in DBMS:

◌◌ Multivalued Dependency
◌◌ Trivial Functional Dependency
ity

◌◌ Non-Trivial Functional Dependency

◌◌ Transitive Dependency

Multivalued Dependency in DBMS:

m

Multivalued dependency occurs in the situation where there are multiple

independent multivalued attributes in a single table. It happens when there are more
than one independent multivalued attributes in a table. A multivalued dependency is a
)A

finished constraint between two sets of attributes in a relation. It requires that certain
tuples be accessible in a relation.

●● Multivalued dependency happens when two attributes in a table are independent

of one another be that as it may, both depend on a third attribute.
(c

●● A multivalued dependency comprises of at least two attributes that are dependent

on a third attribute that’s the reason it generally needs at least three attributes• It is
represented by the symbol →→
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 163

●● If a table has attributes P, Q and R, then Q and R are multi-valued facts of P. P

Notes

e
→→ Q P →→ R
●● 4NF is related to multivalued dependency.

in
Eg –Let there be a relation Student StudentName CourseDiscipline Activities
Amit Mathematics Singing Amit Mathematics Dancing Yuvraj Computers Cricket Akash
Literature Dancing Akash

nl
Literature Cricket Akash Literature Singing
●● In the above table, we can see Students Amit and Akash have interest in more

O
than one movement.
●● This is multivalued dependency since Course Discipline of a student are
independent of Activities, however are dependent on the student.

ty
●● Therefore, multivalued dependency – StudentName ->-> CourseDiscipline
StudentName ->-> Activities
Eg:

si
Car_model Maf_year Color
H001 2017 Metallic
H001
H005
2017
2018
r Green
Metallic
ve
H005 2018 Blue
H010 2015 Metallic
H033 2012 Gray
ni

In this Eg, maf_year and color are independent of each other but dependent on
car_model. In this Eg, these two columns are said to be multivalue dependent on car_
model.
U

This dependence can be represented like this:

car_model ->>maf_year
ity

car_model->>colour

n DBMS, a table is characterized to believe Multivalued dependency when the

succeeding conditions follow:
m

◌◌ For a dependency XY, in the event that for the only value of X, many values of
B happens, the table might hold multivalued dependency.
◌◌ Besides, a table ought to incorporate minimum 3 table columns for it to have a
)A

multivalued dependency.
◌◌ Finally, for a relation R(X, Y, Z), if there exists a multivalued dependency
among X and Y, so Y and Z ought to must be independent of one another.
In this way, when every one of these above conditions are fulfilled for any relation
in a table then it is characterized to have a multivalued dependency.
(c

Amity Directorate of Distance & Online Education

 164 Introduction to Data Base Management Systems

Multivalued Dependency is likewise an extraordinary Eg of a Join Dependency,

Notes

e
having only two groups of values contained for example can be said as a binary join
dependency.

in
Trivial Functional Dependency in DBMS
The Trivial dependency is a set of attributes which are known as a trivial if the set
of attributes are associated with that property.

nl
Along these lines, X - > Y is a trivial valuable dependency If Y is a subset of X.

How about we comprehend with a Trivial Functional Dependency Eg.

O
o A → B has trivial functional dependency if B is a subset of A.

o The following dependencies are also trivial like: A → A, B → B

ty
For Eg1:

Emp_id E_NAME

si
AS555 H
AS811 George
AS999 Kevin
r
Consider this table of with two columns Emp_id and E_NAME.
ve
{Emp_id, E_NAME} -> Emp_id is a trivial functional dependency as Emp_id is a
subset of {Emp_id,E_NAME}.

Eg2: Consider a table with two columns Student_id and Student_Name.

ni

{Student_Id, Student_Name} -> Student_Id is a trivial functional dependency as

Student_Id is a subset of {Student_Id, Student_Name}. That makes sense because if
we know the values of Student_Id and Student_Name then the value of Student_Id can
U

be uniquely determined.

Also, Student_Id -> Student_Id & Student_Name -> Student_Name are trivial
dependencies too.
ity

Non Trivial Functional Dependency

Functional dependency which also known as a nontrivial dependency happens
when A->B remains constant where B isn’t a subset of A. In a relationship, in the event
m

that attribute B isn’t a subset of attribute A, it is considered as a non-trivial dependency.

◌◌ A → B has a non-trivial functional dependency if B is not a subset of A.

)A

◌◌ When A intersection B is NULL, then A → B is called as complete non-trivial.

On the off chance that a FD X->Y remains constant where X intersection Y is null
then this dependency is supposed to be completely non trivial function dependency.

Company CEO Age

(c

Microsoft Satya Nadella 51

Google Sundar Pichai 46
Apple Tim Cook 57
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 165

Eg:
Notes

e
(Company} -> {CEO} (if we know the Company, we knows the CEO name)

But CEO is not a subset of Company, and hence it’s non-trivial functional

in
dependency.

Eg:

nl
This kind of design can contribute to data anomalies and inconsistencies, for
example:

◌◌ If you deleted the two books Children of the Mind and Ender’s Game, you

O
would delete the author “Orson Scott Card” and his nationality completely
from the database.
◌◌ You cannot add a new author to the database unless you also add a book.

ty
What if the author is yet unpublished or you don’t know the name of a book
they authored?
◌◌ If “Orson Scott Card” changed his citizenship, you would have to change his

si
citizenship in all records in which he appears. Having multiple records with
the same author can result in inaccurate data. What if the data entry person
doesn’t realize there are multiple records for someone, and changes the data

◌◌
in only one record?
r
You can’t delete a book such as The Handmaid’s Tale without also completely
ve
deleting the author.
An employee table with three attributes: emp_ie, emp_name, emp_address.
The following functional dependencies are non-trivial:
ni

emp_id->emp_name (emp_name is not a subset of emp_id)

emp_id->emp_address (emp_address is not a subset of emp_id)

Transitive Dependency
U

A functional dependency is said to be transitive if it is indirectly formed by two

functional dependenceies. For e.g. X->Z is a transitive dependency if the following
functional dependencies hold tru: X->Y. Y does not->X.
ity

How do you find transitive dependency?

When an indirect ralationship causes functional dependency it is called Transitive

Dependency. IfP->Q and Q-> R is true, then P-> R is a transitive dependency. To
achieve 3NF, eliminate the Transitive Dependency
m

A Transitive Dependency is a type of functional dependency which happends

when ‘t’ is indirectly formed by two functional dependencies. Let’s understand with the
)A

following Transitive Dependency Example.

Example:

Company CEO Age

Microsoft Satya Nadela 51
(c

Google Sundar Pichai 46

Alibaba Jack Ma 54

Amity Directorate of Distance & Online Education

 166 Introduction to Data Base Management Systems

{Company}->}CEO} (if we know the compay, we know its CEO’s name)

Notes

e
{CEO} ->(Age) If we know the CEO, we know the Age

Therefore according to the rule of rule of transitive dependency:

in
{Company}->(Age} should hold that makes sense because if we know the company
name, we can know his age.

nl
Note: You need to remember that transitive dependency can only occur in a
relation of three or more attributes.

O
Authors

Authro_ID Author Book Author_Nationality

Auth_001 Orson Scott Card Ender’s Game United States

ty
Auth_001 Orson Scott Card Children of the Mind United States
Auth_002 Margaret Atwood The Hnadmaid’s Tale Canada

In the Authors example above:

si
◌◌ Book: Author; Here, the book attribute determines the Author attribute. If you
know the book name, you can learn the author;s name. however, author
r
doesn’t determine book, because an author can write multiple books. For
example, just because we know the book name.
ve
◌◌ Author: Author_Nationality: Likewise, the Author attribute determines the
Author_Nationality, but not the other way around-just because we know the
author’s nationality doesn’t mean we can determine the author.
ni

But this table introduces a transitive dependency:

◌◌ Book: Author_nationality; If we know the book name, we can determine the

author’s antionality via the Author column.
U

Why Transitive Dependencies Are bad database design

What is the value of avoiding transitive dependencies to help ensure 3NF? Let’s
consider our first table again and see the issues it creates:
ity

Authors

Authro_ID Author Book Author_Nationality

Auth_001 Orson Scott Card Ender’s Game United States
m

Auth_001 Orson Scott Card Children of the Mind United States

Auth_002 Margaret Atwood The Hnadmaid’s Tale Canada
)A

This kind of design can contribute to data anomalies and inconsistencies, for
example:

◌◌ If you deleted the two books Children of the Mind and Ender’s Game, you
would delete the author “Orson Scott Card” and his nationality completely
from the database.
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 167

◌◌ You cannot add a new author to the database unless you also add a book.
Notes

e
What if the author is yet unpublished or you don’t know the name of a book
they authored?

in
◌◌ If “Orson Scott Card” changed his citizenship, you would have to change his
citizenship in all records in which he appears. Having multiple records with
the same author can result in inaccurate data. What if the data entry person
doesn’t realize there are multiple records for someone, and changes the data

nl
in only one record?
◌◌ You can’t delete a book such as The Handmaid’s Tale without also completely
deleting the author.

O
Advantages of Functional Dependency
●● Functional Dependency avoids data redundancy. Therefore same data do not

ty
repeat at multiple locations in that database
●● It helps you to maintain the quality of data in the database
●● It helps you to defined meanings and constraints of databases

si
●● It helps you to identify bad designs
It helps you to find the facts regarding the database design

Check your understanding

r
ve
A. Fill in the blanks:
1. The arrow notation ‘→’ in FD is read as ________.
2. The term full functional dependency (FFD) is used to indicate the ________ set of
ni

attributes .
3. A functional dependency in the set is redundant if it can be derived from the other
________ in the set.
U

4. A functional decomposition is the process of ________ the functions .

5. A functional dependency establishes a relationship between _____sets of attributes..
ity

B. State true or false:

1. Armstrong’s axioms allow us to generate all F+ for any given F
2. Functional dependencies are transitive
m

3. A functional dependency is a relationship between an attribute “Y” and a determinant

(1 or more other attributes) such that for a given value of a determinant the value of
the attribute Y is uniquely defined.
)A

4. Transitivity of functional dependencies means that if a →b and a→c then c→b.

5. Functional dependencies are reflexive, i.e., if a →b then b →a.

Activity
(c

1. Make a presentation on the topic Types of Functional Dependencies.

Amity Directorate of Distance & Online Education

 168 Introduction to Data Base Management Systems

Summary
Notes

e
●● A functional dependency between two columns, X and Y, means that for any two
records R1 and R2 in the table, if field X of record R1 contains value x and field

in
X of record R2 contains the same value x, then if field Y of record R1 contains
the value y, then field Y of record R2 must contain the value y. We can say that
attribute Y is determined by attribute X.

nl
●● Functional dependencies may exist between multiple source columns.
●● A functional dependency establishes a relationship between two sets of attributes.
●● A transitive dependency exists when you have the following functional dependency

O
pattern:
o A→B and B→C; therefore A→C

ty
●● A transitive dependency therefore exists only when the determinant that is not the
primary key is not a candidate key for the relation.
●● A second normal form relation that has no transitive dependencies is, of course,
automatically in third normal form.

si
Glossary
●●
r
The attribute set on the left side of the arrow, X is called Determinant
ve
●● The attribute set on the right side, Y is called the Dependent.
●● Types of FD:
o Trivial functional dependency
ni

o Non-Trivial functional dependency

o Multivalued functional dependency
o Transitive functional dependency
U

●● In Trivial Functional Dependency, a dependent is always a subset of the

determinant.
●● In Non-trivial functional dependency, the dependent is strictly not a subset of the
ity

determinant.
●● In Multivalued functional dependency, entities of the dependent set are not
dependent on each other.
m

●● In transitive functional dependency, dependent is indirectly dependent on

determinant.

Check your Understanding (answers)

)A

A. Fill In The Blanks:

1. Functional Dependency
2. Relationship
(c

3. Dependencies

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 169

4. analyzing the links and relationships between all the components that create a
Notes

e
functional relationship
5. Two

in
B. State True or False:
1. True

nl
2. False
3. True
4. True

O
5. True

ty
r si
ve
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 170 Introduction to Data Base Management Systems

Unit -4.2: Normalization

Notes

e
Objectives

in
At the end of this unit, you will be able to:

●● Learn about Design of Database Schema Anomalies

nl
●● Know about the Process of Normalization
●● Understand First, Second, Third, Forth and Fifth Normal Form

O
●● Know about Boyce -CODD Normal Form
●● Learn Multivalued Dependency
●● Know about Join Dependency

ty
Introduction
Normalization is a procedure for getting sorted out the data in the database which

si
helps you with keeping away from data redundancy, insertion, update and deletion
anomaly. It is a course of analyzing the relation schemas dependent on their diverse
useful dependencies and primary key. Normalization is inborn to relational database
r
theory. It may duplicate similar data inside the database which may achieve the
development of extra tables.
ve
In the event that a database configuration isn’t great, it might contain anomalies,
which resemble a terrible dream for any database administrator. Dealing with a
database with anomalies is close to unimaginable.
ni

4.2.1 Design of Database Schema Anomalies

U

Anomalies
There are various kinds of anomalies which can happen in referencing and
referenced relation which can be examined as:
ity

Insertion anomaly: If a tuple is inserted in referencing relation and referencing

attribute value is absent in referenced attribute, it will not permit installing in referencing
relation. For Eg, If we endeavor to install a record in STUDENT_COURSE with STUD_
NO =7, it will not permit.
m

Deletion and Updation anomaly: If a tuple is deleted or updated from referenced

relation and referenced attribute value is used by referencing attribute in referencing
relation, it will not permit erasing the tuple from referenced relation. For Eg, If we
endeavor to delete a record from STUDENT with STUD_NO =1, it will not permit. To
)A

keep away from this, following can be utilized in query:

◌◌ ON DELETE/UPDATE SET NULL: If a tuple is deleted or updated from

referenced relation and referenced attribute value is utilized by referencing
attribute in referencing relation, it will erase/update the tuple from referenced
(c

relation and set the value of referencing attribute to NULL.

◌◌ ON DELETE/UPDATE CASCADE: If a tuple is deleted or updated from

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 171

referenced relation and referenced attribute value is utilized by referencing

Notes

e
attribute in referencing relation, it will erase/update the tuple from referenced
relation and referencing relation too.

in
◌◌ Update anomalies – If data things are dissipated and are not linked to
one another appropriately, then, at that point, it could prompt unusual
circumstances. For Eg, when we attempt to update one data thing having
its duplicates dispersed more than a few places, a couple of occasions get

nl
updated appropriately while a couple of others are left with old values. Such
occurrences leave the database in an inconsistent state.
Normalization is a strategy to eliminate this large number of anomalies and carry

O
the database to a consistent state.

Database normalization is a stepwise proper cycle that licenses us to decompose

database tables so the two data dependency and update anomalies are limited. It

ty
uses useful dependency that exists in the table and primary key or candidate key in
analyzing the tables. Normal forms were at first proposed called First Normal Form
(INF), Second Normal Form (2NF), and Third Normal Form (3NF).

si
Consequently, R, Boyce, and E. F. Codd introduced a more grounded importance
of 3NF called Boyce-Codd Normal Form. Except for 1NF, this heap of normal forms rely
upon practical dependency among the attributes of a table. Higher normal forms that go
r
past BCNF were introduced later like Fourth Normal Form (4NF) and Fifth Normal Form
ve
(5NF). Regardless, these later normal forms oversee conditions that are outstandingly
unprecedented.

Benefits:
ni

There are many advantages to normalizing a database. A portion of the principle

benefits are:

◌◌ By utilizing normalization redundancy of database or data duplication can be

U

settled.
◌◌ We can limit null values by utilizing normalization.
◌◌ Results in a more minimized database (because of less data redundancy/
ity

zero).
◌◌ Minimize/keep away from data adjustment issues.
◌◌ It works on the query.
◌◌ The database structure is more clear and more obvious.
m

◌◌ The database can be extended without influencing existing data.

◌◌ Finding, Sorting, and Indexing can be quicker in light of the fact that the table
is little and more rows can be obliged on the data page.
)A

Normalization
o Normalization is the process of organizing the data in the database.
o Normalization is used to minimize the redundancy from a relation or set of relations.
(c

It is also used to eliminate the undesirable characteristics like Insertion, Update and
Deletion Anomalies.

Amity Directorate of Distance & Online Education

 172 Introduction to Data Base Management Systems

o Normalization divides the larger table into the smaller table and links them using
Notes

e
relationship.
o The normal form is used to reduce redundancy from the database table.

in
Types of Normal Forms
There are the four types of normal forms:

nl
O
ty
r si
ve
ni
U
ity
m

4.2.2 Process of Normalization

Database normalization is a stepwise formal process that allows us to decompose
database tables in such a way that both data dependency and update anomalies are
)A

minimized. It makes use of functional dependency that exists in the table and primary
key or candidate key in analyzing the tables. Normal forms were initially proposed
called First Normal Form (INF), Second Normal Form (2NF), and Third Normal Form
(3NF).
(c

Subsequently, R, Boyce, and E. F. Codd introduced a stronger definition of 3NF

called Boyce-Codd Normal Form. With the exception of 1NF, all these normal forms are
based on functional dependency among the attributes of a table. Higher normal forms

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 173

that go beyond BCNF were introduced later such as Fourth Normal Form (4NF) and
Notes

e
Fifth Normal Form (5NF). However, these later normal forms deal with situations that
are very rare.

in
nl
O
ty
si
Summary of Normalization in a Nutshell

Normal
Form
Test
r Remedy (Normalization)
ve
1NF Relation should have no non-atomic Form name relation for each
attributes or nested relations. non-atomic attribute or nested
relation.
2NF For relations where primary key contains Decompose and set up a new
ni

multiple attributes, no non-key attributes relation for each partial key

should be functionally dependent on a part with its dependent attributes.
of the primary key. Make sure to keep a relation
U

with the original primary key

and any attributes that are fully
functionally dependent on it.
ity

3NF Relations should not have a non-key Decompose and set up a

attribute functionally determined by another relation that includes the non-
non-key attribute (or by a sets of non-key key attributes(s) that functionally
attributes) i.e, there should be no transitive determine(s) other non-key
dependency of a non-key attribute of the attribute(s).
m

primary key.
BCNF Relations should not have any attribute Decompose the table into two
in Functional Dependency which is non- subtables.
)A

prime, the attribute that doesn’t occur in

any candidate key.
4NF The relation should not have a multi-value Decompose the table into two
dependency means it occur when two subtables.
(c

attributes of a table are independent of each

other but both depend on a third attribute.

Amity Directorate of Distance & Online Education

 174 Introduction to Data Base Management Systems

5NF The relations should not have join Decompose all the tables into
Notes

e
dependency means if a table can be as many as possible numbers in
recreated by joining multiple tables and order to avoid dependency.

in
each of the tables has a subset of the
attributes of the table, then the table is in
Join Dependency.

nl
4.2.3 First Normal Form
First Normal Form

O
1NF (First Normal Form) Rules

◌◌ Each table cell should contain a single bvalue.

ty
◌◌ Each record needs to be unique.

r si
ve
ni
U
ity

First Normal Form is defines in the definition of relations (tables) itself. This rule
defines that all the attributes in a relation must have atomic domains. The values in an
atomic domain are indivisible units.
m
)A

We re-arrange the relation (table) as below, to convert it to First Normal Form.

(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 175

Notes

e
in
nl
O
Each attribute must contain only a single value from its pre-defines domain.

ty
4.2.4 Second Normal Form
2NF (Second Normal Form) Rules

si
◌◌ Rule 1: Be in 1NF.
◌◌ Rule 2: Single Column Primary Key that does not functionally dependent on
any subset of candidate key relation.
r
It is clear that we cannot move forward to make our simple database in 2nd
ve
Normalization form unless we partition the table above.
ni
U
ity
m

We have divided our 1NF table into two tables viz Table 1 and Table 2. Table 1
contains member information. Table 2 contains information on movies rented.
)A

We have introduced a new column called Membership-id which is the primary key
for table .

1. Records can be uniquely identified in table 1 using Membership id

(c

Before we learn about the second normal form, we need to understand the
following:

Amity Directorate of Distance & Online Education

 176 Introduction to Data Base Management Systems

◌◌ Prime attribute: an attribute, which is part of the candidate-key, is known as a

Notes

e
prime attribute.
◌◌ Non-prime attribute: an attribute, which is not a part of the prime-key, is said to

in
be a non-prime attribute.
If we follow second normal form, then every non-prime attribute should be fully
functionally dependent on prime key attribute. That is, if X = A holds, then there should

nl
not be any proper subset Y of X, for which Y = A also holds true.

O
ty
r si
ve
We broke the relation in two as depicted in the above picture. So there exists no
ni

partial dependency.

4.2.5 Third Normal Form

U

3NF (Third Normal Form) Rules

◌◌ Rule 1: Be in 2NF
ity

◌◌ Rule 2: Has no transitive functional dependencies

To move our 2NF table into 3NF, we again need to again divide our table.
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 177

Notes

e
in
nl
O
We have again divided our tables and created a new table which stores

ty
Salutations.

There are no transitive functional dependencies, and hence our table is in 3NF

In table 3 Salutation ID is primary key, and in Table 1 Salutation ID is foreign to

si
primary key in Table 3.

For a relation to be in Third Normal Form, it must be in Second Normal form and
the following must satisfy: r
ve
◌◌ No non-prime attribute is transitively dependent on prime key attribute.
◌◌ For any non-trivial functional dependency, , then either:
o X is a superkey or,
ni

o A is prime attribute.
U
ity

We find that in the above Student-detail relation, Stu-ID is the key and only prime
key attribute. We find that City can be identified by Stu-ID as well as Zip itself. Neither
Zip is a superkey nor is City a prime attribute. Additionally, , so there exists transitive
m

dependency.

To bring this relation into third normal form, we break the realtion into two relations
as follow:
)A
(c

Amity Directorate of Distance & Online Education

 178 Introduction to Data Base Management Systems

Notes

e
in
nl
4.2.6 Boyce-Codd Normal Form

O
Boyce-Codd Normal Form (BCNF) is an extension of Third Normal Form on strict
terms. BCNF states that –

●● For any non-trivial functional dependency, X → A, X must be a super-key.

ty
In the above image, Stu_ID is the super-key in the relation Student_Detail and Zip
is the super-key in the relation ZipCodes. So,

Stu_ID → Stu_Name, Zip

si
and

Zip → City
r
Which confirms that both the relations are in BCNF.
ve
Boyce Codd normal form (BCNF)
o BCNF is the advance version of 3NF. It is stricter than 3NF.
ni

o A table is in BCNF if every functional dependency X → Y, X is the super key of the

table.
o For BCNF, the table should be in 3NF, and for every FD, LHS is super key.
U

Eg: Let’s assume there is a company where EMPs work in more than one
department.
ity

EMP table:

EMP_ID EMP_ EMP_DEPT DEPT_TYPE EMP_DEPT_

COUNTRY NO
264 India Designing D394 283
m

264 India Testing D394 300

364 UK Stores D283 232
364 UK Developing D283 549
)A

In the above table Functional dependencies are as follows:

◌◌ EMP_ID → EMP_COUNTRY
◌◌ EMP_DEPT → {DEPT_TYPE, EMP_DEPT_NO}
(c

Candidate key: {EMP-ID, EMP-DEPT}

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 179

16M
Notes

e
362

How to find Nth Highest SAL in SQL

in
The table is not in BCNF because neither EMP_DEPT nor EMP_ID alone are keys.

To convert the given table into BCNF, we decompose it into three tables:

nl
EMP_COUNTRY table:

EMP_ID EMP_COUNTRY

O
264 India
264 India

ty
EMP_DEPT table:

EMP_DEPT DEPT_TYPE EMP_DEPT_NO

Designing D394 283

si
Testing D394 300
Stores D283 232
Developing D283 549
r
ve
EMP_DEPT_MAPPING table:

EMP_ID EMP_DEPT
D394 283
D394 300
ni

D283 232
D283 549
U

Functional dependencies:
◌◌ EMP_ID → EMP_COUNTRY
◌◌ EMP_DEPT → {DEPT_TYPE, EMP_DEPT_NO}
ity

Candidate keys:
Forthefirsttable: EMP_ID

Forthesecondtable: EMP_DEPT
m

For the third table: {EMP_ID, EMP_DEPT}

Now, this is in BCNF because left side part of both the functional dependencies is a
)A

key.
(c

Amity Directorate of Distance & Online Education

 180 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
r si
ve
ni
U
ity
m

4.2.7 Multivalued Dependency

MVD or multivalued dependency means that for a single value of attribute ‘a’
)A

multiple values of attribute ‘b’ exist. We write it as,

a → --> b

It is read as: a is multi-valued dependent on b.

(c

Multivalued dependency (MVD) is having the presence of one or more rows

in a table. It implies the presence of one or more other rows in that same table. A
multivalued dependency prevents fourth normal form. A multivalued dependency

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 181

involves at least three attributes of a table.

Notes

e
It is represented with a symb“l “-”->” in DBMS.

X->Y relates one value of X to one value of Y.

in
X->->Y (read as X multidetermines Y) relates one value of X to many values of Y.

A Nontrivial MVD occurs when X->->Y and X->->z where Y and Z are not

nl
dependent are independent to each other. Non-trivial MVD produces redundancy.

Suppose a person named Geeks is working on 2 projects Microsoft and Oracle

and has 2 hobbies namely Reading and Music. This can be expressed in a tabular

O
format in the following way.

ty
r si
ve
Project and Hobby are multivalued attributes as they have more than one value for
a single person i.e., Geeks.
ni

Multi Valued Dependency (MVD) :

We can say that multivalued dependency exists if the following conditions are met.
U

Conditions for MVD:

Any attribute say a multiple define another attribute b; if any legal relation , for
all pairs of tuples t1 and t2 in r, such that,
ity

t1[a]=t2[a]

Then there exists t3 and t4 in r such that.

t1[a]=t2[a]= t3[a]=t4[a]
m

t1[b]=t3[b]; t2[b]=t4[b]

t1=t4; t2=t3
)A

Then multivalued (MVD) dependency exists.

To check the MVD in given table, we apply the conditions stated above and we
check it with the values in the given table.
(c

Amity Directorate of Distance & Online Education

 182 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Condition – 1 for MVD

ty
t1[a]=t2[a]=t3[a]=t4[a]

Finding from table,

si
t1[a]=t2[a]=t3[a]=t4[a]=Geeks

So, condition 1 is satisfied.

Condition – 2 for MVD

r
ve
t1[b]=t3[b]

And

t2[b]=t4[b]
ni

Finding from the table,

t1[b]=t3[b]=MS
U

And

t2[b]=t4[b]=Oracle
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 183

Notes

e
in
nl
O
ty
r si
ve
ni

We use multivalued conditions in two different ways:

To test relations to decide if they are lawful under a given arrangement of practical
U

and multivalued dependencies.

To determine limitations on the arrangement of lawful relations. We will concern

ourselves with the relations that fulfil a given arrangement of practical and multivalued
ity

dependencies.

MVD Transitive Rule

m
)A
(c

Amity Directorate of Distance & Online Education

 184 Introduction to Data Base Management Systems

Notes

e
in
nl
O
The 4th Normal Form can cause the Multivalued Dependencies. If a relation is in
Boyce code Normal form, it has to remove the multivalued Dependencies.

ty
Explanation: The multivalued dependencies is that, if there is a dependency or
relation in a table, then on value has multiple dependencies occur.

Let us consider an Example as given below. Consider the following table:

si
id Department Shift
1 Coding day
2 r Hr day
ve
3 Network night

In the above table, id2 has two departments Hr and Network. And shift timing day
and night.
ni

When we select the details with the id2, then it will result the tables as follows:
U

id Department Shift
2 Hr day
2 Network nigh
2 Hr night
ity

2 Network day

This means there exist multivalued dependencies. In this, relation between

department and shift is nothing.
m

This can be rectified by removing the multivalued dependency as, making this data
in to two tables as below:
)A

Table 1

id Department
1 Coding
2 Hr
(c

2 Network

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 185

Table 2
Notes

e
Id Shift
1 Day

in
2 Day
2 Night

nl
The 4th NORMAL FORM is applied to remove the multivalued dependencies
in the information table. The fourth everyday form hence defines the multivalued
dependencies.

O
1. multivalued dependency happens while attributes in a desk are impartial of each
different however, both rely on a third attribute.
2. a multivalued dependency consists of at least attributes that are depending on a

ty
third attribute it really is why it usually calls for as a minimum 3 attributes.
3. Eg: assume there is a motorcycle manufacturer corporation which produces
colors(white and black) of each model every 12 months.

si
BIKE_MODEL MANUF_YEAR COLOR
M2011 2008 White
M2001 2008
r Black
ve
M3001 2013 White
M3001 2013 Black
M4006 2017 White
M4006 2017 Black
ni

In this case. these two columns can be called as multivalued dependention on

U

BIKE_MODEL.

The representation of these dependancies is shown below.

1. BIKE_MODEL → →MANUF_YEAR
ity

2. BIKE_MODLE → → COLOR

This can be read as BIKE_MODLE multi determined MNUF_YEAR and BIKE_

MODLE malidefined COLOR.
m
)A
(c

Amity Directorate of Distance & Online Education

 186 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
In the above table, we can see Students Amit and Akash have interest in more than
one activity.

si
This is multivalued dependency because Course Discipline of a student are
independent of Activities, but are dependent on the student.

r
Therefore, multivalued dependency:
ve
ni

To above relation violates Fourth Normal Form in Normalization.

To correct it, divide the table into two separate tables and break Multivalued
Dependency:
U

<StudentCourse>

StudentName CourseDescipline
ity

Amit Mathematics
Amit Mathematics
Yuvraj Computers
Akash Literature
m

Akash Literature
Akash Literature
)A

<StudentActivities>

StudentName CourseDescipline
Amit Singing
Amit Dancing
(c

Yuvraj Cricket
Akash Dancing

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 187

Akash Cricket
Notes

e
Akash Singing

in
This breaks the multivalued dependency and now we have two functional
dependencies:

nl
O
If at least two impartial relation are stored in a single relation or we will say
multivalued dependency happens whilst the presence of at least one traces in a desk
induces the presence of at least one extraordinary columns in that equal table. Put
some other way, two attributes (or sections) in a table are independent of every other,

ty
however each rely on a third characteristic. A multivalued dependency in every case
wishes someplace around three attributes because it consists of no much less than
attributes that are dependent on a third. For a dependency a - > b, if for a unmarried

si
cost of a, more than one cost of b exists, then, at that factor, the desk might also have
multi-valued dependency. The desk must have as a minimum three attributes and b and
c should be impartial for a - >> b multivalued dependency.

For Eg, r
ve
Person Mobile Food_Likes
Mahesh 9893/9424 Burger / pizza
Resh 9191 Pizza
ni

Person->-> mobile,
Person ->-> food_likes

This is read as “person multidetermines mobile” and “person multidetermines food_

U

likes.”

Note that a functional dependency is a special case of multivalued dependency. In

a functional dependency X -> Y, every x determines exactly one y, never more than one.
ity

4.2.8 Fourth Normal Form (4NF):

Fourth nrml form (4nf) is a diploma of database normalization where there are
no non-trivial multivalued dependencies other than a candidate key. It expands on the
m

initial three everyday forms (1nf, 2nf and 3nf) and the boyce-codd regular shape (bcnf).
It expresses that, notwithstanding a database meeting the necessities of bcnf, it should
now not contain a couple of multivalued dependency .
)A

●● a relation can be in 4nf if’it’s miles in boyce codd ordinary shape and has no multi-
valued dependency.
●● for a dependency a → b, if for a unmarried value of a, more than one values of b
exists, then the relation can be a multi-valued dependency.
(c

Eg

Amity Directorate of Distance & Online Education

 188 Introduction to Data Base Management Systems

Student
Notes

e
STU_ID COURSE HOBBY
21 Computer Dancing

in
21 Math Singing
34 Chemistry Dancing
74 Biology Cricket

nl
59 Physics Hockey

Student_course

O
STU_ID COURSE
21 Computer
21 Math

ty
34 Chemistry
74 Biology
59 Physics

si
Student_hobby

STU_ID HOBBY
21 r Dancing
ve
21 Singing
34 Dancing
74 Cricket
59 Hockey
ni

Characterstics – a relation r is in 4nf if and most effective if the following conditions

are happy:
U

◌◌ It need to be inside the boyce-codd ordinary form (bcnf).

◌◌ The desk should now not have any multi-valued dependency. A desk with
a multivalued dependency violates the normalization preferred of fourth
ity

everyday shape (4nk) because it creates useless redundancies and can

contribute to inconsistent statistics. To convey this up to 4nf,’it’s miles
important to break this facts into tables.
Eg – Consider the database table of a class whaich has two relations R1 contains
student ID(SID) and student name (SNAME) and R2 contains course id(CID) and
m

course name (CNAME).

Table – R1(SID, SNAME)

)A

SID SNAME
S1 A
S2 B
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 189

Table – R2(CID, CNAME)

Notes

e
CID CNAME
C1 C

in
C2 D

When there cross product is done it resulted in multivalued dependencies:

nl
Table – R1 X R2

SID SNAME CID CNAME

O
S1 A C1 C
S1 A C2 D
S2 B C1 C

ty
S2 B C2 D

Multivalued dependencies (MVD) are:

si
SID->->CID; SID->->CNAME; SNAME->->CNAME

4.2.9 Join Dependency

r
Be part of dependency – join decomposition is a in addition generalization of
ve
multivalued dependencies. On the off chance that the be a part of of r1 and r2 over c is
equal to relation r then we can say that a be part of reliance (jd) exists, in which r1 and
r2 are the decomposition r1(a, b, c) and r2(c, d) of a given members of the family r (a, b,
c, d). As an alternative, r1 and r2 are a lossless decomposition of r. A jd r1, r2, … , rn is
ni

said to keep over a relation r if r1, r2, … .., rn is a lossless-be a part of decomposition.
The *(a, b, c, d), (c, d) may be a jd of r if the be a part of of be a part’of’s attribute is
equal to the relation r. Here, *(r1, r2, r3) is applied to show that relation r1, r2, r3, and so
U

on are a jd of r.

Suppose r is a relation schema r1, r2, r3…….. Rn be the decomposition of r. R( r )

is said to satisfy be a part of dependency if and most effective if
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 190 Introduction to Data Base Management Systems

4.2.10 Fifth Normal Form / Projected Normal Form (5NF)

Notes

e
A relation r is in 5nf if and best if each join dependency in r is usually
recommended by way of the candidate keys of r. A relation disintegrated into two

in
relations need to have loss-much less join property, which ensures that no deceptive or
extra tuples are produced, when relations are delivered collectively via a natural be part
of.

nl
●● a relation is in 5nf if’it’s far in 4nf and now not includes any be a part of
dependency and becoming a member of should be lossless.
●● 5nf is glad when all of the tables are damaged into as many tables as possible a

O
good way to avoid redundancy.

ty
r si
ve
Inside the above table, j takes each computer and math elegance for semester
1 however he would not take math magnificence for semester 2. In this situation,
ni

aggregate of most of these fields required to pick out a legitimate facts.

Suppose we add a brand new semester as semester three however do now not
understand about the problem and who might be taking that concern so we depart
U

lecturer and challenge as null. However all 3 columns collectively acts as a number one
key, so we ’an’t go away different two columns blank.

So as to make the above table into 5nf, we will decompose it into three members of
ity

the family p1, p2 & p3:

m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 191

Notes

e
in
nl
O
ty
si
Properties: A relation R is in 5NF if and only if it stisfies following condition:

1. R should be already in 4NF.

r
2. It cannot be further non loss decomposed (join dependency)
ve
Example: Consider the above schema, with a case as ‘If a company makes a
product and an agent is an agent for that company: then he always sells that product for
the company’. Under these circumstances. the ACP table is shown as:

Table-ACP
ni

Agent Company Product

A1 PQR Nut
U

A1 PQR Bolt
A1 XYZ Nut
A1 XYZ Bolt
A2 PQR Nut
ity

Table-R1

Agent Company
m

A1 PQR
A1 XYZ
A2 PQR
)A

Table-R2

Agent Company
A1 NUt
(c

A1 Bolt
A2 Nut

Amity Directorate of Distance & Online Education

 192 Introduction to Data Base Management Systems

Table-R3
Notes

e
Agent Company
PQR Nut

in
PQR Bolt
XYZ Nut
XYZ Bolt

nl
Join Dependency:
Join decomposition is a further generalization of multivalued dependencies. On the

O
off chance that the be part of of r1 and r2 over c is identical to relation r then we are
able to say that a join reliance (jd) exists, in which r1 and r2 are the decomposition r1(a,
b, c) and r2(c, d) of a given relations r (a, b, c, d). Alternatively, r1 and r2 are a lossless

ty
decomposition of r. A jd r1, r2, … , rn is stated to hold over a relation r if r1, r2, … .., rn
is a lossless-be part of decomposition. The *(a, b, c, d), (c, d) might be a jd of r if the be
a part of of be a part’of’s attribute is equal to the relation r. Here, *(r1, r2, r3) is utilized
to show that relation r1, r2, r3, and so on are a jd of r.

si
Allow r is a relation schema r1, r2, r3…….. Rn be the decomposition of r. R( r ) is
stated to satisfy join dependency if and best if

r
ve
If a table may be recreated by joining a couple of tables and each of this desk have
a subset of the attributes of the table, then the table is in be a part of dependency.’It’s
miles a generalization of multivalued dependency
ni

be part of dependency can be related to 5nf, in which a relation is in 5nf, only if it is

already in 4nf and it ’an’t be decomposed in addition.
U

Example
<Employee>

EmpName EmpSkills EmpJob (Assigned

ity

work)
Tom Notworking EJ001
Jarry Web Development EJ002
Katie Programmng EJ002
m

The above table can be decomposed into the following three tables; therefore it is
not in 5NF:
)A

<Employee Skills>

EmpName EmpSkills
Tom Notworking
Jarry Web Development
(c

Katie Programmng

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 193

Notes

e
in
nl
O
ty
r si
ve
The above members of the family have be part of dependency, so t’ey’re now not
in 5nf. That would suggest that a be part of relation of the above three members of the
family is identical to our original relation <employee>.

Join dependency is a constraint that is much like practical dependency or

ni

multivalued dependency.’It’s far glad if and only if the relation involved is the be part
of of a positive range of projections. Such sort of constraint is known as be a part of
dependency. Al’ow’s ’on’t forget a unique class of join dependencies which assist to
U

capture statistics dependencies found in a hierarchical records shape..

Eg1:
ity

Example 1
A relation R satisfies join dependency if R is equal to the join of R1,I.....Rn where Ri
is a subset of the set of attributes of R.
m
)A
(c

here,
Amity Directorate of Distance & Online Education
 194 Introduction to Data Base Management Systems

dept ->-> problem

Notes

e
dept->-> call

the above relation is in 4nf. Anomalies can arise in relation in 4nf if the number

in
one key has 3 or extra fields. The primary key is (dept, problem, call). Now and again
decomposition of a relation into smaller family members does not remove redundancy.
In such instances it is able to be possible to decompose the relation in 3 or greater

nl
relations the usage of 5nf. The above relation says that dept offers many optionally
available topics which might be taken by means of a ramification of students. Students
have the opinion to select subjects. Consequently all three fields are needed to

O
constitute the facts. The above relation does now not display non-trivial mvds for the
reason that attributes concern and call are dependent; they may be related to every
other (a fd situation->call exists). The relati’n can’t be decomposed in relations (dept,
challenge) and (dept,sname). Therefore the relation may be decomposed into following

ty
three rela–ions −

r1(dept, issue)

si
r2(dept, call) and

r3(situation, name) and it is able to be shown that decomposition is lossless.

r Dept
R1
Subject
ve
CSE C
CSE Java
IT C
ni

R2
Dept Name
CSE Ammu
U

CSE Amar
IT Bhanu

R3
ity

Subject Name
C Ammu
Java Amar
C Bhanu
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 195

Case Study
Notes

e
in
nl
O
ty
The above hierarchical company informs regarding ward and sufferers currently
admitted to a ward depend most effective at the health facility but no longer the facilities
present in that health center. Because hospitals have a couple of wards, practical
dependencie’ aren’t good enough to describe the records dependency among hospitals

si
and wards or facilities. In this situation, multivalued dependencies,

health facility->-> ward or

r
clinic ->-> centers maintain. The usage of first order hierarchical decomposition
ve
could enable us to represent facts dependencies present in hierarchical records shape
in a more natural way. For this reason we will keep health facility database as lossless
be part of of the subse–uent −

hospital_facility(health center, facilities),

ni

hospital_ward(hospital, ward, affected person, court cases, treatment, doctor)

Check your understanding:

U

A. Fill in the blanks:

1. Normalization is a process of ________ a set of relations with anomalies to produce
ity

smaller and well-structured relations that contain minimum or no ________.

2. ________ first developed the process of normalization.
3. A relation is said to be in 1NF if the values in the domain of each attribute of the
relation are ________.
m

4. A relation R is said to be in 2NF if it is in ________ and every non-prime key attributes

of R is ________ on each relation key of R.
5. ________ is the formal process for deciding which attributes should be grouped
)A

together.

B. State true or false:

1. For a relation schema R(A, B, C, D), if AB → C and C → D, then AB is a key.
(c

2. If a relation is in 3NF, then it is also in BCNF.

3. BCNF decomposition of a relation R(A, B, C, D) with functional dependencies A → B

Amity Directorate of Distance & Online Education

 196 Introduction to Data Base Management Systems

and C → D leads to three relations, R1(A, B), R2(C, D), and R3(C, A).
Notes

e
4. From A → B, we can derive AC → BC, which can further lead to A → BC.
5. For a relation R with attribute ABCDE and set of functional dependencies F = {A →

in
B, BC → D, D → E}, the closure for attribute combination {AC} is ABCDE.

Activity

nl
1. Prepare a power point presentation on the difference between First Normal Form
and Second Normal Form based on normalization.

O
Summary
●● database normalization is a manner of organizing the facts inside the database.
Normalization is a scientific approach of decomposing tables to cast off records
redundancy(repetition) and unwanted characteristics like insertion, update and

ty
deletion anomalies’
●● it’s far a multi-step process that places records into tabular form, doing away with
duplicated facts from the relation tables.

si
●● Normalization is used for mainly features,
●● getting rid of redundant(useless) facts. Making sure information dependencies
r
make sense i. E information is logically stored.
ve
●● if a desk is not nicely normalized and have data redundancy t’en it’s going to
now not exceptional consume up extra reminiscence place but can even make it
difficult to deal with and replace the database, without going through records loss.
●● Insertion, updation and deletion anomalies are very common if databa’e isn’t
ni

normalized.
●● normalization recommendations are divided into the following regular bureaucracy:
U

First nf
2nd nf
Third nf
ity

Bcnf
Fourth ordinary form
Fifth nf
m

6th nf

word list:
)A

●● normalization is a database format approach that reduces records redundancy and

eliminates undesirable traits like insertion, replace and deletion anomalies.
●● listing of regular office work
1nf (first regular shape)
(c

2nf (2d ordinary shape)

3nf (1/3 ordinary shape)

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 197

bcnf (boyce-codd everyday form)

Notes

e
4nf (fourth everyday shape)
5nf (5th ordinary shape)

in
6nf (6th regular shape)
●● a key in rectangular is a price used to choose out statistics in a table uniquely.

nl
An square secret is a unmarried column or aggregate of multiple columns used to
uniquely perceive rows or tuples within the table. Rectangular key is used to select
out reproduction information, and it moreover allows set up a relationship among
multiple tables in the database.

O
●● columns in a table tha’ aren’t used to pick out out a report uniquely are referred to
as non-key columns.

ty
●● a number one is a unmarried column price used to identify a database document
uniquely.
●● database designing is essential to the successful implementation of a database
management machine that meets the information requirements of an business

si
enterprise tool.
●● normalization in DBMS is a technique which facilitates produce database systems
r
which can be cost-effective and feature higher protection models.
ve
●● useful dependencies are a very vital aspect of the normalize records way
●● maximum database systems are normalized database as plenty because the zero.
●● a number one key uniquely identifies are report in a desk and can not be null
ni

●● a overseas key enables connect desk and references a primary key

Check your understanding(answers):

U

A. Fill in the blanks:

1. grouping , relations
2. Codd
ity

3. atomic
4. 1NF , fuly dependent
5. Normalisation
m

B. State True or False:

1. True
)A

2. False
3. True
4. False
5. True
(c

Amity Directorate of Distance & Online Education

 198 Introduction to Data Base Management Systems

Check your understanding:

Notes

e
A. Fill In The Blanks:

in
1. The E-R model that is supported with the additional semantic concepts is called the
________.
2. The relationship between a superclass and a subclasses is ________.

nl
3. A total participation specifies that every member in the ________ must participate.
4. A partial participation constraint specifies that a member of a ________ need not
belong to any of its _____. Attribute inheritance is the property by which attribute

O
inherit values of all attributes of the ________.

B State True Or False:

ty
1. Subclasses are the sub-grouping of occurrences of entities in an entity type that
shares common attributes or relationships distinct from other sub-groupings.
2. In case of supertype, objects in one set are grouped or subdivided into one or more

si
classes in many systems.
3. The relationship between a superclass and a subclasses is a one-to-many (1:N)
relationship.
4. r
Attribute inheritance is the property by which supertype entities inherit values of all
ve
attributes of the subty 5. Each member of the subclass is also a member of the
superclass.

Exercising 1 :
ni

●● draw the er diagram for a small database for a book place. The database will
keep statistics about books for sale. Every book has an isbn, identify, fee and brief
description. Every e book is published by way of a writer in a sure publishing yr.
U

For every publisher, the database continues the call, address and make contact
with number. Each book is written by using one or more authors. For each writer,
the database keeps his/her identity, name and a short advent. Each e-book is
saved in exactly one warehouse with a specific quantity. For every warehouse,
ity

the database continues the warehouse call, the region and the cellphone variety.
Every book has one or greater sellers, which can be either agencies (company
carriers) or people (person providers). For every enterprise, the database
maintains a name of the enterprise, its cope with, its phone numbers (there can
m

be more than one phone quantity, each with more than a few and an outline) and
its contact character. For every man or woman dealer, the database continues
a name, a smartphone range and an email cope with. A touch man or woman
whose organisation sells a e bo’k can’t be promoting the identical book as an
)A

person dealer at the same time (he/she may promote different books as an man or
woman seller).

Word list:
(c

●● supertypes and subtypes

●● supe–type - an entity kind that relates to one or extra subtypes.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 199

●● su–type - a subgroup of entities with specific attributes.

Notes

e
●● inheri–ance - the idea that subtype entities inherit the values of all supertype
attributes. Note: subtype times also are categorized as supertype times.

in
Generalization & specialization
●● generaliz–tion - the procedure of defining a fashionable entity kind from a set of
specialised entity kinds.

nl
●● specializ–tion - the opposite of generalization, since it defines subtypes of the
supertype and determines the connection among the 2. Constraints
●● disjointness constr–ints - you will need to determine whether or not a supertype

O
example can also concurrently be a member of two or extra subtypes. The disjoint
rule forces subclasses to have disjoint units of entities. The overlap rule forces a
subclass (additionally referred to as a supertype instance) to have overlapping

ty
units of entities.
●● completeness constr–ints - determine whether or not a supertype example must
additionally be a member of as a minimum one subtype. The overall specialization

si
rule needs that every entity in the superclass belong to a few subclass. Simply
as with a everyday erd, overall specialization is symbolized with a double line
connection among entities. The partial specialization rule lets in an entity to

r
not belong to any of the subclass’s. It’s far represented with a unmarried line
connection. Subtype discriminators
ve
●● a subtype discriminator is an attribute of the supertype that indicates an ’ntity’s
subtype. The att’ibute’s values are what determine the target subtype.
●● disjoint sub–ypes - easy attributes that need to have alternative values to indicate
ni

any viable subtypes.

●● overlapping sub–ypes - composite attributes whose subparts pertain to diverse
subtypes. Each subpart ooleanoolean price that indicates whether or not the
U

instance belongs to the related subtype. Erd

●● conceptual records models set up a vast view of what should be included in the
model set.
ity

●● entities are items or principles that represent important statistics. Entities are
generally nouns which includes product, customer, region, or merchandising.
●● there are 3 forms of entities generally utilized in entity courting diagrams.
●● strong entity,
m

●● susceptible entity and associative entity

●● inside entity-relationship diagrams, relationships are used to record the interaction
)A

between entities.
●● forms of relationships: weak and robust
●● relationships are normally verbs which include assign, accomplice, or tune and
provide beneficial data that could not be discerned with simply the entity sorts.
(c

●● erd attributes are characteristics of the entity that help customers to better
understand the database. Attributes are covered to include information of the

Amity Directorate of Distance & Online Education

 200 Introduction to Data Base Management Systems

numerous entities that are highlighted in a conceptual er diagram.

Notes

e
●● eer creates a layout extra accurate to database schemas.
●● it reflects the records houses and constraints greater precisely.

in
●● it consists of all modeling principles of the er model.
●● diagrammatic technique enables for showing the eer schema.

nl
●● it includes the concept of specialization and generalization.
●● it is used to symbolize a collection of objects this is union of objects of various of
different entity sorts.

O
●● from time to time important to represent a unmarried superclass/subclass courting
with multiple superclass, in which the superclasses represent one of a kind entity
types. In this case, the subclass will represent a group of objects that may be a

ty
subset of the union of distinct entity kinds; we call one of these subclass a union
kind or a class.

Check your understanding(answers):

si
FILL IN THE BLANKS:
1. Enhanced ER Model
2. r
Property Inheritance
ve
3. Entity set, relationship set
4. Entity set, relationship set
5. Parent product class
ni

STATE TRUE OR FALSE:

1. True
U

2. True
3. True
4. True
ity

5. True

Check your understanding:

A. Fill in the blanks:
m

1. The arrow notation ‘→’ in FD is read as ________.

2. The term full functional dependency (FFD) is used to indicate the ________ set of
)A

attributes .
3. A functional dependency in the set is redundant if it can be derived from the other
________ in the set.
4. A functional decomposition is the process of ________ the functions .
(c

5. A functional dependency establishes a relationship between _____sets of attributes..

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 201

B. State true or false:

Notes

e
1. Armstrong’s axioms allow us to generate all F+ for any given F
2. Functional dependencies are transitive

in
3. A functional dependency is a relationship between an attr“b”te “Y” and a determinant
(1 or more other attributes) such that for a given value of a determinant the value of
the attribute Y is uniquely defined.

nl
4. Transitivity of functional dependencies means that if a →b and a→c then c→b.
5. Functional dependencies are reflexive, i.e., if a →b then b →a.

O
Summary
●● a purposeful dependency among columns, x and y, method that for any two
information r1 and r2 inside the table, if field x of record r1 consists of value x and

ty
area x of report r2 consists of the same price x, then if discipline y of file r1 carries
the fee y, then field y of record r2 have to incorporate the value y. We can say that
attribute y is determined by attribute x.

si
●● purposeful dependencies may additionally exist among a couple of source
columns.
●●
r
a practical dependency establishes a courting between sets of attributes.
ve
●● a transitive dependency exists if you have the following useful dependency
pattern: a→b and b→c; consequently a→c
●● a transitive dependency therefore exists most effective when the determinant th’t
isn’t the primary k’y isn’t a candidate key for the relation.
ni

●● a 2nd everyday shape relation that has no transitive dependencies is, of path,
routinely in 1/3 everyday form. Glossary:
◌◌ the characteristic set at the left side of the arrow, x is called determinant
U

◌◌ the attribute set on the right facet, y is known as the based.

◌◌ forms of fd:
trivial functional dependency
ity

non-trivial functional dependency

multivalued functional dependency
transitive useful dependency
m

●● in trivial functional dependency, a structured is constantly a subset of the

determinant.
)A

●● in non-trivial functional dependency, the established is exactly not a subset of the

determinant.
●● in multivalued practical dependency, entities of the structured set are not
dependent on every different.
(c

●● in transitive practical dependency, structured is circuitously depending on

determinant.

Amity Directorate of Distance & Online Education

 202 Introduction to Data Base Management Systems

Further Readings
Notes

e
1. Raghu Ramakrishnan and Johannes Gehrke. (2003). Database Management
Systems. Third Edition. McGraw Hill

in
2. Ramez Elmasri and Shamkant B. Navathe. (2016). Fundamentals of Database
Systems. Seventh Edition. Pearson Education, Inc.
3. Carlos Coronel and Steven Morris. (2018). Database Systems: Design,

nl
Implementation and Management. Thirteenth Edition. Course Technology, Inc.

Check your understanding (answers):

O
A. FILL IN THE BLANKS:
1. Functional Dependency
2. Relationship

ty
3. Dependencies
4. analyzing the links and relationships between all the components that create a

si
functional relationship
5. Two
B. STATE TRUE OR FALSE:
1. True
r
ve
2. False
3. True
4. True
ni

5. True
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 203

Module -5: Data Recovery & Protection

Notes

e
Structure:

in
5.1 Concurrency Control Techniques
5.1.1 Introduction

nl
5.1.2 Transaction and System Concept
5.1.3 Properties of Transactions

O
5.1.4 Locking Concepts and Types of Locking
5.1.5 Two Phase Locking Techniques
5.1.6 Concurrency Control Based on Time Stamping

ty
5.1.7 Multivesion Concurrency Control
5.1.8 Granularity of Data Items and Multiple Granularity Locking
5.1.9 Define Schedule and Serializibility

si
5.1.10 Characterizing Schedules Based on Serializability
5.2 Database Recovery and Security
5.2.1 Recovery Concepts r
ve
5.2.2 Recovery Techniques Based on Deferred Update
5.2.3 Recovery Techniques Based on Immediate Update
5.2.4 Shadow Paging
ni

5.2.5 System Recovery

5.2.6 Media Recovery
U

5.2.7 Database Security

ity
m
)A
(c

Amity Directorate of Distance & Online Education

 204 Introduction to Data Base Management Systems

Unit -5.1: Concurrency Control Techniques

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Learn about Transaction and System Concept

nl
●● Know about Properties of Transactions
●● Understand Locking Concepts and Types of Locking

O
●● Learn about Two Phase Locking Techniques
●● Understand about Concurrency Control Based on Time Stamping
●● Know about Multivesion Concurrency Control

ty
●● Understand Granularity of Data Items and Multiple Granularity Locking
●● Learn the definition of Schedule and Serializibility
●● Understand the Characterizing Schedules Based on Serializability

si
5.1.1 Introduction

Transaction r
ve
●● the transaction is a set of logically related operation. It consists of a group of
obligations.
●● a transaction is an motion or series of actions. It’s far performed by way of a
ni

single consumer to perform operations for gaining access to the contents of the
database.
Eg: suppose an emp of financial institution transfers rs 800 from x’s account to y’s
U

account. This small transaction consists of several low-stage responsibilities:

x’s account
1. Open_account(x)
ity

2. Old_balance = x. Balance
3. New_balance = old_balance - 800
4. X. Balance = new_balance
m

5. Close_account(x)

y’s account
)A

1. Open_account(y)
2. Old_balance = y. Stability
3. New_balance = old_balance + 800
(c

4. Y. Stability = new_balance
5. Close_account(y)

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 205

operations of transaction:
Notes

e
following are the main operations of transaction:

Study(x): study operation is used to read the price of x from the database

in
and stores it in a buffer in primary reminiscence. Write(x): write operation is used to
write down the cost again to the database from the buffer. Let’s take an eg to debit
transaction from an account which includes following operations:

nl
◌◌ R(x);
◌◌ X = x - 500;

O
◌◌ W(x);
let’s assume the cost of x earlier than starting of the transaction is 4000.

◌◌ The primary operation reads x’s value from database and shops it in a buffer.

ty
◌◌ The second operation will decrease the fee of x through 500. So buffer will
contain 3500.
◌◌ The third operation will write the buffer’s fee to the database. So x’s final price

si
could be 3500. However it can be possible that due to the failure of hardware,
software or energy, etc. That transaction may additionally fail before finished
all of the operations in the set. For eg: if in the above transaction, the debit
r
transaction fails after executing operation 2 then x’s value will continue to be
4000 in the database which is not applicable via the financial institution. To
ve
solve this problem, we’ve vital operations:
Dedicate: it’s miles used to keep the work executed permanently.

Rollback: it’s far used to undo the work performed.

ni

Concurrency control : concurrency control in database management device is

a technique of dealing with simultaneous operations without conflicting with every
other. It guarantees that database transactions are finished simultaneously and
U

appropriately to provide accurate consequences without violating facts integrity of the

respective database. Concurrent get admission to is quite easy if all users are just
reading statistics. There may be no manner they can intervene with each other. Even
ity

though for any sensible database, it would have a mix of study and write operations
and consequently the concurrency is a mission. DBMS concurrency manage is used
to deal with such conflicts, which in general occur with a multi-user device. Therefore,
concurrency manage is the most critical detail for correct functioning of a database
management system wherein or greater database transactions are carried out
m

concurrently, which require get entry to the same records.

Records restoration : database systems, like some other pc device, are problem
to screw ups but the information saved in it need to be available as and when required.
)A

While a database fails it should own the facilities for instant restoration. It need to
additionally have atomicity i. E. Both transactions are finished efficaciously and
dedicated (the impact is recorded permanently within the database) or the transaction
have to have no effect at the database. There are both automated and non-automatic
methods for both, backing up of facts and healing from any failure conditions. The
(c

techniques used to recover the misplaced statistics because of machine crash,

transaction errors, viruses, catastrophic failure, wrong cmds execution and so on. Are

Amity Directorate of Distance & Online Education

 206 Introduction to Data Base Management Systems

database restoration techniques. So as to prevent facts loss healing strategies based

Notes

e
on deferred replace and immediate replace or backing up facts can be used.

5.1.2 Transaction and System Concept

in
Alternate is a valid unit of labor that addresses true activities of any association
or a mission at the same time as concurrency manipulate is the administration of

nl
concurrent exchange execution. Alternate processing frameworks execute database
transactions with sizable databases and plenty of concurrent users, for eg, rail direction
and air reservations frameworks, banking framework, mastercard processing, economic
trade gazing, shop stock and checkouts, and so forth.

O
Transaction : a trade is an intelligent unit of labor of database processing that
includes at least one database access operations. A change can be characterized as
an activity or collection of activities this is finished by a solitary user or utility software

ty
to perform operations for purchasing to the substance of the database. The operations
can comprise healing, (study), addition (write), cancellation and alteration. An exchange
have to be both finished or cut off. It can both be hooked up inner an utility application
or may be determined intuitively through a high-stage query language like sq.. Its

si
execution protects the consistency of the database. Every trade need to get to shared
facts with out meddling with one-of-a-kind transactions and at something point an trade
efficiently finishes its execution; its effect need to be everlasting.
r
This fundamental abstraction frees the database application programmer from the
ve
following issues :

◌◌ inconsistencies resulting from conflicting updates from concurrent customers.

◌◌ partly finished transactions in the event of systems failure.
ni

◌◌ person-directed undoing of transactions. A transaction is a sequence of read

and write moves which can be grouped collectively to shape a database get
entry to. A transaction may also consist of a very easy choose operation to
generate a list of desk contents, or it can consist of a chain of interrelated
U

replace cmd sequences.

Operations Description
ity

Retrive To retrieve data stored in a databse

Insert To store new data in database
Delete To delete existing data from database
Update To modify existing data in database
m

Commit To save the work done permanently

Rollback To undo the work done

State Description
)A

Active state A transaction goes into an active state immediately after it starts
execution, where it can issue READ and WRITE operations. A transaction
may be aborted when the transaction itself delects an error during
execution which it cannot recover from, for example, a transaction trying
(c

to debit loan amount of an employee from his insufficient gross salary.

Atransaction may also be aborted before it has been committed due to
system failure or any other circumstances beyond control.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 207

Partially When the transaction ends, it moves to the partially committed state.
Notes

e
Committed When the last state is reached.
To this point, some recovery protocols need to nesure that a system

in
failure will not result in an inability to record the changes of the transaction
permanently. Once this check is successful, the transaction is said to
have reached its commit point and enless the committed state.

nl
Aborted When the normal execution can no longer be per formed.
Failed or aborted transactions may be restarted later, either automatically
or after being resubmitted by the user as new transactions.
Committed After successful completion of transaction. A transaction is said to be

O
in a committed state if it has partially committed and it can be ensured
that it will never be aborted.

Transaction that changes the contents of the database should alter the database

ty
starting with one consistent state then onto the next. A consistent database state is one
in which all data integrity constraints are fulfilled.

To guarantee database consistency, each transaction should start with the

si
database in a known consistent state.

Transaction Execution and Problems :

r
A transaction which effectively finishes its execution is said to have been
ve
committed. In any case, the transaction is aborted.

In this way, if a committed transaction plays out any update operation on the
database, its impact should be pondered the database regardless of whether there is a
ni

failure.
U
ity
m

State Description
)A

Active state A transaction goes into an active state immediately after it starts
execution, where it can issue READ and WRITE operations. A transaction
may be aborted when the transaction itself delects an error during
execution which it cannot recover from, for example, a transaction trying
to debit loan amount of an employee from his insufficient gross salary.
(c

Atransaction may also be aborted before it has been committed due to

system failure or any other circumstances beyond control.

Amity Directorate of Distance & Online Education

 208 Introduction to Data Base Management Systems

Partially When the transaction ends, it moves to the partially committed state.
Notes

e
Committed When the last state is reached.
To this point, some recovery protocols need to nesure that a system

in
failure will not result in an inability to record the changes of the transaction
permanently. Once this check is successful, the transaction is said to
have reached its commit point and enless the committed state.

nl
Aborted When the normal execution can no longer be per formed.
Failed or aborted transactions may be restarted later, either automatically
or after being resubmitted by the user as new transactions.
Committed After successful completion of transaction. A transaction is said to be

O
in a committed state if it has partially committed and it can be ensured
that it will never be aborted.

A transaction can be in one of the following states:

ty
Transaction log (or journal):
to support transaction processing, DBMSs hold a transaction document of every

si
exchange made to the database right into a log (additionally referred to as diary).
Log is a file of all transactions and the corresponding adjustments to the database.
The statistics saved inside the log is used by the DBMS for a healing necessity set
r
off through a rollback assertion, that is packages unusual end, a gadget (power
ve
or organisation) failure, or disk crash. Some relational database the executives
structures (rDBMSs) use the transaction log to get better a database forward to an at
gift constant state. The DBMS therefore update the transaction log even as executing
transactions that adjust the database. The transaction log shops earlier after which
ni

in a while facts about the database and any of the tables, rows and characteristic
values that participated within the transaction. The start and the closure (dedicate) of
the transaction also are recorded within the transaction log. For every transaction, the
accompanying information is recorded on the log:
U

◌◌ a start-of-transaction marker.
◌◌ the transaction identifier which can include who and in which statistics.
ity

◌◌ the file identifiers which include the identifiers for the document occurrences.
◌◌ the operation(s) accomplished at the facts (for eg, insert, erase, adjust).
◌◌ the previous price(s) of the modified records. This information is needed for
undoing the modifications made by using a partially completed transaction.
m

It’s far referred to as the undo log. Where the amendment made by the
transaction is the insertion of another file, the previous values can be
assumed to be null.
)A

◌◌ the refreshed cost(s) of the modified file(s). This information is needed for
making sure that the modifications made by way of a dedicated transaction
are truth be instructed reflected in the database and can be used to redo
these modifications. This facts is called the redo a part of the log. In case the
amendment made by way of the transaction is the deletion of a record, the
(c

refreshed values may be assumed to be null.

◌◌ a devote transaction marker if the transaction is committed, in any other case
an abort or rollback transaction marker. The log consists before any updates
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 209

are made to the database. This is referred to as write-beforehand log strategy.

Notes

e
On this strategy, a transaction isn’t always authorized to adjust the physical
database until the undo portion of the log is composed to stable database.
In case of a system failure, the DBMS examines the transaction log for all

in
uncommitted or poor transactions and restores (rollback) the database to
its previous country based totally on the statistics in the transaction log.
At the point while the healing procedure is finished, the DBMS writes in the

nl
transaction log all dedicated transactions that had been now not physically
kept in contact with the physical database before the failure befell. If a rollback
is issued before the stop of a transaction, the DBMS restores the database

O
only for that particular transaction, as opposed to for all transactions,
to preserve up with the durability of the previous transactions. All in all,
committed transactions aren’t rolled lower back.

ty
5.1.3 Properties of Transactions

Transaction properties:

si
a transaction have to have the following four residences, referred to as acid
residences (also called acidity of a transaction), to make sure that a database stays
solid country after the transaction is accomplished:

1. Atomicity. r
ve
2. Consistency.
3. Isolation.
4. Sturdiness.
ni

Atomicity:
the atomicity assets of a transaction requires that everyone operations of a
U

transaction be finished, if now not, the transaction is aborted. As such, a transaction

is treated as single, character logical unit of labor. As a result, a transaction have to
execute and finish each activity in its logic earlier than it commits its modifications as
said earlier than, the transaction is considered as one interest no matter the reality
ity

that there are multiple examine and writes. As a consequence, transaction completes
or fails as one unit. The atomicity belongings of transaction is ensured by way of the
transaction restoration subsystem of a DBMS. In case of a device crash within the midst
of transaction execution, the recuperation techniques undo any consequences of the
m

transaction on the database.

Consistency:
)A

database consistency is the assets that every transaction sees a constant

database example. All in all, execution of a transaction should leave a database in
either its previous solid kingdom or every other strong kingdom that reflects the brand
new adjustments (updates) made with the aid of the transaction. If the transaction fails,
the database need to be gotten returned to the state it become in prior to the execution
(c

of the failed transaction. If the transaction commits, the database need to reflect the
new adjustments. Accordingly, all resources are continually in a regular nation. The
maintenance of consistency is via and massive the obligation of the programmers

Amity Directorate of Distance & Online Education

 210 Introduction to Data Base Management Systems

who compose the database programs or of the DBMS module that enforces integrity
Notes

e
constraints. A database program have to be written such that guarantees that, if the
database is in a regular state prior to executing the transaction, it will likely be in a
steady state after the whole execution of the transaction, assuming that no impedance

in
with one-of-a-kind transactions happen. As such, a transaction need to transform the
database beginning with one constant state then onto the next consistent country.

nl
Isolation:
isolation belongings of a transaction method that the records used for the
duration of the execution of a transaction cannot be utilized by a second transaction

O
till the primary one is completed. This belongings isolates transactions from one
another. In special phrases, if a transaction t1 is being accomplished and is the usage
of the facts element x, that statistics thing can’t be accessed by way of some other
transaction (t2… … .. Tn) until t1 ends. The isolation property is carried out by using the

ty
simultaneousness manipulate subsystem of the DBMS.

Sturdiness:

si
the durability property of transaction suggests the exhibition of the database’s
consistent country. It states that the modifications made by using a transaction are
permanent. They cannot be lost by way of either a device unhappiness or with the aid
r
of the inaccurate pastime of a broken transaction. On the factor while a transaction is
ve
completed, the database reaches a regular nation and that kingdom can’t be misplaced,
even in case of gadget’s disappointment. Sturdiness assets is the responsibility of the
recovery subsystem of the DBMS.
ni

5.1.4 Locking concepts and types of locking

Lock:
U

A lock is a variable related with a data item that portrays the situation with the
item as for potential operations that may be implemented to it. By way of and massive,
there’s one lock for each facts item within the database. Locks are applied as a method
for synchronizing the doorway by means of concurrent transactions to the database
ity

item.

Sorts of locks:
A few types of locks are applied in concurrency manage. To present locking ideas
m

continuously, we to start with speak about binary locks, that are primary however
restrictive as aren’t used by and with the aid of. We then, at that point, look at shared/
restrictive locks, which offer extra extensive locking capacities and are utilized in
)A

affordable database locking schemes.

Binary lock:
A binary lock may have two states or traits: lockend, unlocked. A particular lock is
related with each database object a. At the off threat that the really worth of the lock
(c

on an is 1, item a can not be gotten to via a database operation that demands the
object. At the off danger that the worth of the lock on an is zero, item may be gotten
to when cited. We allude to the modern-day worth of the lock related with item an as
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 211

lock (a). There are two operations, lock item and liberate object are utilized with binary
Notes

e
locking an alternate demands admittance to an object a by first giving a lock object
(a) operation. Within the occasion that lock (a) = 1, the alternate is forced to face by
way of. Inside the event that lock (a) = zero it’s far set to 1 (the exchange locks the

in
item) and the alternate is permitted to get to item a. At the factor whilst the exchange is
through using the item, it issues an unencumber object (a) operation, which sets lock
(a) to 0 (unlocks the object) with the goal that a might be gotten to by using exceptional

nl
transactions. Thus binary lock enforces mutual exclusion at the data item.

Guidelines of binary locks

O
At the off chance that the sincere binary locking plan portrayed right here is utilized,
each change ought to put up to the accompanying.

Guidelines:

ty
An change ought to supply the operation lock: item (a) previous any read_item (a)
or write, object operations are acted in t. 2. An change t must provide the operation
unlock_item (a) in spite of everything read_item (a) and write_item (a) operations are

si
completed in t. 3. An trade t won’t provide a lock _item (a) operation on the off danger
that it already holds the lock on item a. 4. An trade t won’t issue an liberate _item (a)
operation besides if it already holds the lock on object a. 5. The lock director module of
r
the DBMS can authorize these standards. Between the lock_item (a) and unlock_item
ve
(a) operations in exchange t, is said to maintain the lock on object a. At most one trade
can keep the lock on a specific item. Consequently no transactions can get to’ a similar
object simultaneously.
ni

Disservices of binary locks:

As mentioned before, binary locking plan is excessively restrictive for database
items, when you consider that at maximum one change can keep a lock on a given
U

item. Along these strains, binary locking framework can not be applied for common-
sense reason.

Share/specific (for examine/write) locks:

ity

Shared/extraordinary (or examine/write) locks. The primary binary locking scheme

is excessively restrictive for database gadgets due to the fact probably, one transaction
can maintain a lock on a given item. We ought to allow several transactions to get entry
to the same item x if all of them get right of entry to x for studying purposes because
m

it were. This is due to the fact study operations at the equal object via numerous
transactions are not clashing . Nonetheless, if a transaction is to jot down an item x, it
must have unique get admission to to x. For this reason, an change sort of lock called a
more than one-mode lock is used. On this scheme—referred to as shared/exceptional
)A

or study/write lock

There are 3 locking operations:

Read_lock(x), write_lock(x), and liberate(x). A lock associated with an object
(c

x, lock(x), now has three viable states: study-locked, write-locked, or unlocked. A

study-locked item is also called proportion-locked because exclusive transactions
are permitted to examine the object, whereas a write-locked item is referred to as
Amity Directorate of Distance & Online Education
 212 Introduction to Data Base Management Systems

exclusive-locked due to the fact a single transaction exclusively holds the lock on the
Notes

e
object. One method for executing the first operations on a study/write lock is to display
the quantity of transactions that maintain a shared (study) lock on an item within the
lock table. Every record in the lock table could have four fields: <data_item_name, lock,

in
no_of_reads, locking_transaction(s)>. Another time, to shop space, the device needs
to maintain up with lock records just for locked gadgets inside the lock table. The worth
(nation) of lock is both read-locked or write-locked, certainly coded (at the off chance

nl
that we count on no statistics are saved inside the lock table for unlocked gadgets).
If lock(x)=write-locked, the worth of locking_transaction(s) is a single transaction that
holds the special (write) lock on x. If lock(x)=examine-locked, the benefit of locking

O
transaction(s) is a listing of at least one transactions that preserve the shared (examine)
lock on x. The 3 operations read_lock(x), write_lock(x), and unlock(x) are defined
in figure 22. 2. As in the past, every one of the three locking operations need to be
taken into consideration indivisible; no interleaving have to be permitted once one of

ty
the operations is started till either the activity terminates with the aid of conceding the
lock or the transaction is installed a striking tight queue for the item. Whilst we use the
shared/exclusive locking scheme, the machine should implement the following policies:

si
A transaction T mus issue the operation write_lock(x)before any write_item(X)
operation is performed in T.

read_lock(X)
r
ve
If LOCK(X) =’unlocked’

then begin LOCK(X0) ←’read-locked’, no_of_reads(X)←1

end
ni

else if LOCK(x) = ‘read-locked’

thenno_of_read(X)←no_of_reads(X) +1 else begin

U

weit (unit LOCK(X) = ‘unlocked’

and the lock manager wakes up the transacitons); go to B

end; write_lock(X);
ity

If LOCK(X) =’unlocked’

then LOCK(X) ←’write-locke’else begin

wait (until LOCK(X) = ‘unlocked’

m

and the lock manger wakes up the transaction); go to B

unlock (X):
)A

If LOCK(X) = “write-locked”

the begin LOCK(X)←”unlocked”;

wakeup one of the waiting transactions, if any

(c

end

else It LOCK(X) =”read-locked”

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 213

then beign
Notes

e
no_of_reads(X) no_of_reads(X) -1;

If no_of_reads(X)=0

in
then begine LOCK(X) =”unlocked”; wakeup one of the waiting transactions, if any

end shared-exclusive)

nl
end;

A transaction t will not issue a write_lock(x) operation if it already holds a study

O
(shared) lock or write (special) lock on object x. This rule may also be secure, as we
talk shortly. A transaction t will not issue an unlock(x) operation except it already holds a
study (shared) lock or a write (different) lock on item x.

ty
Simplistic lock protocol:
This kind of lock-based totally protocols permits transactions to reap a lock on
each object before starting operation. Transactions may unlock the records item after

si
finishing the ‘write’ operation.

Pre-claiming locking:
r
pre-claiming lock protocol helps to assess operations and create a list of required
ve
information items that are needed to provoke an execution manner. Within the state
of affairs while all locks are granted, the transaction executes. After that, all locks
release when all of its operations are over. Conversion of locks. From time to time
it’s miles applicable to relax situations four and five in the preceding listing in order
ni

to permit lock conversion; that is, a transaction that already holds a lock on item x is
authorized beneath positive conditions to convert the lock from one locked country to
any other. For eg, it is feasible for a transaction t to problem a read_lock(x) and then
later to upgrade the lock by issuing a write_lock(x) operation. If t is the best transaction
U

preserving a study lock on x on the time it issues the write_lock(x) operation, the lock
can be upgraded; in any other case, the transaction ought to wait. It’s also possible for
a transaction t to difficulty a write_lock(x) after which later to downgrade the lock by way
ity

of issuing a read_lock(x) operation. Whilst upgrading and downgrading of locks is used,

the lock desk should include transaction identifiers within the document shape for each
lock (in the locking_transaction(s) discipline) to store the statistics on which transactions
hold locks on the item. The descriptions of the read_lock(x) and write_lock(x) operations
m

Locking operations:
There are 3 locking operations referred to as read_lock(a), write_lock(a) and free
up(a) represented as lock-s(a), lock-x(a), liberate(a) (right here, s suggests shared
)A

lock, x indicates exceptional lock)can be finished on a data object. A lock associated

with an object a, lock (a), now has 3 feasible states: “study-locked”, “write-locked,” or
“unlocked.” a read-locked object is likewise called proportion-locked object because
different transactions are allowed to study the object, whereas a write-locked item is
induced one-of-a-kind-locked, due to the fact a single transaction exclusively holds the
(c

lock at the object.

Amity Directorate of Distance & Online Education

 214 Introduction to Data Base Management Systems

Compatibility of Locks:
Notes

e
Think that there are a and b two one of a kind locking modes. If a transaction t1
requests a lock of mode on item q on which transaction t2 currently maintain a lock of

in
mode b. If transaction can be granted lock, in spite of the presence of the mode b lock,
then we say mode a is compatible with mode b. The sort of function is shown in a single
matrix as proven beneath:

nl
O
The graphs indicates that if two transactions simplest examine the equal records

ty
item they do not conf1ict, but if one transaction writes a information item and any
other both study or write the same facts object, then they struggle with each different.
A transaction requests a shared lock on facts object q through executing the lock-s(q)

si
GUIsdance. Further, an different lock is requested thru the lock- x(q) GUIsdance. A
records object q may be unlocked thru the unlock(q) education. To get right of entry to a
data item, transaction t1 must first lock that object. If the records item is already locked
r
by means of another transaction in an incompatible mode, the concurrency control
manager will not provide the lock till all incompatible locks held with the aid of other
ve
transactions have been launched. Hence, t1 is made to wait till all incompatible locks
held by way of different transactions have been launched

eg: as an example keep in mind the simplified banking gadget. Allow a and b be
ni

two debts which can be accessed by way of transactions t1 and t2. Transaction t1
transfers rs. 50 from account b to account a and is defined as:
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 215

Transaction T2 displays the total amount of money in accounts A and B that is, the
Notes

e
sum A+B and is defined as

in
nl
O
ty
Assume that the values of money owed a and bare rs. One hundred and rs. 200,
respectively. If those transactions are executed serially, either within the order t1 and
t2 or the order t2, t1 then transaction t2 will display the price rs. Three hundred. If

si
however, those transactions are achieved concurrently, as proven in time table 1. In this
example, transaction t2 shows rs. 250, which is incorrect. The motive for this mistake is
that the transaction ti unlocked facts item b too early, because of which t2 suggests an
inconsistent state. r
ve
ni
U
ity
m
)A

Solution of Inconsistency Problem

(c

Suppose now that unlocking is delayed to the end of the transaction. The
transaction T3 corresponds to T I with unlocking delayed and is defined as

Amity Directorate of Distance & Online Education

 216 Introduction to Data Base Management Systems

Notes

e
in
nl
O
ty
Transaction T4 corresponds to T2 with unlocking delayed, and is defined as

r si
ve
ni
U

You should verify that the sequence of reads and writes in schedule I which leads
to an incorrect total of Rs.250 being displayed, is no longer possible with T3 and T4 as
shown in Schedule 2
ity

T3 T4
lock -X(B)
read(B)
m

B;-B-60
Write(B)
lock-S(A)
)A

read(A)
lock - S(B)
wait...
wait...
lock-X(A)
(c

wait...

wait...

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 217

Schedule 2
Notes

e
Alas, the use of locking can cause an undesirable state of affairs. Remember the
partial schedule 2, t3 is conserving an special mode lock on b and t4 is soliciting for a

in
shared mode lock on b i. E. T4 is waiting fort3 to liberate b. Similarly, t4 is conserving
a shared mode lock on a and t3 is requesting an extraordinary mode lock on a,
consequently t3 is awaiting t4 to free up a.

nl
O
ty
Thus we have arrived at a state where neither of these transactions can ever

si
proceed with its normal execution. This situation is called deadlock.

r
ve
ni
U
ity
m
)A

As a result we can say that the answer of inconsistency leads to impasse hassle.
If we do not use locking or liberate information items as soon as possible after studying
(c

or writing them, we may additionally get inconsistent states. On the other hand, if we
do not unlock a information object before requesting a lock on another information item
deadlocks might also arise. There are approaches to keep away from impasse in some

Amity Directorate of Distance & Online Education

 218 Introduction to Data Base Management Systems

conditions. Deadlocks are really superior to inconsistent states, when you consider
Notes that they may be dealt with by using rolling lower back of transactions, wherein as

e
inconsistent states may cause real global troubles that cannot be dealt with with the aid
of the database sysem.

in
5.1.5 Two Phase Locking Techniques

nl
2-phase locking (2pl)
●● the two-phase locking protocol divides the execution segment of the transaction
into three parts.

O
●● in the first component, whilst the execution of the transaction begins, it seeks
permission for the lock it requires.
●● inside the 2nd component, the transaction acquires all of the locks. The 0. 33

ty
segment is began as soon as the transaction releases its first lock.
●● in the 1/3 segment, the transaction cannot call for any new locks. It most effective
releases the received locks.

r si
ve
There are two levels of 2pl:
ni

growing segment: inside the developing section, a new lock at the information
object can be obtained via the transaction, however none can be launched. Shrinking
phase: in the shrinking phase, present lock held by using the transaction may be
U

released, but no new locks may be obtained. Inside the under eg, if lock conversion is
allowed then the following section can appear:

◌◌ Upgrading of lock (from s(a) to x (a)) is authorized in growing segment

◌◌ Downgrading of lock (from x(a) to s(a)) must be accomplished in shrinking
ity

segment..
Eg:

T1 T2
m

0 LOCK-S(A)
1 LOCK-S(A)
2 LOCK-X(b)
)A

3 --------- ---------
4 UNLOCK(A)
5 LOCK-S(C)
6 UNLOCK(B)
7 UNLOCK(A)
(c

8 UNLOCK(C)
9 -------- -------

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 219

The subsequent manner suggests how unlocking and locking paintings with 2-pl.
Notes

e
Transaction t1:

◌◌ growing phase: from step 1-3

in
◌◌ shrinking section: from step five-7
◌◌ lock point: at three

nl
transaction t2:
◌◌ developing section: from step 2-6
◌◌ shrinking phase: from step 8-9

O
◌◌ lock point: at 6

Strict two-section locking (strict-2pl)

ty
◌◌ the primary segment of strict-2pl is similar to 2pl. Within the first phase, after
acquiring all the locks, the transaction keeps to execute generally.
◌◌ the most effective distinction among 2pl and strict 2pl is that strict-2pl does not

si
launch a lock after the use of it.
◌◌ strict-2pl waits until the entire transaction to dedicate, and then it releases all
of the locks at a time.
◌◌ r
strict-2pl protocol does no longer have shrinking section of lock launch. It does
ve
now not have cascading abort as 2pl does.

Centralized 2pl
in centralized 2 pl, a single web site is chargeable for lock management procedure.
ni

It has simplest one lock supervisor for the complete DBMS.

Primary replica 2pl

U

number one replica 2pl mechanism, many lock managers are distributed to one of
a kind web sites. After that, a particular lock supervisor is accountable for dealing with
the lock for a set of statistics objects. While the primary copy has been up to date, the
trade is propagated to the slaves.
ity

Dispensed 2pl
in this form of two-phase locking mechanism, lock managers are disbursed to all
web sites. They’re responsible for handling locks for data at that website online. If no
m

facts is replicated, it’s far equal to number one replica 2pl. Communique fees of allotted
2pl are pretty better than number one copy 2pl

Guaranteeing Serializability by Two-Phase Locking

)A

A transaction is stated to observe the 2-phase locking protocol if all locking

operations (read_lock, write_lock) precede the primary release operation inside the
transaction. One of these transaction can be divided into two levels: an increasing or
developing (first) phase, for the duration of which new locks on items may be obtained
(c

but none may be launched; and a shrinking (2nd) section, for the duration of which
present locks may be released but no new locks may be received. If lock conversion is
authorized, then upgrading of locks (from examine-locked to put in writing-locked) must
Amity Directorate of Distance & Online Education
 220 Introduction to Data Base Management Systems

be done at some stage in the increasing section, and downgrading of locks (from write-
Notes

e
locked to study-locked) ought to be carried out

in
nl
O
ty
r si
ve
ni
U

shrinking segment. Hence, a read_lock(x) operation that downgrades an already

held write lock on x can seem most effective inside the shrinking section.

Transactions t1 and t2 in figure 22. Three(a) do no longer follow the two-phase

ity

locking protocol due to the fact the write_lock(x) operation follows the release(y)
operation in t1, and further the write_lock(y) operation follows the free up(x) operation
in t2. If we put into effect -segment locking, the transactions can be rewritten as t1 and
t2 , as proven in figure 22. Four. Now, the time table proven in parent 22. 3(c) isn’t
always authorized for t1 and t2 (with their modified order of locking and unlocking
m

operations) below the regulations of locking defined in phase 22. 1. 1 due to the fact
t1 will trouble its write_lock(x) before it unlocks item y; consequently, while t2 issues
its read_lock(x), it’s miles forced to wait till t1 releases the lock via issuing an unlock
)A

(x) inside the agenda. It can be proved that, if every transaction in a time table follows
the 2-section locking protocol, the schedule is assured to be serializable, obviating the
want to check for serializability of schedules. The locking protocol, by enforcing -section
locking GUIsdelines, also enforces serializability.
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 221

Notes

e
in
nl
O
section locking may also restriction the quantity of concurrency that could arise in a
schedule due to the fact a transaction t might not be capable of launch an item x after it
is thru the usage of it if t must lock an extra item y later; or conversely, t should lock the

ty
additional item y earlier than it wishes it so that it is able to launch x. For this reason,
x ought to remain locked by means of t until all objects that the transaction desires to
examine or write have been locked; most effective then can x be released through t.

si
Meanwhile, every other transaction seeking to get admission to x can be compelled
to wait, even though t is done with x; conversely, if y is locked in advance than it is
wished, some other transaction looking for to access y is compelled to attend even
r
though t isn’t always using y but. That is the fee for making sure serializability of all
ve
schedules while not having to check the schedules themselves. Although the 2-phase
locking protocol guarantees serializability this is, each time table that is permitted is
serializable, it does not allow all feasible serializable schedules this is, someserializable
schedules can be prohibited by using the protocol.
ni

Basic, conservative, strict, and rigorous two-segment locking. There are some of
variations of two-phase locking (2pl). The technique simply described is referred to as
primary 2pl. A variation known as conservative 2pl (or static 2pl) requires a transaction
U

to lock all of the objects it accesses earlier than the transaction starts offevolved
execution, via predeclaring its examine-set and write-set. The examine-set of a
transaction is the set of all gadgets that the transaction reads, and the write-set is the
set of all objects that it writes. If any of the predeclared items wished cannot be locked,
ity

the transaction does not lock any item; as an alternative, it waits until all the objects are
to be had for locking. Conservative 2pl is a deadlock-unfastened protocol,but, it’s far
hard to use in practice because of the want to predeclare the examine-set and write-set,
which is not possible in many situations. The most famous variation of 2pl is strict 2pl,
m

which guarantees strict schedules

in this modification, a transaction t does not launch any of its unique (write) locks
till after it commits or aborts. Therefore, no different transaction can examine or write an
)A

object that is written by means of t except t has committed, main to a strict agenda for
recoverability. Strict 2pl is not deadlock-free. A greater restrictive variation of strict 2pl
is rigorous 2pl, which also guarantees strict schedules. In this change, a transaction t
does not release any of its locks (exceptional or shared) till after it commits or aborts,
and so it’s miles less complicated to implement than strict 2pl. Notice the distinction
(c

between conservative and rigorous 2pl: the former ought to lock all its objects earlier
than it starts offevolved, so once the transaction starts it is in its shrinking segment; the
latter does not release any of its objects till after it terminates (by using committing or

Amity Directorate of Distance & Online Education

 222 Introduction to Data Base Management Systems

aborting), so the transaction is in its increasing section till it ends. In lots of instances,
Notes

e
the concurrency control subsystem itself is answerable for generating the read_lock
and write_lock requests. For eg, suppose the machine is to put in force the stern 2pl
protocol. Then, each time transaction t issues a read_item(x), the system calls the

in
read_lock(x) operation on behalf of t. If the country of lock(x) is write_locked by way
of some different transaction t , the system places t inside the waiting queue for item
x; in any other case, it offers the read_lock(x) request and permits the read_item(x)

nl
operation of t to execute. Alternatively, if transaction t problems a write_item(x), the
machine calls the write_lock(x) operation on behalf of t. If the nation of lock(x) is write_
locked or read_locked through some other transaction t , the system locations t inside

O
the ready queue for object x; if the country of lock(x) is read_locked and t itself is the
most effective transaction retaining the study lock on x, the machine enhancements the
lock to write_locked and permits the write_item(x) operation via t. Sooner or later, if the
state of lock(x) is unlocked, the system offers the write_lock(x) request and lets in the

ty
write_item(x) operation to execute. After every motion, the machine need to update its
lock desk as it should be. The use of locks can motive two extra problems: deadlock
and starvation.

si
Managing impasse and hunger
Impasse refers to a selected state of affairs where two or more techniques are
r
expecting every different to launch a useful resource or extra than two methods are
awaiting the resource in a round chain. Deadlock happens while every transaction t in
ve
a set of or greater transactions is looking ahead to some object this is locked via a few
different transaction t inside the set. Hence, every transaction in the set is in a ready
queue, anticipating one of the other trans-actions within the set to release the lock on
an item. But due to the fact the alternative transaction is likewise ready, it’s going to
ni

never launch the lock. A simple eg is proven in determine 22. Five(a), in which the 2
transactions t1 and t2 are deadlocked in a partial agenda; t1 is in the ready queue for x,
that’s locked by means of t2 , at the same time as t2 is in the ready queue for y, that’s
U

locked with the aid of t1 . In the meantime, neither t1 nor t2 nor every other transaction
can get entry to gadgets x and y.

Impasse prevention protocols

ity

One manner to prevent impasse is to apply a impasse prevention protocol. One

impasse prevention protocol, that’s used in conservative -section locking, requires
that each transaction lock all the objects it desires in advance (that is usually not a
practical assumption)—if any of the objects cannot be received, none of the gadgets
m

are locked. Instead, the transaction waits and then tries again to lock all of the objects
it needs. Manifestly this solution further limits concurrency. A 2d protocol, which also
limits concurrency, entails ordering all the gadgets within the database and making sure
)A

that a transaction that needs several objects will lock them consistent with that order.
This requires that the progrmer (or the gadget) is aware about the selected order of the
items, which is also now not sensible within the database context.

Some of different impasse prevention schemes had been proposed that make a
choice about what to do with a transaction involved in a likely impasse situation: need
(c

to or not it’s blocked and made to wait or need to or not it’s aborted, or ought to the
transaction preempt and abort another transaction? Some of those strategies use the

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 223

concept of transaction timestamp ts(t), that’s a completely unique identifier assigned to

Notes

e
each transaction. The timestamps are normally based totally at the order wherein trans-
actions are started; therefore, if transaction t1 starts offevolved earlier than transaction
t2, then ts(t1) < ts(t2). Note that the older transaction (which starts offevolved first) has

in
the smaller time-stamp fee. Schemes that prevent deadlock are known as wait-die and
wound-wait. Think that transaction ti tries to lock an object x but isn’t always capable of
due to the fact x is locked with the aid of a few different transaction tj with a conflicting

nl
lock.

The policies observed by means of these schemes are:

O
◌◌ Wait-die. If ts(ti) < ts(tj), then (ti older than tj) ti is permitted to attend; in any
other case (ti more youthful than tj) abort ti (ti dies) and restart it later with the
identical timestamp.

ty
◌◌ Wound-wait. If ts(ti) < ts(tj), then (ti older than tj) abort tj (ti wounds tj) and
restart it later with the same timestamp; otherwise (ti more youthful than tj) ti is
authorized to wait.

si
In wait-die, an older transaction is authorized to anticipate a more youthful
transaction, while a younger transaction requesting an item held via an older
transaction is aborted and restarted. The wound-wait method does the opposite: a

r
younger transaction is authorized to look forward to an older one, whereas an older
transaction inquiring for an item held with the aid of a younger transaction re-empts
ve
the more youthful transaction via aborting it. Both schemes turn out to be aborting the
more youthful of the two transactions (the transaction that started later) that can be
worried in a deadlock, assuming that this could waste much less processing. It could
be shown that those strategies are impasse-unfastened, since in wait-die, transactions
ni

handiest look ahead to more youthful transactions so no cycle is created. In addition,

in wound-wait, transactions best wait for older transactions so no cycle is created.
However, each techniques may purpose a few transactions to be aborted and restarted
U

needlessly, even though those transactions may additionally in no way surely reason
a deadlock. Some other organization of protocols that save you impasse do now not
require timestamps.

These include the no ready (nw) and cautious ready (cw) algorithms. In the no
ity

ready set of rules, if a transaction is not able to achieve a lock, it’s miles immediately
aborted and then restarted after a sure time postpone with out checking whether or not
a deadlock will definitely arise or now not. In this case, no transaction ever waits, so no
deadlock will occur. However, this scheme can cause transactions to abort and restart
m

needlessly. The cautious waiting set of rules was proposed to try to lessen the variety of
pointless aborts/restarts. Assume that transaction ti attempts to fasten an object x but is
not able to achieve this because x is locked by way of some other transaction tj with a
)A

conflicting lock.

The cautious ready rules are as follows:

Cautious waiting. If tj isn’t always blocked (not expecting a few other locked
item), then ti is blocked and allowed to wait; in any other case abort ti. It could be
(c

shown that careful waiting is deadlock-free, because no transaction will ever wait for
any other blocked transaction. Through thinking about the time b(t) at which each
blocked transaction t was blocked, if the 2 transactions ti and tj above each grow to
Amity Directorate of Distance & Online Education
 224 Introduction to Data Base Management Systems

be blocked, and ti is anticipating tj, then b(ti) < b(tj), in view that ti can best look ahead
Notes

e
to tj at a time whilst t j is not blocked itself. Therefore, the blockading instances form
a total ordering on all blocked transactions, so no cycle that causes deadlock can
occur. Impasse detection. A 2nd, more practical technique to dealing with impasse is

in
deadlock detection, wherein the device assessments if a state of deadlock genuinely
exists. This answer is appealing if we recognise there might be little interference
some of the transactions—that is, if one-of-a-kind transactions will hardly ever access

nl
the identical items at the identical time. This may take place if the transactions are
short and every transaction locks only some objects, or if the transaction load is light.
However, if transactions are long and every transaction makes use of many objects, or

O
if the transaction load is pretty heavy, it could be positive to use a impasse prevention
scheme.

A easy manner to stumble on a nation of impasse is for the system to construct

ty
and preserve a wait-for graph. One node is created within the wait-for graph for each
transaction this is presently executing. Whenever a transaction ti is ready to lock
an object x this is currently locked by using a transaction tj, a directed part (ti → tj)
is created inside the wait-for graph. While tj releases the lock(s) on the items that ti

si
was awaiting, the directed facet is dropped from the wait-for graph. We’ve a state of
useless-lock if and handiest if the wait-for graph has a cycle. One hassle with this
technique is the matter of figuring out while the machine should check for a deadlock.
r
One opportunity is to test for a cycle every time an aspect is delivered to the wait-
ve
for graph, however this may cause immoderate overhead. Criteria which include the
quantity of currently executing transactions or the period of time several transactions
have been ready to lock objects may be used as a substitute to test for a cycle. Discern
22. 5(b) suggests the wait-for graph for the (partial) schedule shown in figure 22. 5(a).
ni

If the system is in a nation of impasse, some of the transactions inflicting the deadlock
ought to be aborted. Choosing which transactions to abort is referred to as sufferer
choice. The set of rules for victim choice should generally keep away from choosing
U

transactions that have been walking for a long term and that have performed many
updates, and it should strive as an alternative to pick transactions which have now not
made many modifications (younger transactions).

Timeouts. Any other easy scheme to address impasse is using timeouts. This
ity

method is sensible due to its low overhead and simplicity. On this approach, if a
transaction waits for a length longer than a system-defined timeout length, the device
assumes that the transaction can be deadlocked and aborts it—irrespective of whether
a deadlock certainly exists or now not.
m

Starvation
Hunger : every other problem that may arise while we use locking is hunger, which
)A

happens when a transaction cannot continue for an indefinite time frame even as
different transactions in the gadget continue generally. This could occur if the waiting
scheme for locked items is arbitrary, giving precedence to a few transactions over
others.
(c

One answer for hunger is to have a truthful waiting scheme, including using a
first-come-first-served queue; transactions are enabled to lock an item within the order
in which they in the beginning requested the lock. Every other scheme allows a few

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 225

transactions to have priority over others however increases the priority of a transaction
Notes

e
the longer it waits, till it eventually receives the highest priority and proceeds. Hunger
can also arise due to victim choice if the algorithm selects the equal transaction as
victim repeatedly, therefore inflicting it to abort and by no means end execution.

in
The algorithm can use better priorities for transactions which have been aborted
multiple times to keep away from this problem. The wait-die and wound-wait schemes

nl
mentioned previously keep away from starvation, due to the fact they restart a
transaction that has been aborted with its equal unique timestamp, so the possibility
that the identical transaction is aborted repeatedly is narrow.

O
5.1.6 Concurrency Control Based on Time Stamping
A transaction T that issues a read_item(X) or write_item(X) such that TS(T) >
write_TS(X) has its read or write operation delayed until the transaction T that wrote

ty
the value of X (hence TS(T ) = write_TS(X)) has committed or aborted. To implement
this algorithm, it is necessary to simulate the locking of an item X that has been written
by transaction T until T is either com-mitted or aborted. This algorithm does not cause
deadlock, since T waits for T only if TS(T) > TS(T ).

si
Thomas’s Write Rule.

r
A modification of the basic TO algorithm, known as Thomas’s write rule, does not
enforce conflict serializability, but it rejects fewer write operations by modifying the
ve
checks for the write_item(X) operation as follows:

5.1.7 Multivesion Concurrency Control

ni

Timestamps
Timestamp is a unique identifier created by means of the DBMS to perceive a
transaction. Usually, timestamp values are assigned within the order in which the
U

transactions are submitted to the device, so a timestamp can be thought of because

the transaction begin time. We are able to talk over with the timestamp of transaction t
as ts(t). Concurrency manage techniques primarily based on timestamp ordering do no
ity

longer use locks; therefore, deadlocks can not arise. Timestamps can be generated in
numerous methods. One opportunity is to use a counter this is incremented whenever
its fee is assigned to a transaction. The transaction time-stamps are numbered 1, 2,
three, ... In this scheme. A laptop counter has a finite max-imum value, so the gadget
need to periodically reset the counter to zero while no transactions are executing for
m

some quick time frame. Some other way to implement timestamps is to use the present
day date/time price of the system clock and make certain that no timestamp values
are generated in the course of the same tick of the clock. Timestamp primarily based
)A

protocol in DBMS is an algorithm which makes use of the device time or logical counter
as a timestamp to serialize the execution of concurrent transactions. The timestamp-
based protocol guarantees that every conflicting examine and write operations are
carried out in a timestamp order. The older transaction is usually given precedence
in this technique. It makes use of system time to determine the time stamp of the
(c

transaction. This is the maximum commonly used concurrency protocol. Lock-based

totally protocols help you to manipulate the order between the conflicting transactions

Amity Directorate of Distance & Online Education

 226 Introduction to Data Base Management Systems

when they’ll execute. Timestamp-based totally protocols manage conflicts as quickly as

Notes

e
an operation is created.

Eg:

in
think there are there transactions t1, t2, and t3. T1 has entered the gadget at time
0010

nl
t2 has entered the device at 0020

t3 has entered the machine at 0030

precedence might be given to transaction t1, then transaction t2 and finally

O
transaction t3.

Advantages:

ty
◌◌ schedules are serializable similar to 2pl protocols.
◌◌ no watching for the transaction, which gets rid of the opportunity of deadlocks.

si
Dangers:
◌◌ starvation is viable if the identical transaction is restarted and usually aborted.

r
The timestamp ordering algorithm
ve
The idea for this scheme is to order the transactions primarily based on their
timestamps. A schedule in which the transactions take part is then serializable, and
the only equal serial time table approved has the transactions so as of their timestamp
values. That is known as timestamp ordering (to). Word how this differs from 2pl, where
ni

a agenda is serializable by way of being equal to a few serial agenda allowed via the
locking protocols. In timestamp ordering, but, the schedule is equal to the particular
serial order corresponding to the order of the transaction time-stamps. The algorithm
need to make certain that, for every object accessed by using conflicting operations
U

inside the time table, the order in which the object is accessed does not violate the
timestamp order. To try this, the algorithm pals with each database object x two
timestamp (ts) values:
ity

◌◌ read_ts(x). The examine timestamp of object x is the most important

timestamp among all of the timestamps of transactions which have
efficaciously examine item x—that is, read_ts(x) = ts(t), wherein t is the
youngest transaction that has examine x successfully. Write_ts(x). The write
m

timestamp of object x is the biggest of all the time-

◌◌ stamps of transactions which have effectively written object x—that is,
write_ts(x) = ts(t), wherein t is the youngest transaction that has written x
)A

efficaciously. Fundamental timestamp ordering (to). On every occasion a few

transaction t tries to difficulty a read_item(x) or a write_item(x) operation, the
basic to algorithm compares the timestamp of t with read_ts(x) and write_ts(x)
to make sure that the timestamp order of transaction execution is not violated.
If this order is violated, then transac-tion t is aborted and resubmitted to the
(c

system as a new transaction with a brand new timestamp. If t is aborted and

rolled again, any transaction t1 that can have used a value written with the
aid of t should additionally be rolled again. Further, any transaction t2 which

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 227

could have used a cost written by t1 need to additionally be rolled back,

Notes

e
and so on. This effect is called cascading rollback and is one of the troubles
related to fundamental to, since the schedules produced aren’t guaranteed to
be recoverable. An additional protocol should be enforced to make certain that

in
the schedules are recoverable, cascadeless, or strict. We first describe the
basic to algorithm here. The concurrency manipulate set of rules need to take
a look at whether or not conflicting operations violate the timestamp order-ing

nl
within the following cases:
◌◌ on every occasion a transaction t problems a write_item(x) operation, the
following is checked:

O
if read_ts(x) > ts(t) or if write_ts(x) > ts(t), then abort and roll lower back t and reject
the operation. This should be achieved because some more youthful transaction with
a timestamp greater than ts(t)—and hence after t inside the timestamp ordering—has

ty
already examine or written the cost of item x earlier than t had a hazard to put in writing
x, accordingly violating the timestamp ordering. If the circumstance in element (a) does
not arise, then execute the write_item(x) operation of t and set write_ts(x) to ts(t). On
every occasion a transaction t problems a read_item(x) operation, the subsequent is

si
checked:

◌◌ if write_ts(x) > ts(t), then abort and roll again t and reject the opera-tion. This

r
must be executed due to the fact some younger transaction with time-stamp
more than ts(t)—and subsequently after t inside the timestamp ordering—has
ve
already written the cost of object x earlier than t had a chance to examine x.
●● if write_ts(x) ≤ ts(t), then execute the read_item(x) operation of t and set read_
ts(x) to the larger of ts(t) and the modern read_ts(x). Each time the primary to
algorithm detects conflicting operations that occur within the wrong order, it
ni

rejects the later of the 2 operations via aborting the transaction that issued it. The
schedules produced via primary to are consequently guaranteed to be conflict
serializable, like the 2pl protocol. However, some schedules are possible under
U

every protocol that are not allowed beneath the alternative. Accordingly, neither
protocol permits all feasible serializable schedules. As mentioned in advance,
impasse does no longer occur with timestamp ordering. However, cyclic restart
(and consequently hunger) may additionally occur if a transaction is usually
ity

aborted and restarted. Strict timestamp ordering (to). A variant of basic to called
strict to ensures that the schedules are both strict (for easy recoverability) and
(battle) serializable. In this modification,
m

Strict Timestamp Ordering (TO):

A variation of basic TO called strict TO ensures that the schedules are both strict
(for easy recoverability) and (conflict) serializable. In this variation, a transaction T that
)A

issues a read_item(X) or write_item(X) such that TS(T) > write_TS(X) has its read or
write operation delayed until the transaction T that wrote the value of X (hence TS(T ) =
write_TS(X)) has committed or aborted. To implement this algorithm, it is necessary to
simulate the locking of an item X that has been written by transaction T until T is either
com-mitted or aborted. This algorithm does not cause deadlock, since T waits for T only
(c

if TS(T) > TS(T ).

Amity Directorate of Distance & Online Education

 228 Introduction to Data Base Management Systems

Thomas’s Write Rule:

Notes

e
A modification of the basic TO algorithm, known as Thomas’s write rule, does not
enforce conflict serializability, but it rejects fewer write operations by modifying the

in
checks for the write_item(X) operation as follows:

Multivesion Concurrency Control

nl
Other protocols for concurrency control keep the vintage values of a facts
object when the item is updated. Those are referred to as multiversion concurrency
manage, due to the fact several variations (values) of an object are maintained. While

O
a transaction calls for access to an object, the correct version is chosen to keep the
serializability of the currently executing schedule, if viable. The concept is that some
read operations that might be rejected in different strategies can still be typical through
studying an older model of the item to preserve serializability. When a transaction writes

ty
an item, it writes a new edition and the antique model(s) of the object are retained. A
few multiver-sion concurrency control algorithms use the concept of view serializability
as opposed to warfare serializability.

si
An apparent drawback of multiversion techniques is that extra garage is needed
to hold more than one versions of the database items. However, older variations can
also have to be maintained anyway—as an example, for recovery purposes. Similarly,
r
a few database applications require older variations to be saved to hold a records of
the evolution of information item values. The acute case is a temporal database which
ve
continues track of all adjustments and the times at which they took place. In such
cases, there’s no extra storage penalty for multiversion strategies, seeing that older
variations are already maintained.
ni

●● multiversion approach based totally on timestamp ordering

in this technique, several variations x1, x2, ..., xk of each statistics item x
are maintained. For every version, the value of model xi and the subsequent two
U

timestamps are stored:

◌◌ read_ts(xi). The study timestamp of xi is the biggest of all the timestamps

of transactions which have successfully examine model xi. Write_ts(xi).
ity

The write timestamp of xi is the timestamp of the transaction that wrote the
price of model xi. Each time a transaction t is allowed to execute a write_
item(x) operation, a new edition xk+1 of object x is created, with each the
write_ts(xk+1) and the read_ts(xk+1) set to ts(t). Correspondingly, while a
transaction t is authorized to examine the cost of version xi, the value of read_
m

ts(xi) is ready to the larger of the current read_ts(xi) and ts(t). To make certain
serializability, the following regulations are used:
◌◌ if transaction t troubles a write_item(x) operation, and model i of x has the
)A

highest write_ts(xi) of all variations of x that is additionally much less than or

identical to ts(t), and read_ts(xi) > ts(t), then abort and roll back transaction t;
in any other case, create a new version xj of x with read_ts(xj) = write_ts(xj)
= ts(t). If transaction t issues a read_item(x) operation, locate the model i of x
that has the very best write_ts(xi) of all variations of x that is also much less
(c

than or identical to ts(t); then return the cost of xi to transaction t, and set the
price of read_ts(xi) to the larger of ts(t) and the contemporary read_ts(xi).

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 229

As we are able to see in case 2, a read_item(x) is continually successful,

Notes

e
because it unearths the ideal version xi to examine based on the write_ts of
the numerous existing variations of x. In case 1, however, transaction t may
be aborted and rolled returned. This takes place if t tries to put in writing a

in
version of x that ought to were examine by another transaction t whose
timestamp is read_ts(xi); but, t has already examine model xi, which turned
into written by using the transaction with timestamp equal to write_ts(xi). If this

nl
war happens, t is rolled returned; otherwise, a new version of x, written with
the aid of transaction t, is created. Be aware that if t is rolled again, cascading
rollback can also occur. As a result, to make sure recoverability, a transaction

O
t must not be allowed to dedicate until after all the transactions which have
written some model that t has read have devoted.
●● multiversion -section locking using certify locks

ty
on this multiple-mode locking scheme, there are three locking modes for an object:
examine, write, and certify, as opposed to simply the 2 modes (study, write) discussed
previously. Consequently, the kingdom of lock(x) for an object x can be one of study-
locked, write-locked, certify-locked, or unlocked. Within the wellknown locking scheme,

si
with simplest study and write locks , a write lock is an special lock. We are able to
describe the relationship between read and write locks in the general scheme by the
lock compatibility table . An entry of sure way that if a transaction t holds the type of lock
distinctive within the column header r
ve
ni
U
ity

Other protocols for concurrency control keep the vintage values of a facts
object when the item is updated. Those are referred to as multiversion concurrency
manage, due to the fact several variations (values) of an object are maintained. While
m

a transaction calls for access to an object, the correct version is chosen to keep the
serializability of the currently executing schedule, if viable. The concept is that some
read operations that might be rejected in different strategies can still be typical through
)A

studying an older model of the item to preserve serializability. When a transaction writes
an item, it writes a new edition and the antique model(s) of the object are retained. A
few multiver-sion concurrency control algorithms use the concept of view serializability
as opposed to warfare serializability.

An apparent drawback of multiversion techniques is that extra garage is needed

(c

to hold more than one versions of the database items. However, older variations can
also have to be maintained anyway—as an example, for recovery purposes. Similarly,

Amity Directorate of Distance & Online Education

 230 Introduction to Data Base Management Systems

a few database applications require older variations to be saved to hold a records of

Notes

e
the evolution of information item values. The acute case is a temporal database which
continues track of all adjustments and the times at which they took place. In such
cases, there’s no extra storage penalty for multiversion strategies, seeing that older

in
variations are already maintained.

●● multiversion approach based totally on timestamp ordering

nl
in this technique, several variations x1, x2, ..., xk of each statistics item x
are maintained. For every version, the value of model xi and the subsequent two
timestamps are stored:

O
read_ts(xi). The study timestamp of xi is the biggest of all the timestamps of
transactions which have successfully examine model xi. Write_ts(xi). The write
timestamp of xi is the timestamp of the transaction that wrote the price of model xi.

ty
Each time a transaction t is allowed to execute a write_item(x) operation, a new edition
xk+1 of object x is created, with each the write_ts(xk+1) and the read_ts(xk+1) set to
ts(t). Correspondingly, while a transaction t is authorized to examine the cost of version
xi, the value of read_ts(xi) is ready to the larger of the current read_ts(xi) and ts(t). To

si
make certain serializability, the following regulations are used:

if transaction t troubles a write_item(x) operation, and model i of x has the highest

r
write_ts(xi) of all variations of x that is additionally much less than or identical to
ts(t), and read_ts(xi) > ts(t), then abort and roll back transaction t; in any other case,
ve
create a new version xj of x with read_ts(xj) = write_ts(xj) = ts(t). If transaction t issues
a read_item(x) operation, locate the model i of x that has the very best write_ts(xi) of
all variations of x that is also much less than or identical to ts(t); then return the cost
of xi to transaction t, and set the price of read_ts(xi) to the larger of ts(t) and the
ni

contemporary read_ts(xi). As we are able to see in case 2, a read_item(x) is continually

successful, because it unearths the ideal version xi to examine based on the write_
ts of the numerous existing variations of x. In case 1, however, transaction t may be
U

aborted and rolled returned. This takes place if t tries to put in writing a version of x
that ought to were examine by another transaction t whose timestamp is read_ts(xi);
but, t has already examine model xi, which turned into written by using the transaction
with timestamp equal to write_ts(xi). If this war happens, t is rolled returned; otherwise,
ity

a new version of x, written with the aid of transaction t, is created. Be aware that if t is
rolled again, cascading rollback can also occur. As a result, to make sure recoverability,
a transaction t must not be allowed to dedicate until after all the transactions which
have written some model that t has read have devoted.
m

●● multiversion -section locking using certify locks

on this multiple-mode locking scheme, there are three locking modes for an object:
examine, write, and certify, as opposed to simply the 2 modes (study, write) discussed
)A

previously. Consequently, the kingdom of lock(x) for an object x can be one of study-
locked, write-locked, certify-locked, or unlocked. Within the wellknown locking scheme,
with simplest study and write locks , a write lock is an special lock. We are able to
describe the relationship between read and write locks in the general scheme by the
lock compatibility table . An entry of sure way that if a transaction t holds the type of lock
(c

distinctive within the column header

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 231

On object x and if transaction t requests the kind of lock indicated within the row
Notes

e
header on the equal item x, then, at that factor, t can get the lock in light of the truth that
the locking modes are compatible. However, a passage of no in the table demonstrates
that the locks are not like minded, so t should postpone till t offers the lock.

in
In the standard locking plan, while a transaction gets a write lock on an object, no
different transactions can get to that item. The concept behind multiversion 2pl is to

nl
allow different transactions t to study an object x even as a single transaction t holds
a write lock on x. That is cultivated through permitting two versions for each item x;
one version must always had been composed through a few committed transaction.
The second version x is made whilst a transaction t procures a write lock at the item.

O
One of a kind transactions can keep on reading the devoted version of x at the same
time as t holds the write lock. Transaction t can write the worth of x depending at the
situation, without influencing the really worth of the committed version x. Anyhow, when

ty
t is prepared to dedicate, it have to acquire a assure lock on all gadgets that it at gift
holds write locks on earlier than it can commit. The affirm lock isn’t always like minded
with study locks, so the transaction would possibly need to defer its dedicate until all
its write-locked objects are introduced by using any reading transactions to acquire the

si
guarantee locks. Once the make certain locks—that are selective locks—are procured,
the committed model x of the data item is about to the really worth of model x , model x
is disposed of, and the confirm locks are then introduced.
r
Inn this multiversion 2pl plan, reads can maintain concurrently with a unmarried
ve
write interest—a route of movement not allowed under the standard 2pl plans. The
rate is that a transaction would possibly need to postpone its devote till it acquires
restrictive verify locks on each one of the objects it has refreshed. It thoroughly can be
shown that this plan attempts no longer to path cuts brief, given that transactions are
ni

clearly authorised to read the version x that was composed with the aid of a dedicated
transaction.
U

5.1.8 Granularity of Data Items and Multiple Granularity Locking

All concurrency control techniques anticipate that the database is formed of
some of named records objects. A database object could be selected to be one of the
ity

following:

1. A database report
2. A discipline price of a database report
3. A disk block
m

4. A whole record
5. The complete database
)A

the granularity can have an effect on the overall performance of concurrency

control and recuperation.

Granularity degree concerns for locking

(c

The dimensions of facts gadgets is regularly referred to as the data object

granularity. Excellent granularity alludes to little item sizes, even though coarse
granularity alludes to huge item sizes. A few tradeoffs have to be taken into

Amity Directorate of Distance & Online Education

 232 Introduction to Data Base Management Systems

consideration in choosing the statistics object length. We are able to communicate

Notes

e
about statistics item length with reference to locking, albeit comparative arguments
can be made for other concurrency control techniques. In the first area, word that the
bigger the statistics item size is, the lower the extent of concurrency allowed. For eg,

in
if the data item size is a disk block, a transaction t that necessities to lock a record b
have to lock the entire disk block x that contains b because a lock is associated with
the whole facts item (block). Presently, assuming every other transaction s desires to

nl
lock an alternate document c that ends up living in a similar block x in a clashing lock
mode, it’s miles pressured to face by way of. On the off hazard that the facts item length
became a solitary file, transaction s could have the option to retain, considering that

O
it might lock an change data object (report). On the other hand, the extra modest the
facts object size is, the extra the amount of objects inside the database. Due to the fact
each item is related with a lock, the machine may have a larger variety of energetic
locks to be taken care of via the lock administrator. Extra lock and open responsibilities

ty
might be done, inflicting a better overhead. Also, extra more room may be needed for
the lock table. For timestamps, capacity is needed for the read_ts and write_ts for every
data object, and there will be similar overhead for managing endless items. Given the

si
above tradeoffs, an plain inquiry can be posed: what’s the satisfactory object length?
The correct reaction is that it is based upon the forms of transactions protected. In the
event that a regular transaction receives to few information, it is profitable to have the

r
statistics item granularity be one report. Then again, assuming a transaction normally
gets to many records in a comparable file, it could be smarter to have block or report
ve
granularity with the purpose that the transaction will reflect onconsideration on that
huge wide variety of data as one (or a pair) information items.

A couple of granularity stage locking:

ni

As the pleasant granularity length relies upon on the given transaction, it seems
appropriate that a database machine need to aid multiple tiers of granularity, where the
granularity degree can be different for diverse mixes of transactions. Figure 22. 7 shows
U

a simple granularity hierarchy with a database containing documents, each record

containing numerous disk pages, and every page containing numerous records. This
may be used to illustrate a a couple of granularity stage 2pl protocol, in which a lock
may be requested at any degree. But, additional styles of locks might be had to aid any
ity

such protocol correctly

m
)A
(c

Don’t forget the subsequent situation, with handiest shared and one of a kind lock

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 233

kinds, that refers to the example in discern 22. 7. Think transaction t1 desires to update
Notes

e
all of the statistics in record f1, and t1 requests and is granted an unique lock for f1.
Then all of f1’s pages (p11 through p1n)—and the facts contained on those pages—are
locked in exceptional mode. This is useful for t1 because placing a single document-

in
stage lock is extra efficient than placing n web page-level locks or having to fasten each
character file. Now think some other transaction t2 only wants to read record r1nj from
page p1n of record f1; then t2 could request a shared document-level lock on r1nj.

nl
However, the database machine (this is, the transaction supervisor or more especially
the lock manager) should verify the compatibility of the asked lock with already held
locks. One manner to verify this is to traverse the tree from the leaf r1nj to p1n to f1

O
to db. If at any time a conflicting lock is hung on any of those objects, then the lock
request for r1nj is denied and t2 is blocked and need to wait. This traversal could be
pretty efficient.

ty
However, what if transaction t2’s request came earlier than transaction t1’s
request? In this case, the shared document lock is granted to t2 for r1nj, but whilst t1’s
report-stage lock is requested, it’s far pretty tough for the lock manager to check all
nodes (pages and facts) which can be descendants of node f1 for a lock war. This will

si
be very inefficient and could defeat the cause of having a couple of granularity level
locks. To make more than one granularity level locking practical, additional forms of
locks, known as aim locks, are wished. The concept in the back of purpose locks is for
r
a transaction to indicate, alongside the route from the root to the preferred node, what
ve
form of lock (shared or one-of-a-kind) it will require from one of the node’s descendants.
There are 3 types of aim locks:

◌◌ Purpose-shared (is) suggests that one or extra shared locks may be asked on
a few descendant node(s).
ni

◌◌ Purpose-one of a kind (ix) indicates that one or extra specific locks may be
asked on a few descendant node(s).
◌◌ Shared-purpose-one of a kind (six) suggests that the cutting-edge node
U

is locked in shared mode but that one or extra exceptional locks will be
requested on a few descendant node(s).
The compatibility desk of the three purpose locks, and the shared and one of a
ity

kind locks, is proven in figure 22. Eight. Except the advent of the three varieties of aim
locks, the perfect locking protocol need to be used. The a couple of granularity locking
(mgl) protocol consists of the following regulations:

◌◌ The lock compatibility (based on parent 22. 8) should be adhered to.

m

◌◌ The root of the tree must be locked first, in any mode.

◌◌ A node n may be locked by using a transaction t in s or is mode simplest if the
determine node n is already locked with the aid of transaction t in both is or ix
)A

mode.
◌◌ A node n may be locked through a transaction t in x, ix, or six mode handiest if
the parent of node n is already locked via transaction t in either ix or six mode.
◌◌ A transaction t can lock a node only if it has not unlocked any node (to put in
(c

force the 2pl protocol).

◌◌ A transaction t can free up a node, n, best if not one of the youngsters of node
n are currently locked via t.

Amity Directorate of Distance & Online Education

 234 Introduction to Data Base Management Systems

Notes

e
in
nl
O
Rule 1 in reality states that conflicting locks can not be granted. GUIsdelines 2, 3,
and four nation the situations while a transaction might also lock a given node in any

ty
of the lock modes. Policies five and 6 of the mgl protocol enforce 2pl regulations to
supply serializable schedules. To illustrate the mgl protocol with the database hierarchy
in parent 22. 7, bear in mind the subsequent three transactions:

si
◌◌ T1 wants to replace record r111 and document r211.
◌◌ T2 desires to replace all facts on web page p12.
◌◌
r
T3 wants to read record r11j and the complete f2 file.
ve
5.1.9 Define Schedule and Serializibility
Routine: routine/schedule is a procedure of lining the transactions and executing
them one after the other. Whilst there are a couple of transactions which can be
running in a concurrent way and the order of operation is needed to be set so that the
ni

operations do no longer overlap each different, scheduling is brought into play and the
transactions are timed thus.
U

Forms of schedule:

Types of Schedules in DBMS

ity
m
)A
(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 235

Serial Schedules:
Notes

e
Schedules in which the transactions are executed non-interleaved, i.e., a serial
schedule is one in which no transaction starts until a running transaction has ended are

in
called serial schedules.

Example: Consider the following schedule involving two transactions T1 and T2.

nl
T1 T2
R(A)
W(A)

O
R(B)
W(B)
R(A)
R(B)

ty
where R(A) denotes that a read operation is performed on some data item ‘A’

This is a serial schedule since the transactions perform serially in the order T1 —>

si
T2.

Non-Serial Schedule:
r
This is a type of scheduling where the operations of more than one transactions
ve
are interleaved. This can prompt an ascent in the concurrency trouble. The transactions
are accomplished in a non-serial manner, preserving the very last product proper
and identical as the serial time table. Not at all just like the serial agenda where one
transaction must cling tight for one extra to complete all its operation, within the non-
ni

serial schedule, the opposite transaction keeps without trusting that the beyond
transaction will finish. This sort of time table does not give any advantage of the
concurrent transaction. It tends to be of two sorts particularly, serializable and non-
serializable agenda. The non-serial schedule may be divided further into serializable
U

and non-serializable.

Serializable:
ity

that is applied to hold up with the consistency of the database. It’s miles frequently
applied inside the non-serial scheduling to verify whether or not or no longer the
scheduling will activate any inconsistency. Then again, a serial schedule need
not bother with the serializability since it follows a transaction just whilst the beyond
transaction is finished. The non-serial agenda is meant to be in a serializable agenda
m

simply while it’s miles similar to the serial schedules, for a n quantity of transactions.
For the reason that concurrency is authorized for this case as a result, a couple of
transactions can execute simultaneously. A serializable agenda facilitates in further
)A

growing each useful resource use and cpu throughput. Those are of sorts : conflit
serializable:

a schedule is known as conflict serializable if it can be transformed right into a

serial agenda by way of swapping non-conflicting operations. Two operations are said
(c

to be conflicting if all conditions satisfy:

◌◌ they belong to unique transactions

Amity Directorate of Distance & Online Education

 236 Introduction to Data Base Management Systems

◌◌ they perform at the identical information object

Notes

e
◌◌ at least one in all them is a write operation

View serializable:

in
A agenda is called view serializable if it is view equal to a serial time table
(no overlapping transactions). A struggle time table is a view serializable but if the

nl
serializability contains blind writes, then the view serializable does not warfare
serializable.

Non-serializable:

O
The non-serializable schedule is split into two types, recoverable and non-
recoverable time table.

ty
Recoverable schedule:
Schedules in which transactions commit only in the end transactions whose
modifications they examine dedicate are known as recoverable schedules. In different

si
words, if some transaction tj is studying fee updated or written through some other
transaction ti, then the commit of tj have to arise after the commit of ti

r
There can be 3 sorts of recoverable agenda/schedule:
ve
Cascading agenda:
Also known as avoids cascading aborts/rollbacks (aca). While there’s a failure in
a single transaction and this leads to the rolling lower back or aborting other structured
ni

transactions, then such scheduling is referred to as cascading rollback or cascading

abort.

Eg:
U
ity
m
)A

Cascadeless Schedule:
Schedules wherein transactions examine values solely in the end transactions
(c

whose transforms they may examine dedicate are referred to as cascadeless

schedules. Continues faraway from that a solitary transaction abort prompts a

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 237

sequence of transaction rollbacks. A method to stop cascading aborts is to deny a

Notes

e
transaction from studying uncommitted modifications from one greater transaction in a
comparable timetable.

in
As such, if a few transaction tj desires to examine fee updated or written by using
a few different transaction ti, then, at that point, the commit of tj have to read it after the
dedicate of ti

nl
Example: Consider the following schedule involving two transactions T1 and T2.

T1 T2

O
R(A)
W(A)
W(A)
commit

ty
R(A)
commit

This schedule is cascadeless. Since the updated value of A is read by T2 only after

si
the updating transaction i.e. T1 commits.

Example: Consider the following schedule involving two transactions T1 and T2.

T1 r T2
ve
R(A)
W(A)
R(A)
W(A)
ni

abort
abort
U

It’s miles a recoverable time table but it does now not avoid cascading aborts. It
may be visible that if t1 aborts, t2 will need to be aborted too so as to maintain the
correctness of the agenda as t2 has already examine the uncommitted price written by
ity

t1.

Strict Scheduling:
A schedule is strict if for any two transactions ti, tj, if a write operation of ti precedes
m

a conflicting operation of tj (either read or write), then the dedicate or abort event of ti
also precedes that conflicting operation of tj. In other words, tj can examine or write up
to date or written price of ti handiest after ti commits/aborts.
)A

Example: Consider the following schedule involving two transactions T1 and T2.

T1 T2
R(A)
R(A)
(c

W(A)
commit

Amity Directorate of Distance & Online Education

 238 Introduction to Data Base Management Systems

W(A)
Notes

e
R(A)
commit

in
This is a strict schedule since T2 reads and writes A which is written by T1 only after
the commit of T1

nl
Non-Recoverable Schedule:
Example: Consider the following schedule involving two transactions T1 and T2.

O
T1 T2
R(A)
W(A)
W(A)

ty
R(A)
commit
abort

si
T2 examine the cost of a written with the aid of t1, and dedicated. T1 later aborted,
consequently the price study by means of t2 is inaccurate, but when you consider that t2

r
committed, this schedule is non-recoverable.
ve
●● Cascadeless schedules are stricter than recoverable schedules or are a subset of
recoverable schedules.
●● Strict schedules are stricter than cascadeless schedules or are a subset of
cascadeless schedules.
ni

●● Serial schedules fulfill constraints of all recoverable, cascadeless and strict

schedules and as a result is a subset of strict schedules.
U

The relation among various styles of schedules may be depicted as :

ity
m

Eg: Consider the following schedule:

)A

S:R1(A), W2(A), Commit2, W1(A), W3(A), Commit3, Commit1

5.1.10 Characterizing Schedules Based on Serializability

(c

The idea of serializability of schedules is utilized to distinGUIssh which schedules

are right when transaction executions have interleaving of their operations in the
schedules

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 239

Notes

e
in
nl
O
ty
r si
ve
ni

Serial, Nonserial, and Conflict-Serializable Schedules

U

Schedules an and b in discern 21. 5(a) and (b) are known as serial in mild of
the fact that the operations of every transaction are carried out consecutively, with
practically no interleaved operations from the alternative transaction. In a serial
ity

schedule, whole transactions are acted in serial request: t1 and in a while t2 in discern
21. 5(a), and t2 and afterward t1 in figure 21. Five(b). Schedules c and d in parent
21. Five(c) are known as nonserial in mild of the fact that each sequence interleaves
operations from the two transactions.
m

Officially, a time table s is serial if, for every transaction t taking an interest inside
the schedule, each one of the operations of t are completed consecutively inside
the schedule; anyways, the schedule is called nonserial. In this manner, in a serial
schedule, only a unmarried transaction at a time is energetic—the dedicate (or abort)
)A

of the lively transaction starts execution of the subsequent transaction. No interleaving

happens in a serial agenda. One realistic presumption we can make, on the off danger
that we agree with the transactions to be self sustaining, is that each serial time table
is considered proper. We will count on this on the grounds that every transaction is
thought to be proper each time accomplished all alone . Henceforth, it doesn’t make
(c

any distinction which transaction is executed first. However long every transaction is

Amity Directorate of Distance & Online Education

 240 Introduction to Data Base Management Systems

done from begin to finish in isolation from the operations of different transactions, we
Notes

e
get a proper outcome on the database.

The problem with serial schedules is that they restriction concurrency via

in
denying interleaving of operations. In a serial agenda, if a transaction hangs tight for
an i/o activity to finish, we can’t change the cpu processor to every other transaction,
subsequently burning thru essential cpu processing time. Moreover, if some transaction

nl
t may be very long, distinctive transactions ought to agree with that t will end every one
among its operations previous to starting. Eventually, serial schedules are viewed as
unsatisfactory by using and with the aid of. Notwithstanding, inside the occasion that we
are able to determine out which one-of-a-kind schedules are corresponding to a serial

O
sched-ule, we are able to allow those schedules to take place.

To expose our verbal exchange, don’t forget the schedules in determine 21. 5,
and expect that the underlying values of database things are x = ninety and y = ninety

ty
and that n = 3 and m = 2. Next to executing transactions t1 and t2, we might count
on that the database values have to be x = 89 and y = ninety three, as indicated by
way of the importance of the transactions. Effectively positive, executing each of the

si
serial schedules an or b gives the proper effects. Presently bear in mind the nonserial
schedules c and d. Time table c (that’s as old as 21. Three(a)) offers the consequences
x = ninety two and y = ninety three, in which the x worth is unsuitable, at the same time

r
as time table d gives the proper results.
ve
Schedule c offers an incorrect final results because of the misplaced replace
trouble tested in section three; transaction t2 peruses the worth of x earlier than it’s
far changed by transaction t1, so simply the effect of t2 on x is pondered inside the
database. The impact of t1 on x is misplaced, overwritten via t2, prompting the wrong
ni

outcome for aspect x. Be that as it could, a few nonserial schedules supply the proper
expected outcome, for instance, agenda d. We’d want to figure out which of the
nonserial schedules always provide a right outcome and which might provide wrong
consequences. The concept used to explain schedules as such is that of serializability
U

of a agenda/schedule.

The meaning of serializable schedule is as consistent with the subsequent: a time

table s of n transactions is serializable in case it is equal to some serial schedule of
ity

similar n transactions. We can characterize the concept of equivalence of schedules

right away. Observe that there are n! Possible serial schedules of n transactions and
loads extra manageable nonserial schedules. We will form two disjoint gatherings of
the nonserial schedules—those which might be identical to (at the least one) of the
serial schedules and henceforth are serializable, and those that aren’t corresponding
m

to any serial agenda and thus aren’t serializable. Saying that a nonserial agenda s is
serializable is similar to saying that it’s far cor-rect, in light of the truth that it’s far equal
to a serial time table, that is taken into consideration right. The extra inquiry is: while are
)A

schedules concept about same?

There are more than one methods of characterizing agenda equivalence. The
most trustworthy but least suitable definition consists of searching at the affects of the
schedules on the database. Two schedules are called result equal in the event that
(c

they produce a similar final condition of the database. Be that as it could, two wonderful
schedules may additionally coincidentally create a similar final state. For eg, schedules
s1 and s2 will deliver a comparable last database country inside the occasion that
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 241

they execute on a database with an underlying well worth of x = 100; anyways, for
Notes

e
different beginning values of x, the schedules aren’t final results identical. Moreover,
these schedules execute various transactions, so they simply ought now not be
viewed as equal. Therefore, end result equivalence alone can’t be utilized to represent

in
equivalence of schedules. The most cozy and maximum huge manner to address
characterizing time table equivalence isn’t always to make any suspicions about the
sorts of operations remembered for the transactions. For two schedules to be identical,

nl
the operations carried out to each information element impacted through the schedules
must be implemented to that aspect within the schedules in a similar request.

Two meanings of equivalence of schedules are for the maximum component

O
utilized: warfare equivalence and look at equivalence. We communicate about war
equivalence subsequent, that’s the more normally utilized definition. The which means
of battle equivalence of schedules is as per the following: two schedules are purported

ty
to be struggle equivalent if the request for any two conflicting operations is some thing
very comparable in the two schedules. Evaluate operations in a timetable are stated to
conflict in the event that they have a place with various transactions, get right of entry
to a comparable database object, and both both are write_item operations or one is

si
a write_item and the alternative a read_item. Inside the occasion that two conflicting
operations are implemented in diverse orders in sched-ules, the effect can be
distinctive on the database or on the transactions within the sched-ule, and henceforth
r
the schedules are not war equivalent. For eg , if a examine and write pastime take
ve
place in the request r1(x), w2(x) in plan s1, and within the contrary request w2(x), r1(x)
in plan s2, the really worth read by r1(x) can be special inside the schedules. Moreover,
if two write operations
ni
U

Occur within the order w1(x), w2(x) in s1, and inside the reverse order w2(x),
w1(x) in s2, the subsequent r(x) operation inside the schedules will examine potentially
ity

one-of-a-kind values; or if these are the remaining operations writing object x in the
schedules, the very last price of item x in the database may be exclusive. Utilizing the
notion of war equivalence, we symbolize a agenda s to be war serializable in case it’s
far (warfare) similar to some serial schedule s . In this type of case, we can reorder
m

the non conflicting operations in s until we shape the identical serial schedule s . As
indicated by means of this definition, time table d in discern 21. 5(c) is equal to the
serial time table an in parent 21. 5(a). Within the two schedules, the read_item(x) of
t2 reads the well worth of x composed with the aid of t1, at the same time as special
)A

operations examine the database esteems from the underlying database nation.
Additionally, t1 is the last transaction to write down y, and t2 is the last transaction to put
in writing x inside the schedules. Due to the fact that a could be a serial schedule and
time table d is equal to a, d is a serializable agenda. Notice that the operations r1(y)
(c

and w1(y) of schedule d don’t war with the operations r2(x) and w2(x), in view that they
get admission to diverse facts gadgets. As a result, we can pass r1(y), w1(y) before
r2(x), w2(x), prompting the identical serial schedule t1, t2.

Amity Directorate of Distance & Online Education

 242 Introduction to Data Base Management Systems

Schedule c in discern 21. 5(c) is not same to both of the two capacity serial
Notes

e
schedules an and b, and henceforth isn’t serializable. Trying to reorder the operations
of agenda c to observe an identical serial schedule falls flat considering r2(x) and
w1(x) conflict, which means that we can’t move r2(x) right down to get the equal serial

in
schedule t1, t2. Basically, in light of the fact that w1(x) and w2(x) struggle, we cannot
circulate w1(x) right down to get the same serial schedule t2, t1.

nl
Check your Understanding

A. Fill in the Blanks

O
1. A _____________ is an action or series of actions. It is performed by a single user
to perform operations for accessing the contents of the database.
2. Some relational database management systems use the ____________ to recover

ty
a database forward to a currently consistent state.
3. The ___________ property of a transaction requires that all operations of a
transaction be completed.

si
4. A ______ is a variable associated with a data item that describes the status of the
item with respect to possible operations that can be applied to it.
5. A binary lock can have two states ________ and _________ .
r
ve
B. State True or False
1. The two-phase locking protocol divides the execution phase of the transaction into
four parts.
2. In the shrinking phase , a new lock on the data item may be acquired by the
ni

transaction, but none can be released.

3. Strict-2PL waits until the whole transaction to commit, and then it releases all the
U

locks at a time.
4. An obvious advantage of multiversion techniques is that more storage is needed to
maintain multiple versions of the database items.
ity

5. The size of data items is often called the data item granularity .

Activity
1. Prepare a case study on the topic ‘Two Phase Locking Techniques’.
m

Summary
A transaction is a logical unit of work of database processing that includes one or
extra database get entry to operations. A transaction can be described as an action or
)A

series of actions this is accomplished with the aid of a unmarried user or application
application to perform operations for accessing the contents of the database. The
operations can include retrieval, (examine), insertion (write), deletion and change.
A transaction ought to be either finished or aborted. It is able to both be embedded
within an software application or may be designated interactively through a high-degree
(c

query language which include square. Its execution preserves the consistency of the
database. Each transaction ought to access shared information without interfering with

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 243

the other transactions and on every occasion a transaction successfully completes its
Notes

e
execution; its effect have to be everlasting. This primary abstraction frees the database
utility programmer from the subsequent issues:

in
◌◌ inconsistencies as a result of conflicting updates from concurrent users.
◌◌ partially finished transactions within the occasion of structures failure.
◌◌ person-directed undoing of transactions. A transaction is a chain of study and

nl
write actions which are grouped together to from a database get entry to. A
transaction may encompass a easy choose operation to generate a listing of
table contents, or it can encompass a sequence of related replace command

O
sequences. A transaction which successfully completes its execution is
stated to had been committed. Otherwise, the transaction is aborted. As a
consequence, if a dedicated transaction performs any replace operation at the
database, its effect should be reflected on the database even though there

ty
may be a failure.

Transaction log (or magazine) :

si
◌◌ To assist transaction processing, DBMSs hold a transaction file of each
exchange made to the database right into a log (also referred to as journal).
Log is a report of all transactions and the corresponding adjustments to the
r
database. The statistics stored in the log is used by the DBMS for a healing
requirement caused via a rollback statement, that’s application’s strange
ve
termination, a system (energy or network) failure, or disk crash. A few
relational database management structures (rDBMSs) use the transaction
log to get better a database ahead to a currently steady state. The DBMS
automatically replace the transaction log at the same time as executing
ni

transactions that modify the database. The transaction log stores earlier than-
and-after information about the database and any of the tables, rows and
characteristic values that participated within the transaction. The start and the
U

finishing (devote) of the transaction also are recorded inside the transaction
log. For every transaction, the subsequent statistics is recorded at the log:
◌◌ a start-of-transaction marker.
ity

◌◌ the transaction identifier which can consist of who and in which information.
◌◌ the record identifiers which encompass the identifiers for the record
occurrences.
◌◌ the operation(s) executed on the data (for instance, insert, delete, adjust).
m

◌◌ the preceding cost(s) of the modified statistics. This statistics is required

for undoing the modifications made with the aid of a partially completed
transaction. It’s far known as the undo log. Where the modification made by
)A

the transaction is the insertion of a brand new file, the previous values can be
assumed to be null.
◌◌ the updated fee(s) of the modified record(s). This statistics is needed for
making sure that the changes made by way of a devoted transaction are
in truth pondered within the database and may be used to redo these
(c

adjustments. This statistics is known as the redo a part of the log. In case the
change made by using the transaction is the deletion of a document, the up to
date values can be assumed to be null.
Amity Directorate of Distance & Online Education
 244 Introduction to Data Base Management Systems

◌◌ A commit transaction marker if the transaction is devoted, otherwise an abort

Notes

e
or rollback transaction marker. The log is written earlier than any updates
are made to the database. This is called write-in advance log method. In
this approach, a transaction is not allowed to alter the bodily database till

in
the undo part of the log is written to stable database. In case of a system
failure, the DBMS examines the transaction log for all uncommitted or
incomplete transactions and restores (rollback) the database to its previous

nl
nation based totally on the information inside the transaction log. While the
recuperation procedure is finished, the DBMS writes within the transaction log
all committed transactions that had been not physically written to the physical

O
database earlier than the failure took place. If a rollback is issued before
the termination of a transaction, the DBMS restores the database best for
that precise transaction, instead of for all transactions, so one can keep the
sturdiness of the preceding transactions. In other words, devoted transactions

ty
aren’t rolled lower back.

Transaction homes:

si
a transaction ought to have the following 4 properties, referred to as acid homes
(also called acidity of a transaction), to make sure that a database stays strong nation
after the transaction is achieved:

◌◌ r
Atomicity, Consistency, Isolation, Sturdiness
ve
Atomicity :
The atomicity belongings of a transaction requires that all operations of a
transaction be completed, if not, the transaction is aborted. In other words, a transaction
ni

is dealt with as unmarried, person logical unit of work. Therefore, a transaction must
execute and complete every operation in its good judgment before it commits its
modifications. As said in advance, the transaction is taken into consideration as one
U

operation even though there are a couple of examine and writes. Thus, transaction
completes or fails as one unit. The atomicity assets of transaction is ensured by the
transaction healing subsystem of a DBMS. Within the occasion of a system crash in
the midst of transaction execution, the healing strategies undo any outcomes of the
ity

transaction on the database.

Consistency :
database consistency is the property that each transaction sees a consistent
m

database example. In different phrases, execution of a transaction have to leave a

database in either its previous strong state or a new solid nation that reflects the new
adjustments (updates) made with the aid of the transaction. If the transaction fails,
the database should be back to the nation it was in previous to the execution of the
)A

failed transaction. If the transaction commits, the database should replicate the
brand new changes. For that reason, all resources are continually in a steady state.
The protection of consistency is typically the duty of the programmers who write the
database applications or of the DBMS module that enforces integrity constraints. A
(c

database software must be written in a manner that guarantees that, if the database
is in a consistent country before executing the transaction, it’ll be in a steady kingdom
after the whole execution of the transaction, assuming that no interference with different

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 245

transactions occur. In other phrases, a transaction ought to remodel the database from
Notes

e
one consistent kingdom to another regular country.

Isolation :

in
isolation belongings of a transaction method that the information used at some
stage in the execution of a transaction cannot be utilized by a 2d transaction till the

nl
first one is finished. This belongings isolates transactions from each other. In different
phrases, if a transaction t1 is being accomplished and is the usage of the records item
x, that facts item can not be accessed through some other transaction (t2……….. Tn)
until t1 ends. The isolation assets is enforced by means of the concurrency manipulate

O
subsystem of the DBMS.

Sturdiness :

ty
the durability assets of transaction shows the overall performance of the
database’s steady state. It states that the adjustments made by a transaction are
everlasting. They cannot be misplaced by using either a gadget failure or via the
erroneous operation of a faulty transaction. When a transaction is completed, the

si
database reaches a steady nation and that country can not be lost, even inside the
occasion of machine’s failure. Durability property is the responsibility of the healing
subsystem of the DBMS. Locking ideas and sorts of locking
r
Lock : a lock is a variable associated with a information object that describes
ve
the status of the item with appreciate to viable operations that may be applied to it.
Normally, there is one lock for each information item inside the database. Locks are
used as a way of synchronizing the get entry to through concurrent transactions to the
database object.
ni

Kinds of locks:
Several types of locks are utilized in concurrency control. To introduce locking
U

ideas gradually, we first speak binary locks, that are simple however restrictive and
so aren’t used in exercise. We then discuss shared/unique locks, which offer extra
widespread locking talents and are utilized in sensible database locking schemes.
ity

Binary lock :
A binary lock will have states or values: locked and unlocked. A distinct lock is
associated with every database object a. If the cost of the lock on a is 1, item a can’t
m

be accessed by a database operation that requests the object. If the price of the lock
on a is 0 then object may be accessed when requested. We consult with the current
cost of the lock related to object a as lock (a). There are two operations, lock object
and unencumber item are used with binary locking a transaction requests get entry
)A

to to an object a by means of first issuing a lock item (a) operation. If lock (a) = 1,
the transaction is compelled to wait. If lock (a) = zero it’s far set to 1 (the transaction
locks the object) and the transaction is permitted to get admission to item a. Whilst the
transaction is through the use of the object, it troubles an liberate object (a) operation,
which sets lock (a) to zero (unlocks the item) in order that a can be accessed by way
(c

of different transactions. Therefore binary lock enforces mutual exclusion at the facts
object.

Amity Directorate of Distance & Online Education

 246 Introduction to Data Base Management Systems

Guidelines of binary locks

Notes

e
If the easy binary locking scheme described here is used, each transaction need to
obey the subsequent policies:

in
◌◌ A transaction must difficulty the operation lock_item (a) before any read_item
(a) or write, item operations are performed in t.
◌◌ A transaction t must difficulty the operation unlock_item (a) in the end read_

nl
item (a) and write_item (a) operations are completed in t.
◌◌ A transaction t will now not difficulty a lock _item (a) operation if it already
holds the lock on object a

O
◌◌ A transaction t will now not trouble an unencumber _item (a) operation unless
it already holds the lock on object a.
◌◌ The lock manager module of the DBMS can implement these policies. Among

ty
the lock_item (a) and unlock_item (a) operations in transaction t, is said to
keep the lock on object a. At most one transaction can keep the lock on a
selected object. Accordingly no two transactions can access the’ same object

si
concurrently.

Hazards of binary locks

r
Binary locking scheme is too restrictive for database items, because at most one
transaction can keep a lock on a given item. So, binary locking gadget can’t be used for
ve
realistic purpose.

Percentage/exclusive (for examine/write) locks:

Shared/one of a kind (or examine/write) locks. The preceding binary locking
ni

scheme is just too restrictive for database objects due to the fact at maximum,
one transaction can preserve a lock on a given object. We ought to allow several
transactions to get right of entry to the equal object x if they all get right of entry to x for
U

analyzing functions handiest. This is because read operations on the equal object via
distinctive transactions aren’t conflicting . But, if a transaction is to write down an item
x, it must have specific get entry to to x. For this purpose, a unique kind of lock called
a multiple-mode lock is used. In this scheme—known as shared/specific or examine/
ity

write locks—there are 3 locking operations: read_lock(x), write_lock(x), and release(x).

A lock associated with an item x, lock(x), now has three possible states: study-locked,
write-locked, or unlocked. A study-locked item is likewise known as proportion-locked
due to the fact different transactions are allowed to read the object, while a write-locked
m

object is called distinctive-locked because a unmarried transaction exclusively holds

the lock at the object. One approach for implementing the preceding operations on a
study/write lock is to maintain tune of the range of transactions that maintain a shared
)A

(examine) lock on an item inside the lock desk. Each report inside the lock table can
have four fields: <data_item_name, lock, no_of_reads, locking_transaction(s)>. Again,
to shop area, the machine wishes to maintain lock statistics simplest for locked gadgets
within the lock desk. The fee (kingdom) of lock is either examine-locked or write-locked,
definitely coded (if we expect no facts are stored within the lock table for unlocked
(c

objects). If lock(x)=write-locked, the fee of locking_transaction(s) is a unmarried

transaction that holds the exceptional (write) lock on x. If lock(x)=read-locked, the price
of locking transaction(s) is a listing of one or extra transactions that keep the shared
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 247

(study) lock on x. The 3 operations read_lock(x), write_lock(x), and release(x) are

Notes

e
described in discern 22. 2. As earlier than, every of the 3 locking operations need to be
taken into consideration indivisible; no interleav-ing should be allowed once one of the
operations is started out until either the opera-tion terminates through granting the lock

in
or the transaction is located in a ready queue for the object.

Simplistic lock protocol:

nl
This sort of lock-based totally protocols permits transactions to acquire a lock
on each item before beginning operation. Transactions may additionally unlock the
information object after completing the ‘write’ operation.

O
Pre-claiming locking:
Pre-claiming lock protocol facilitates to evaluate operations and create a list of

ty
required facts items which might be had to provoke an execution manner. Within the
situation whilst all locks are granted, the transaction executes. After that, all locks
release when all of its operations are over.

si
Locking operations:
There are 3 locking operations called read_lock(a), write_lock(a) and release(a)
r
represented as lock-s(a), lock-x(a), unlock(a) (right here, s suggests shared lock, x
shows one-of-a-kind lock)may be carried out on a information object. A lock related
ve
to an item a, lock (a), now has 3 feasible states: “read-locked”, “write-locked,” or
“unlocked.” a read-locked item is likewise referred to as share-locked object due to the
fact different transactions are allowed to examine the object, while a write-locked object
is brought about unique-locked, because a single transaction solely holds the lock on
ni

the item. Compatibility of locks :suppose that there are a and b two unique locking
modes. If a transaction t1 requests a lock of mode on object q on which transaction
t2 presently hold a lock of mode b. If transaction can be granted lock, despite the
U

presence of the mode b lock, then we are saying mode a is compatible with mode b.

-phase locking (2pl)

ity

◌◌ the two-phase locking protocol divides the execution segment of the

transaction into 3 parts.
◌◌ within the first component, whilst the execution of the transaction begins, it
seeks permission for the lock it requires.
m

◌◌ inside the 2nd part, the transaction acquires all of the locks. The third section
is commenced as quickly because the transaction releases its first lock.
◌◌ in the third segment, the transaction cannot call for any new locks. It most
)A

effective releases the obtained locks.

There are two stages of 2pl:

growing section : within the growing segment, a brand new lock on the information
item can be acquired by the transaction, however none may be released.
(c

Shrinking segment : in the shrinking section, current lock held by the transaction
may be released, but no new locks can be acquired.

Amity Directorate of Distance & Online Education

 248 Introduction to Data Base Management Systems

Inside the under instance, if lock conversion is authorized then the subsequent
Notes

e
phase can show up:

◌◌ Upgrading of lock (from s(a) to x (a)) is allowed in developing segment.

in
◌◌ Downgrading of lock (from x(a) to s(a)) must be done in shrinking segment.
◌◌ Strict -section locking (strict-2pl)
●● the first phase of strict-2pl is just like 2pl. In the first phase, after acquiring all the

nl
locks, the transaction maintains to execute typically.
●● the handiest distinction between 2pl and strict 2pl is that strict-2pl does no longer

O
release a lock after the usage of it.
●● strict-2pl waits until the entire transaction to devote, after which it releases all the
locks at a time.

ty
●● strict-2pl protocol does now not have shrinking section of lock release.
●● it does not have cascading abort as 2pl does. In centralized 2 pl, a unmarried
website online is liable for lock control system. It has simplest one lock supervisor

si
for the complete DBMS. Number one reproduction 2pl mechanism, many lock
managers are distributed to one-of-a-kind websites. After that, a specific lock
manager is liable for handling the lock for a hard and fast of information objects.
Whilst the primary replica has been updated, the trade is propagated to the slaves.
r
ve
Disbursed 2pl
In locking mechanism, lock managers are dispensed to all web sites. They may
be chargeable for managing locks for data at that site. If no records is replicated, it’s
miles equal to number one replica 2pl. Communique charges of distributed 2pl are quite
ni

better than primary copy 2pl

Impasse takes place when each transaction t in a set of two or extra transactions
is looking forward to a few object this is locked via a few other transaction t inside the
U

set. Subsequently, every transaction inside the set is in a ready queue, watching for
one of the different trans-actions within the set to release the lock on an item. However
due to the fact the alternative transaction is likewise waiting, it will never launch the
ity

lock. Hunger : another trouble which could occur whilst we use locking is starvation,
which occurs while a transaction cannot continue for an indefinite period of time whilst
different transactions in the machine continue commonly. This could occur if the ready
scheme for locked items is arbitrary, giving priority to a few transactions over others.
One solution for hunger is to have a fair ready scheme, which includes using a first-
m

come-first-served queue; transactions are enabled to lock an object within the order
wherein they originally asked the lock. Every other scheme permits a few transactions
to have precedence over others however increases the concern of a transaction the
)A

longer it waits, until it finally receives the very best priority and proceeds. Starvation
can also occur because of sufferer choice if the algorithm selects the equal transaction
as victim repeatedly, hence causing it to abort and in no way finish execution. The set
of rules can use higher priorities for transactions which have been aborted a couple
of times to keep away from this trouble. The wait-die and wound-wait schemes
(c

mentioned previously avoid starvation, due to the fact they restart a transaction that has
been aborted with its identical authentic timestamp, so the opportunity that the same
transaction is aborted repeatedly is slender.
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 249

Questions & sporting events

Notes

e
◌◌ Give an explanation for making certain serializability by way of two-section
locking.

in
◌◌ What are the 2 segment locking techniques?
◌◌ Explain percentage/special locks.
◌◌ What are locks? Explain forms of locks.

nl
◌◌ Give an explanation for the transaction homes.

Thesaurus

O
●● atomicity : the atomicity assets of a transaction calls for that each one operations
of a transaction be completed, if no longer, the transaction is aborted. In other
phrases, a transaction is handled as single, character logical unit of labor.

ty
Therefore, a transaction need to execute and entire each operation in its good
judgment earlier than it commits its modifications. As stated earlier, the transaction
is considered as one operation even though there are more than one read and
writes. Hence, transaction completes or fails as one unit. The atomicity assets of

si
transaction is ensured by using the transaction restoration subsystem of a DBMS.
Inside the occasion of a machine crash in the midst of transaction execution, the
recovery strategies undo any consequences of the transaction on the database.
●●
r
consistency : database consistency is the belongings that each transaction sees
ve
a constant database example. In other phrases, execution of a transaction must
go away a database in both its prior solid country or a new stable kingdom that
displays the brand new modifications (updates) made through the transaction. If
the transaction fails, the database need to be again to the kingdom it become in
ni

previous to the execution of the failed transaction. If the transaction commits, the
database need to mirror the new adjustments. Accordingly, all assets are usually
in a regular nation. The protection of consistency is typically the obligation of
U

the programmers who write the database programs or of the DBMS module that
enforces integrity constraints. A database application should be written in a way
that ensures that, if the database is in a consistent state earlier than executing
the transaction, it will likely be in a regular country after the whole execution of
ity

the transaction, assuming that no interference with other transactions arise. In

different words, a transaction need to rework the database from one steady nation
to every other steady nation.
●● Isolation : isolation belongings of a transaction way that the statistics used
m

throughout the execution of a transaction can’t be used by a 2nd transaction till

the primary one is completed. This belongings isolates transactions from one
another. In different words, if a transaction t1 is being carried out and is the use of
)A

the records item x, that records object can’t be accessed by means of every other
transaction (t2……….. Tn) till t1 ends. The isolation belongings is enforced with
the aid of the concurrency manipulate subsystem of the DBMS.
●● sturdiness :the durability assets of transaction suggests the overall performance of
the database’s consistent kingdom. It states that the modifications made by means
(c

of a transaction are permanent. They can’t be lost by using both a machine failure
or by the faulty operation of a faulty transaction. Whilst a transaction is finished,
the database reaches a consistent country and that kingdom can not be lost, even
Amity Directorate of Distance & Online Education
 250 Introduction to Data Base Management Systems

inside the event of system’s failure. Durability assets is the duty of the recovery
Notes

e
subsystem of the DBMS.
●● lock : a lock is a variable associated with a records object that describes the

in
reputation of the object with appreciate to possible operations that can be applied
to it. Usually, there is one lock for each data object inside the database. Locks
are used as a method of synchronizing the get entry to with the aid of concurrent

nl
transactions to the database object.
●● study(x): study operation is used to read the cost of x from the database and
shops it in a buffer in important memory.

O
o write(x): write operation is used to jot down the fee back to the database from
the buffer.
o commit: it’s far used to shop the paintings performed permanently.

ty
o rollback: it’s far used to undo the work finished.
●● concurrency manipulate :concurrency manipulate in database control system is a
manner of handling simultaneous operations with out conflicting with each other. It

si
guarantees that database transactions are achieved concurrently and as it should
be to supply correct consequences with out violating information integrity of the
respective database. Concurrent get admission to is quite clean if all users are
r
simply studying facts. There is no manner they are able to intervene with one
ve
another. Although for any realistic database, it’d have a mix of read and write
operations and for this reason the concurrency is a task. DBMS concurrency
manipulate is used to address such conflicts, which commonly arise with a multi-
consumer device. Consequently, concurrency manipulate is the most critical
detail for proper functioning of a database management gadget in which or more
ni

database transactions are performed simultaneously, which require get right of

entry to to the identical data.
U

●● transaction :a transaction is a logical unit of work of database processing that

consists of one or greater database get entry to operations. A transaction can be
defined as an action or series of moves that is achieved through a single person
or software program to carry out operations for having access to the contents of
ity

the database. The operations can consist of retrieval, (examine), insertion (write),
deletion and modification. A transaction should be both completed or aborted.

Further Readings
1. Raghu Ramakrishnan and Johannes Gehrke. (2003). Database Management
m

Systems. Third Edition. McGraw Hill

2. Ramez Elmasri and Shamkant B. Navathe. (2016). Fundamentals of Database
)A

Systems. Seventh Edition. Pearson Education, Inc.

3. C.J. Date. (2004). An Introduction to Database Systems. Pearson Education,
Inc.

Check your Understanding-Answers

(c

A. Fill in the Blanks

1. transaction
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 251

2. transaction log
Notes

e
3. atomicity
4. lock

in
5. locked ,unlocked
B. True or False

nl
1. False
2. False

O
3. True
4. False
5. True

ty
si
r
ve
ni
U
ity
m
)A
(c

Amity Directorate of Distance & Online Education

 252 Introduction to Data Base Management Systems

Unit -5.2: Database Recovery and Security

Notes

e
Objectives:

in
At the end of this unit, you will be able to:

●● Learn about Recovery Concepts

nl
●● Understand about Recovery Techniques Based on Deferred Update and
Immediate Update
●● Know about Shadow Paging

O
●● Learn about System and Media Recovery
●● Understand about Database Security

ty
5.2.1 Recovery Concepts
A big responsibility of the database administrator is to plot for the threat of
hardware, software program, community, process, or machine failure. If one of these

si
failure affects the operation of a database machine, you ought to generally recover the
database and go back to ordinary operation as fast as might be anticipated. Restoration
need to defend the database and related customers from useless problems and avoid
r
or lessen the danger of copying paintings physically.
ve
Restoration procedures shift contingent upon the sort of failure that came about,
the buildings impacted, and the form of restoration that you carry out. If no files are
misplaced or harmed, recovery would possibly add up to shut to restarting a case. If
statistics has been lost, restoration requires extra advances.
ni

Crash recuperation
DBMS is an incredibly complicated system with many transactions being
U

accomplished every second. The sturdiness and robustness of a DBMS is based upon
its mind boggling structure and its essential hardware and system software program. If
it fails or crashes inside the midst of transactions, it’s miles ordinary that the machine
ity

might comply with a type of algorithm or strategies to get well misplaced records.

Failure class
to peer wherein the problem has befell, we sum up a failure into different
classifications, as follows −
m

Transaction Failure
a transaction needs to abort when it fails to execute or while it arrives at a point
)A

from in which it cannot pass any further. That is known as transaction failure wherein a
couple of transactions or strategies are harmed. Purposes at the back of a transaction
failure can be −

●● logical errors − wherein a transaction can not finish because it has a few code
(c

mistake or any internal blunder situation.

●● machine errors − where the database device itself ends an lively transaction

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 253

considering the DBMS can not execute it, or it wishes to forestall attributable to
Notes

e
some device condition. For eg, if there have to be an prevalence of deadlock or
resource inaccessibility, the machine aborts an lively transaction.

in
Gadget crash
there are troubles − outside to the gadget − that could motive the gadget to stop
suddenly and cause the machine to crash.

nl
For eg, interruptions in electricity deliver may additionally motive the failure of
underlying hardware or software failure.

O
Gadget crash
there are troubles − outside to the device − which could motive the device to
prevent all at once and motive the device to crash. As an instance, interruptions in

ty
energy supply may also reason the failure of underlying hardware or software program
failure. Examples may additionally encompass working system mistakes.

si
Disk failure
in early days of generation evolution, it was a commonplace hassle in which
tough-disk drives or storage drives used to fail frequently. Disk screw ups encompass
r
formation of horrific sectors, unreachability to the disk, disk head crash or another
ve
failure, which destroys all or a part of disk garage.

Storage structure
we’ve got already described the garage gadget. In quick, the storage shape may
ni

be divided into two categories −

●● volatile garage − because the name shows, a volatile garage can’t live to tell the
tale device crashes. Volatile storage devices are located very near the cpu; usually
U

they’re embedded onto the chipset itself. As an example, foremost reminiscence

and cache memory are examples of risky storage. They’re rapid however can
keep only a small quantity of facts.
ity

●● non-volatile storage − these recollections are made to live to tell the tale
machine crashes. They may be massive in information garage capacity, but
slower in accessibility. Examples may include tough-disks, magnetic tapes, flash
reminiscence, and non-volatile (battery sponsored up) ram.
m

Recovery and atomicity

Whilst a gadget crashes, it could have several transactions being completed and
diverse files opened for them to alter the statistics gadgets. Transactions are made
)A

from numerous operations, which might be atomic in nature. However in keeping with
acid residences of DBMS, atomicity of transactions as a whole ought to be maintained,
that is, either all the operations are completed or none. Whilst a DBMS recovers from a
crash, it should preserve the following −
(c

●● it have to take a look at the states of all the transactions, which had been being
carried out.

Amity Directorate of Distance & Online Education

 254 Introduction to Data Base Management Systems

●● a transaction can be within the middle of some operation; the DBMS need to make
Notes

e
sure the atomicity of the transaction in this case.
●● it need to check whether or not the transaction may be completed now or it needs

in
to be rolled again.
●● no transactions would be allowed to leave the DBMS in an inconsistent state.
There are two types of techniques, that can assist a DBMS in recuperating in

nl
addition to retaining the atomicity of a transaction −
●● preserving the logs of each transaction, and writing them onto some stable garage
earlier than without a doubt editing the database.

O
●● keeping shadow paging, wherein the adjustments are completed on a unstable
memory, and later, the real database is updated. Log-primarily based restoration
log is a sequence of statistics, which maintains the facts of actions carried out

ty
through a transaction. It’s miles vital that the logs are written previous to the actual
change and stored on a stable storage media, that’s failsafe. Log-based recovery works
as follows −

si
●● the log record is kept on a solid storage media.
●● whilst a transaction enters the gadget and starts execution, it writes a log about it.
<tn, start> r
ve
◌◌ while the transaction modifies an object x, it write logs as follows −
<tn, x, v1, v2>

it reads tn has changed the fee of x, from v1 to v2.

ni

◌◌ when the transaction finishes, it logs −

<tn, commit>
U

the database may be changed using processes −

◌◌ deferred database amendment − all logs are written directly to the strong
storage and the database is up to date when a transaction commits.
ity

◌◌ immediately database amendment − every log follows an real database

modification.
This is, the database is changed immediately after each operation. Restoration with
concurrent transactions
m

while more than one transaction are being executed in parallel, the logs are
interleaved. On the time of recuperation, it’d emerge as tough for the recovery device to
go into reverse all logs, after which begin recuperating. To ease this example, maximum
)A

modern DBMS use the concept of ‘checkpoints’.

Checkpoint
keeping and maintaining logs in real time and in actual environment might also fill
out all the reminiscence space available within the system. As time passes, the log file
(c

may also develop too huge to be dealt with in any respect. Checkpoint is a mechanism
wherein all of the preceding logs are eliminated from the device and saved completely

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 255

in a garage disk. Checkpoint proclaims a factor earlier than which the DBMS was in
Notes

e
regular state, and all the transactions were committed.

Restoration

in
while a device with concurrent transactions crashes and recovers, it behaves within
the following manner −.

nl
◌◌ the recovery machine reads the logs backwards from the cease to the last
checkpoint.
◌◌ it keeps lists, an undo-list and a redo-list.

O
◌◌ if the recovery system sees a log with <tn, start> and <tn, commit> or simply
<tn, commit>, it puts the transaction inside the redo-list.
◌◌ if the recovery device sees a log with <tn, start> however no commit or abort

ty
log determined, it puts the transaction in undo-list. All of the transactions
inside the undo-list are then undone and their logs are eliminated. All the
transactions inside the redo-listing and their previous logs are eliminated after
which redone earlier than saving their logs.

si
5.2.2 Recovery Techniques Based on Deferred Update

Deferred update r
ve
a considerable duty of the database administrator is to devise for the chance of
hardware, software, network, manner, or system failure. If this kind of failure impacts
the operation of a database system, you ought to normally get better the database and
go back to ordinary operation as speedy as can be predicted. Recovery need to guard
ni

the database and associated users from unnecessary problems and avoid or lessen the
risk of copying paintings physically.

Recovery approaches shift contingent upon the kind of failure that came about, the
U

constructions impacted, and the type of healing that you perform. If no files are lost
or harmed, recovery would possibly upload up to shut to restarting a case. If data has
been misplaced, healing calls for more advances.
ity

Crash recuperation
DBMS is an exceptionally puzzling machine with many transactions being
performed every 2d. The durability and robustness of a DBMS relies upon its thoughts
boggling structure and its essential hardware and device software. If it fails or crashes
m

in the midst of transactions, it’s far ordinary that the machine could comply with a type
of set of rules or techniques to recover misplaced statistics.
)A

Failure type
to look wherein the difficulty has passed off, we sum up a failure into specific
classifications, as follows −

Transaction Failure
(c

a transaction desires to abort whilst it fails to execute or when it arrives at a factor

from wherein it can’t cross any similarly. That is known as transaction failure where

Amity Directorate of Distance & Online Education

 256 Introduction to Data Base Management Systems

multiple transactions or strategies are harmed. Purposes behind a transaction failure

Notes

e
may be −

◌◌ logical blunders − wherein a transaction cannot end since it has a few code

in
mistake or any inner blunder circumstance.
◌◌ device mistakes − where the database machine itself ends an lively
transaction considering that the DBMS can not execute it, or it wishes to

nl
forestall as a result of some machine circumstance. For eg, if there should
be an occurrence of deadlock or resource inaccessibility, the device aborts an
energetic transaction.

O
The protocol and the way one-of-a-kind entries are affected can be pleasant
summarised as proven:

Log entry Log written to disk Changes written Changes written

ty
to database buffer on disk
start_transactionT) No N/A N/A
read_item(T,x) No N/A N/A

si
wirte_itme(T,x) No N/A N/A
commit(T) Yes Yes *Yes
chekpoint Yes
r Underfined
*Yes: writing back to disk may occur not immediately
Yes(of committed Ts)
ve
Figure 14.6

Deferred update in a single-user surroundings

ni

we first discuss recovery based on deferred update in single-person structures,

wherein no concurrency manipulate is wanted, so that we are able to apprehend the
U

restoration technique independently of any concurrency control method. In such an

surroundings, the healing set of rules can be instead simple.

It really works as follows.

ity

Use lists to preserve the transactions: the dedicated transactions listing, which
contains all the devoted transactions because the ultimate checkpoint, and the
energetic transactions list (at most one transaction falls on this category, because the
system is a unmarried-person one). Practice the redo operation to all of the write_item
operations of the devoted transactions from the log inside the order in which they were
m

written to the log. Restart the energetic transactions.

The redo technique is described as follows:

)A

redoing a write_item operation consists of analyzing its log entry write_item(t, x,

old_value, new_value) and setting the value of item x within the database to new_value.
The redo operation is needed to be idempotent, as mentioned earlier than. Word that
the transaction inside the active list will have no impact on the database due to the
(c

deferred replace protocol, and is disregarded completely by the restoration process. It

is implicitly rolled lower back, because none of its operations have been reflected inside

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 257

the database. However, the transaction must now be restarted, either robotically via the
Notes

e
recuperation manner or manually by the user.

Action Entry in log

in
start_transaction(T) commit(T)
Re-submit Yes No

nl
Redo Yes Yes

Fig:14.17

O
The diagram under suggests an example of healing in a unmarried-person
surroundings, wherein the primary failure occurs at some stage in execution of
transaction t2. The recovery process will redo the write_item(t1, d, 20) entry in the log
by resetting the fee of object d to twenty (its new value). The write(t2, …) entries in the

ty
log are neglected by using the healing procedure due to the fact t2 is not committed. If
a second failure takes place all through restoration from the primary failure, the same
recuperation system is repeated from begin to finish, with identical outcomes.

r si
ve
Figure 14.8
ni
U

Figure 14.9
ity

Action Entry in log

start_transaction(T) commit(T)
Re-submit Yes No
m

Redo Yes Yes

)A
(c

Amity Directorate of Distance & Online Education

 258 Introduction to Data Base Management Systems

Notes

e
in
nl
Deferred update in a multi-user environment
for a multi-user device with concurrency control, the healing system can be more
complicated, depending at the protocols used for concurrency manipulate. In many

O
cases, the concurrency manage and recovery processes are interrelated. In general,
the extra the diploma of concurrency we want to attain, the greater difficult the
undertaking of recovery becomes.

ty
Take into account a system wherein concurrency manipulate makes use of -phase
locking (primary 2pl) and prevents impasse through pre-assigning all locks to items
needed by way of a transaction earlier than the transaction starts execution. To mix the
deferred update techniques for restoration with this concurrency manipulate method,

si
we are able to preserve all the locks on objects in effect till the transaction reaches its
commit point. After that, the locks can be released.

r
This ensures strict and serialisable schedules. Assuming that checkpoint entries
are blanketed within the log, a possible healing set of rules for this situation is given
ve
underneath. Use lists of transactions maintained via the gadget: the devoted
transactions list which includes all dedicated transactions because the closing
checkpoint, and the active transactions list.
ni

Redo all the write operations of the devoted transactions from the log, inside the
order in which they were written into the log. The transactions in the active list which are
active and did now not devote are effectively cancelled and should be resubmitted. The
redo technique is similar to defined in advance inside the deferred update within the
U

single-consumer environment.

The diagram below indicates an instance schedule of executing transactions.

When the checkpoint become taken at time t1, transaction t1 had dedicated, whereas
ity

transaction t3 and t4 had no longer. Earlier than the machine crash at time t2, t3
and t2 have been committed but not t4 and t5. In keeping with the deferred replace
method, there may be no need to redo the write operations of transaction t1 or any
transactions devoted before the final checkpoint time t1. Write operations of t2 and t3
should be redone, however, because each transactions reached their devote factors
m

after the closing checkpoint. Recall that the log is pressure-written before committing
a transaction. Transaction t4 and t5 are ignored: they may be effectively cancelled
and rolled again because none in their write operations have been recorded inside the
)A

database beneath the deferred replace protocol.

(c

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 259

Notes

e
in
nl
O
Figure 14.10

ty
5.2.3 Recovery Techniques Based on Immediate Update

Immediate update

si
Within the on the spot update strategies, the database may be up to date with the
aid of the operations of a transaction right away, earlier than the transaction reaches its
devote point. However, these operations are normally recorded inside the log on disk
r
with the aid of pressure-writing before they may be applied to the database, in order
ve
that recovery is feasible.

Whilst immediately replace is authorized, provisions should be made for undoing

the impact of update operations at the database, due to the fact a transaction can fail
after it has applied some updates to the database itself.
ni

Consequently recuperation schemes based on immediate update need to

encompass the functionality to roll returned a transaction by means of undoing the
U

impact of its write operations.

Whilst a transaction starts off evolved, write an entry start_transaction(t) to the log;

While any operation is achieved with the intention to exchange values in the
ity

database, write a log entry write_item(t, x, old_value, new_value);

◌◌ Write the log to disk;

◌◌ As soon as the log record is written, write the replace to the database buffers;
◌◌ Whilst handy, write the database buffers to the disk;
m

◌◌ When a transaction is about to dedicate, write a log record of the form

dedicate(t);
)A

◌◌ Write the log to disk.

The protocol and the way extraordinary entries are affected may be quality
summarised underneath:
(c

Amity Directorate of Distance & Online Education

 260 Introduction to Data Base Management Systems

Log entry Log written to disk Changes written Changes written

Notes

e
to database buffer on disk
start_transactionT) No N/A N/A

in
read_item(T,x) No N/A N/A
wirte_itme(T,x) Yes Yes *yes
commit(T) Yes undefined Undefined

nl
chekpoint Yes Underfined Yes(of committed Ts)
*Yes: writing back to disk may occur not immediately

O
Figure 14.15

By and large, we can recognize two principle classes of immediate update

algorithms. If the recovery technique guarantees that all updates of a transaction are

ty
recorded in the database on disk before the transaction commits, there will never be
a need to redo any operations of committed transactions. Such a calculation is called
UNDO/NO-REDO. Then again, if the transaction is permitted to commit before the

si
entirety of its progressions are kept in touch with the database, we have the UNDO/
REDO method, the most broad recovery calculation. This is likewise the most mind
boggling technique. Recovery exercises are summed up beneath:
r
ve
Action Entry in log
start_transaction(T) commit(T)
Undo and resubmit T Yes No
ni

Redo Yes Yes

Figure 14.16
U

Immediate update in a single-user environment

We first keep in mind a single-user gadget with the intention that we will examine
the recovery manner one after the other from concurrency manage. In the event that
a failure happens in a single-user machine, the executing transaction on the time of
ity

failure would possibly have recorded a few adjustments inside the database. The
impact of all such operations need to be undone as part of the recovery manner.
Consequently, the restoration set of rules needs an undo method, depicted sooner
or later, to undo the effect of positive write operations that have been applied to the
m

database following exam in their machine log access. The restoration set of rules
likewise utilizes the redo approach characterized earlier than. Recuperation takes place
within the accompanying manner.
)A

Utilize two lists of transaction maintained via the device: the committed
transactions since the remaining checkpoint, and the energetic transactions (at
maximum one transaction will fall on this category, in mild of the truth that the machine
is unmarried user). Undo all of the write operations of the active transaction from the
log, utilising the undo technique portrayed hereafter. Redo all of the write operations of
(c

the devoted transactions from the log, within the request where they had been written
inside the log, using the redo method.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 261

The undo method is described as follows:

Notes

e
undoing a write operation consists of analyzing its log entry write_item(t, x, old_
value, new_value) and placing the price of x inside the database to old_value. Undoing

in
some of such write operations from one or extra transactions from the log should
proceed within the reverse order from the order in which the operations have been
written within the log.

nl
Instant replace in a multi-person environment:
Whilst concurrency execution is authorized, the restoration manner again relies

O
upon on the protocols used for concurrency control. The technique below outlines
a recuperation approach for concurrent transactions with instantaneous replace.
Expect that the log consists of checkpoints and that the concurrency manage protocol
produces strict schedules – as, as an instance, the strict 2pl protocol does. Consider

ty
that a strict schedule does not permit a transaction to study or write an object unless the
transaction that closing wrote the item has committed. However, deadlocks can arise in
strict 2pl, accordingly requiring undo of transactions.

si
Use lists of transaction maintained by using the machine:
The dedicated transactions for the reason that final checkpoint, and the energetic
r
transactions. Undo all the write operations of the active (uncommitted) transaction
from the log, the usage of the undo method. The operations should be undone within
ve
the reverse of the order wherein they had been written into the log. Redo all the write
operations of the dedicated transactions from the log, in the order wherein they have
been written in the log, the use of the redo procedure.
ni

5.2.4 Shadow Paging

In the shadow web page scheme, the database isn’t always immediately changed
U

but a duplicate, saved on permanent storage (e. G. Disk), is fabricated from the part of
the database to be modified and all changes are made to this replica. In the meantime,
the vintage version of the database stays intact. As soon as the transaction commits,
the modified reproduction replaces the unique in an atomic way, i. E. The substitute is
ity

finished in its entirety or on no account. If a system crashes at this point, the vintage
version remains available for restoration.

Web page control

m

the paging scheme as used inside the working device for digital memory
management is suitable. The reminiscence this is accessed by way of a technique (a
software in execution is a system) is called digital reminiscence. Virtual reminiscence is
split into pages which are all of a sure length (normally 4096 bytes or 4k). The virtual or
)A

logical pages are mapped onto physical reminiscence blocks (i. E. Disk bodily facts) of
the same length because the pages. The mapping is executed by way of consulting a
web page table (or directory). The web page desk lists every logical web page identifier
and the cope with of the physical blocks that in reality stores the logical web page.
(c

The advantage of this scheme is that the consecutive logical pages want now not be
mapped onto consecutive bodily blocks.

Amity Directorate of Distance & Online Education

 262 Introduction to Data Base Management Systems

Shadow paging scheme in a single-consumer surroundings

Notes

e
In the shadow page scheme, two page tables are used. The unique web page table
(shadow page table) and the cutting-edge page desk. First of all, each web page tables

in
factor to the same blocks of physical storage. The modern-day web page desk is the
best path thru which a transaction can advantage get entry to to the statistics saved
on disk. This is, a transaction always makes use of the cutting-edge web page table to

nl
retrieve the right database blocks.

O
ty
r si
ve
In the course of transaction execution, the shadow web page table is in no way
changed. Whilst a write operation is finished, a new copy of the changed database
page is created, but the antique replica of that page is not overwritten. Instead, the
brand new page is written some place else on a few previously unused disk block.
ni

The cutting-edge web page table access is changed to point to the new disk block,
whereas the shadow page desk isn’t always changed and keeps to point to the vintage
unmodified disk block. The diagram above illustrates the principles of a shadow web
U

page table and a present day page desk. For pages up to date through the transaction,
variations are stored. The antique model is referenced by way of the shadow web page
desk and the new version by way of the cutting-edge web page table.
ity

To get over a failure at some stage in transaction execution, it’s far enough to free
the modified database pages and to discard the current page desk. The country of the
database before transaction execution is available via the shadow web page desk, and
that country is recovered by using reinstating the shadow page table in order that it
becomes the modern-day page desk yet again. The database thus is again to its state
m

previous to the transaction that was executing while the crash befell, and any modified
pages are discarded.
)A

Committing a transaction corresponds to discarding the previous shadow web

page table and liberating antique web page tables that it references. On account that
recuperation involves neither undoing nor redoing information gadgets, this method is
referred to as the no-undo/no-redo recovery technique. The benefit of shadow paging
is that it makes undoing the impact of the executing transaction very simple. There may
(c

be no want to undo or redo any transaction operations. In a multi-person surroundings

with concurrent transactions, logs and checkpoints have to be integrated into the
shadow paging technique. One drawback of shadow paging is that the up to date
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 263

database pages trade area on disk. This makes it difficult to hold associated database
Notes

e
pages close collectively on disk without complex garage control techniques.

Furthermore, if the web page desk (directory) is big, the overhead of writing

in
shadow page tables to disk as transactions commit is large. A in addition worry is a way
to handle garbage series whilst a transaction commits. The vintage pages referenced
by the shadow web page that has been up to date ought to be launched and brought to

nl
a listing of free pages for future use. Those pages are not wanted after the transaction
commits, and the modern-day page desk replaces the shadow web page desk to turn
out to be the valid web page desk.

O
5.2.5 System Recovery
Transactions (or gadgets of work) in opposition to a database may be interrupted
abruptly. If a failure happens before all the modifications that are part of the unit of work

ty
are finished, dedicated, and written to disk, the database is left in an inconsistent and
unusable kingdom. Crash restoration is the system with the aid of which the database
is moved back to a regular and usable state. This is completed by using rolling lower

si
back incomplete transactions and completing committed transactions that were still in
memory while the crash passed off

r
ve
ni
U
ity

if the database or the database supervisor fails, the database can be left in an
inconsistent nation. The contents of the database would possibly consist of adjustments
m

made by means of transactions that have been incomplete on the time of failure. The
database can also be missing adjustments that have been made via transactions
that finished before the failure but which had been not but flushed to disk. A crash
)A

restoration operation must be accomplished so that it will roll lower back the in part
finished transactions and to write down to disk the adjustments of finished transactions
that have been previously made best in reminiscence. Conditions that may necessitate
a crash healing encompass:

◌◌ a power failure at the device, inflicting the database supervisor and the
(c

database walls on it to go down.

◌◌ a hardware failure which includes memory, disk, cpu, or network failure.

Amity Directorate of Distance & Online Education

 264 Introduction to Data Base Management Systems

◌◌ a serious working machine errors that causes the database example to quit
Notes

e
abnormally.
If you need crash restoration to be finished automatically by the database

in
supervisor, enable the automatic restart (autorestart) database configuration parameter
via putting it to on. (that is the default value.) in case you do not want automated restart
behavior, set the autorestart database configuration parameter to off. As a end result,
you ought to issue the restart database command while a database failure takes

nl
place. If the database i/o turned into suspended before the crash passed off, you must
specify the write resume option of the restart database command so as for the crash
recovery to keep. There are two particular sorts of crash healing to be aware about:

O
member crash restoration and organization crash restoration. Member crash recovery
is the process of improving a part of a database using the log flow of a unmarried
member after a member failure. Member crash restoration, that’s commonly initiated
mechanically as part of a member restart, is a web operation-which means that different

ty
participants can still access the database.

Multiple members can be present process member crash restoration on the same
time. Group crash recuperation is the system of recuperating a database using more

si
than one members’ log streams after a failure that reasons no possible cluster caching
facility to stay within the cluster. Organization crash recovery is also commonly initiated
mechanically (as a part of a group restart) and the database is inaccessible even as
r
it’s miles in progress, as with db2 crash recovery operations outside of a database
ve
purescale surroundings. If crash healing takes place on a database this is enabled
for rollforwardrecovery , and an mistakes takes place throughout crash recovery that
is attributable to an individual desk area, that table space is taken offline, and cannot
be accessed till it’s far repaired. Crash restoration keeps on different table areas. On
ni

the completion of crash restoration, the opposite table areas within the database are
available, and connections to the database may be mounted. However, if the table
area that is taken offline is the table space that carries the device catalogs, it must be
U

repaired earlier than any connections are authorized.

This conduct does no longer follow to database purescale environments. If an error

takes place for the duration of member crash restoration or institution crash recovery,
the crash healing operation fails. If the database is configured for connectivity at some
ity

stage in crash recuperation, the database may grow to be connectable at the same time
as crash healing is in development. Tables, indexes or objects that are nonetheless
present process rollback will be locked in special mode or exceptional exclusive mode.
m

5.2.6 Media Recovery

An errors can stand up when seeking to write or study a report this is required
to perform an oracle database. This prevalence is called media failure due to the fact
)A

there may be a physical trouble reading or writing to files at the garage medium. A not
unusual instance of media failure is a disk head crash, which reasons the lack of all
documents on a disk pressure. All documents associated with a database are prone
to a disk crash, such as datafiles, online redo log documents, and control documents.
The appropriate healing from a media failure relies upon at the files affected. In case
(c

you repair the archived redo log documents and facts documents, then you definitely
ought to perform media restoration before you can open the database. Any database

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 265

transactions inside the archived redo log files no longer reflected in the statistics files
Notes

e
are carried out to the data files, bringing them to a transaction-consistent country earlier
than the database is opened. Media recovery calls for a manage report, information
files (commonly restored from backup), and on-line and archived redo log documents

in
containing changes because the time the records documents have been backed up.
Media healing is most customarily used to recover from media failure, which includes
the lack of a document or disk, or a user mistakes, which include the deletion of the

nl
contents of a table. Media recuperation may be a whole healing or a point-in-time
restoration. Whole healing can apply to character datafiles, tablespaces, or the whole
database. Factor-in-time recovery applies to the complete database

O
in a complete restoration, you repair backup records files and apply all adjustments
from the archived and on line redo log documents to the statistics files. The database
is back to its kingdom at the time of failure and can be opened without a lack of

ty
information. In a factor-in-time healing, you come back a database to its contents at a
consumer-decided on time in the beyond. You repair a backup of statistics documents
created earlier than the goal time and an entire set of archived redo log documents
from backup creation thru the goal time. Restoration applies modifications among the

si
backup time and the goal time to the facts files. All adjustments after the goal time are
discarded. Rman allows you to perform both a entire and a factor-in-time recuperation
of your database. However, this documentation makes a speciality of whole recovery.
r
ve
5.2.7 Database Security
Database security includes a diffusion of measures used to relaxed database
control structures from malicious cyber-attacks and illegitimate use. Database
protection applications are designed to protect now not most effective the information
ni

within the database, but additionally the information control machine itself, and
every utility that accesses it, from misuse, harm, and intrusion. Database security
encompasses equipment, strategies, and methodologies which establish security
U

inside a database surroundings. Database protection threats many software program

vulnerabilities, misconfigurations, or patterns of misuse or carelessness may want to
bring about breaches. Here are some of the maximum recognized causes and forms of
database protection cyber threats.
ity

Insider threats
An insider hazard is a safety chance from one of the following three resources,
each of which has privileged way of entry to the database:
m

◌◌ a malicious insider with sick-purpose

◌◌ a negligent character inside the corporation who exposes the database to
assault through careless moves
)A

◌◌ an intruder who obtains credentials via social engineering or other strategies,

or gains get entry to to the database’s credentials
◌◌ an insider hazard is one of the most standard causes of database security
breaches and it regularly happens because a number of employees have
(c

been granted privileged consumer access.

Amity Directorate of Distance & Online Education

 266 Introduction to Data Base Management Systems

Human mistakes
Notes

e
weak passwords, password sharing, unintended erasure or corruption of data,
and other undesirable user behaviors are nonetheless the motive of virtually half of of

in
statistics breaches mentioned. Exploitation of database software vulnerabilities

Attackers continuously attempt to isolate and goal vulnerabilities in software

program, and database management software is a distinctly treasured target. New

nl
vulnerabilities are located day by day, and all open source database control platforms
and industrial database software providers issue security patches regularly. But, in case
you don’t use those patches speedy, your database is probably uncovered to assault.

O
Even in case you do follow patches on time, there is always the risk of zero-day attacks,
when attackers find out a vulnerability, however it has now not but been discovered and
patched through the database vendor.

ty
Sq./nosql injection attacks
a database-precise risk includes the usage of arbitrary non-square and square
assault strings into database queries. Typically, these are queries created as an

si
extension of web application bureaucracy, or obtained through http requests. Any
database device is vulnerable to those assaults, if builders do no longer adhere
to comfortable coding practices, and if the corporation does not carry out ordinary
vulnerability testing. r
ve
Buffer overflow attacks
buffer overflow takes area whilst a system attempts to write a big amount of
statistics to a set-period block of memory, extra than it is accepted to keep. Attackers
ni

would possibly use the excess statistics, saved in adjacent reminiscence addresses, as
the start line from which to release assaults.

Denial of carrier (dos/ddos) attacks

U

Inn a denial of service (dos) assault, the cybercriminal overwhelms the target
provider—in this example the database server—the use of a large amount of faux
requests. The result is that the server cannot perform actual requests from actual
ity

users, and regularly crashes or will become unstable. In a allotted denial of service
assault (ddos), fake traffic is generated via a huge variety of computer systems, taking
part in a botnet managed by means of the attacker. This generates very large traffic
volumes, which can be difficult to prevent with out a tremendously scalable protective
architecture.
m

Cloud-based totally ddos protection offerings can scale up dynamically to address

very large ddos assaults.
)A

Malware
malware is software program written to take benefit of vulnerabilities or to
cause harm to a database. Malware could arrive thru any endpoint tool linked to
the database’s community. Malware safety is important on any endpoint, however
(c

particularly so on database servers, due to their high price and sensitivity.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 267

An evolving it environment
Notes

e
●● the evolving it environment is making databases extra at risk of threats. Right
here are tendencies that can lead to new styles of assaults on databases, or may

in
additionally require new protecting measures:
●● growing data volumes—garage, facts capture, and processing is growing
exponentially throughout almost all companies. Any information safety practices

nl
or gear need to be tremendously scalable to address remote and near-destiny
requirements.
●● distributed infrastructure—network environments are increasing in complexity,

O
in particular as organizations switch workloads to hybrid cloud or multi-cloud
architectures, making the deployment, control, and desire of protection answers
extra hard.

ty
●● increasingly tight regulatory requirements—the global regulatory compliance
panorama is developing in complexity, so following all mandates are becoming
more hard.

si
●● cybersecurity skills scarcity—there’s a global shortage of skilled cybersecurity
professionals, and groups are locating it tough to fill security roles. This may make
it more difficult to guard vital infrastructure, including databases. A way to relaxed

r
your database server? A database server is a bodily or digital gadget going for
walks the database. Securing a database server, also known as “hardening”, is a
ve
method that consists of physical safety, community protection, and secure running
machine configuration.

Make certain physical database safety

ni

Refrain from sharing a server for net programs and database packages, in case
your database consists of sensitive statistics. Although it may be cheaper, and easier,
to host your web page and database collectively on a web hosting issuer, you’re
U

putting the security of your statistics in a person else’s hands. In case you do rely on
a web website hosting carrier to manipulate your database, you must ensure that it’s
miles a organization with a sturdy safety song document. It’s miles great to live clear
of unfastened website hosting offerings due to the possible lack of protection. If you
ity

manipulate your database in an on-premise information middle, take into account that
your information center is also susceptible to attacks from outsiders or insider threats.
Make certain you have physical safety features, consisting of locks, cameras, and
protection employees for your physical facility. Any get entry to to bodily servers need
m

to be logged and most effective granted to authorized people. In addition, do no longer

depart database backups in locations that are publicly available, along with transient
partitions, web folders, or unsecured cloud garage buckets.
)A

Lock down debts and privileges

Permit’s do not forget the oracle database server. After the database is mounted,
the oracle database configuration assistant (dbca) robotically expires and locks
maximum of the default database user accounts. If you set up an oracle database
(c

manually, this doesn’t show up and default privileged money owed gained’t be expired
or locked. Their password stays similar to their username, by default. An attacker will try
and use those credentials first to connect with the database. It is vital to make certain

Amity Directorate of Distance & Online Education

 268 Introduction to Data Base Management Systems

that each privileged account on a database server is configured with a strong, specific
Notes

e
password. If bills are not wished, they ought to be expired and locked. For the last bills,
get right of entry to must be constrained to the absolute minimal required. Each account
should handiest have access to the tables and operations (for example, pick or insert)

in
required by way of the consumer. Keep away from developing person debts with get
right of entry to to every table within the database. Often patch database servers

nl
make certain that patches stay modern-day. Effective database patch control is a
vital safety exercise because attackers are actively seeking out new security flaws in
databases, and new viruses and malware appear on a each day basis. A well timed
deployment of up-to-date variations of database service packs, vital protection hotfixes,

O
and cumulative updates will enhance the steadiness of database overall performance.

Disable public community get entry to

ty
Companies shop their packages in databases. In most real-global eventualities, the
quit-consumer doesn’t require direct get right of entry to to the database. Therefore, you
ought to block all public community get right of entry to to database servers except you
are a website hosting issuer. Preferably, an organisation have to installation gateway

si
servers (vpn or ssh tunnels) for faraway directors.

Encrypt all documents and backups

r
Regardless of how stable your defenses are, there’s continually a opportunity that
ve
a hacker may additionally infiltrate your device. But, attackers aren’t the only hazard
to the security of your database. Your employees may additionally pose a danger on
your business. There’s continually the possibility that a malicious or careless insider
will benefit get admission to to a record they don’t have permission to get admission
ni

to. Encrypting your statistics makes it unreadable to both attackers and personnel.
Without an encryption key, they can’t access it, this affords a closing line of protection
against unwelcome intrusions. Encrypt all-crucial application documents, data files, and
U

backups in order that unauthorized users can’t examine your crucial facts.

Database safety first-rate practices

Here are numerous exceptional practices you may use to improve the security of
ity

touchy databases.

Actively manipulate passwords and consumer get admission to

when you have a big organisation, you need to think about automating access
m

management via password control or get entry to control software. This can offer
authorized customers with a short-time period password with the rights they need every
time they want to benefit access to a database. It additionally maintains song of the
)A

activities completed at some stage in that time body and prevents administrators from
sharing passwords. Whilst administrators may experience that sharing passwords
is handy, but, doing so makes powerful database responsibility and safety nearly not
possible. The subsequent safety features are endorsed:

◌◌ robust passwords must be enforced

(c

◌◌ password hashes need to be salted and stored encrypted

◌◌ bills have to be locked following multiple login attempts
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 269

◌◌ debts must be frequently reviewed and deactivated if body of workers flow to

Notes

e
exclusive roles, depart the organization, or now not require the same stage of
get right of entry to

in
Take a look at your database protection
As soon as you have got put in place your database protection infrastructure,
you ought to test it in opposition to a real danger. Auditing or performing penetration

nl
checks in opposition to your own database will assist you get into the mind-set of a
cybercriminal and isolate any vulnerabilities you can have omitted. To ensure the check
is complete, contain moral hackers or diagnosed penetration checking out services

O
in your security testing. Penetration testers provide vast reports listing database
vulnerabilities, and it’s miles essential to speedy check out and remediate those
vulnerabilities. Run a penetration test on a critical database gadget as a minimum as
soon as in line with yr.

ty
Use real-time database tracking whuch helps in continually scanning your
database for breach tries will increase your safety and lets you rapidly react to feasible
attacks. In particular, record integrity tracking (fim) will let you log all movements carried

si
out on the database’s server and to provide you with a warning of capability breaches.

While fim detects a alternate to critical database files, make sure protection teams
r
are alerted and able to research and reply to the hazard. Use web application and
database firewalls
ve
You ought to use a firewall to defend your database server from database
protection threats. Through default, a firewall does now not permit get right of entry to
to site visitors. It wishes to also stop your database from starting outbound connections
ni

until there is a specific cause for doing so. As well as safeguarding the database with
a firewall, you ought to install an internet software firewall (waf). That is because
attacks aimed toward web applications, which include square injection, may be used
to advantage illicit get right of entry to for your databases. A database firewall will now
U

not stop most net software assaults, due to the fact traditional firewalls operate at the
community layer, while internet software layers function on the software layer . A waf
operates at layer 7 and is able to come across malicious net utility visitors, together with
ity

sq. Injection attacks, and block it earlier than it is able to damage your database.

Check your Understanding

A. Fill in the Blanks

m

1. _______ storage cannot survive system crashes.

2. __________ is a mechanism where all the previous logs are removed from the
system and stored permanently in a storage disk.
)A

3. The __________ system reads the logs backwards from the end to the last
checkpoint.
4. The _________ is needed in case the system fails after the transaction commits but
before all its changes are recorded in the database.
(c

5. _________ enables you to perform both a complete and a point-in-time recovery of

your database.

Amity Directorate of Distance & Online Education

 270 Introduction to Data Base Management Systems

State True or False

Notes

e
1. Crash recovery is the process by which the database is moved back to a consistent
and usable state.

in
2. In the deferred update techniques, the database may be updated by the operations
of a transaction immediately, before the transaction reaches its commit point.
3. The logs are written prior to the actual modification and stored on a stable storage

nl
media.
4. Recovery is a mechanism where all the previous logs are removed from the system
and stored permanently in a storage disk.

O
5. The memory that is accessed by a process is called virtual memory.

Activity

ty
1. Make a report on the topic ‘Database Recovery and Security’.

Summary

si
●● a prime responsibility of the database administrator is to put together for the
opportunity of hardware, software program, network, method, or system failure. If
the sort of failure affects the operation of a database gadget, you should typically
r
recover the database and return to ordinary operation as quick as possible.
ve
Recuperation ought to protect the database and associated users from pointless
troubles and avoid or lessen the opportunity of having to copy work manually.
●● crash restoration
●● DBMS is a tremendously complicated gadget with hundreds of transactions being
ni

carried out each second. The durability and robustness of a DBMS relies upon
on its complicated architecture and its underlying hardware and gadget software
program. If it fails or crashes amid transactions, it’s far expected that the gadget
U

would comply with a few sort of algorithm or strategies to recover lost data.
●● failure category
●● to peer in which the problem has happened, we generalize a failure into diverse
ity

classes, as follows −
●● transaction failure
●● a transaction has to abort while it fails to execute or when it reaches a factor from
wherein it may’t pass any in addition. That is called transaction failure wherein only
m

a few transactions or procedures are harm.

●● motives for a transaction failure will be −
)A

●● ogical mistakes − where a transaction can’t entire because it has some code error
or any inner error situation.
●● device errors − where the database machine itself terminates an lively transaction
due to the fact the DBMS isn’t able to execute it, or it has to forestall due to a few
machine situation. As an instance, in case of impasse or resource unavailability,
(c

the system aborts an energetic transaction.

●● there are problems − external to the gadget − that could cause the gadget to
Amity Directorate of Distance & Online Education
 Introduction to Data Base Management Systems 271

forestall all at once and motive the device to crash. For example, interruptions
Notes

e
in strength supply may additionally cause the failure of underlying hardware or
software program failure.

in
●● examples may encompass operating machine errors.
●● disk failure
●● in early days of technology evolution, it became a common problem where difficult-

nl
disk drives or storage drives used to fail regularly.
●● disk disasters include formation of bad sectors, unreachability to the disk, disk
head crash or some other failure, which destroys all or a part of disk storage.

O
●● garage structure
●● we have already described the storage system. In brief, the storage structure can

ty
be divided into two classes −
●● risky garage − because the name suggests, a volatile storage can’t continue to
exist gadget crashes. Unstable garage devices are positioned very close to the
cpu; typically they’re embedded onto the chipset itself. As an instance, most

si
important reminiscence and cache reminiscence are examples of unstable
storage. They’re speedy but can store simplest a small amount of facts.
●● r
on-unstable storage − those reminiscences are made to survive gadget crashes.
They may be massive in statistics garage ability, but slower in accessibility.
ve
Examples might also encompass difficult-disks, magnetic tapes, flash memory,
and non-unstable (battery backed up) ram.
●● whilst a system crashes, it is able to have numerous transactions being carried out
ni

and diverse files opened for them to adjust the records gadgets. Transactions are
fabricated from diverse operations, which might be atomic in nature. However in
keeping with acid properties of DBMS, atomicity of transactions as a whole have to
be maintained, this is, both all the operations are accomplished or none.
U

●● whilst a DBMS recovers from a crash, it must keep the subsequent −

●● it have to take a look at the states of all the transactions, which were being
finished.
ity

●● a transaction can be within the middle of a few operation; the DBMS need to make
certain the atomicity of the transaction in this case.
●● it ought to take a look at whether or not the transaction can be completed now or it
m

needs to be rolled again.

●● no transactions could be allowed to leave the DBMS in an inconsistent country.
●● there are forms of techniques, which could assist a DBMS in improving as well as
)A

keeping the atomicity of a transaction −

●● preserving the logs of each transaction, and writing them onto a few solid garage
before certainly modifying the database.
●● preserving shadow paging, where the changes are achieved on a risky
(c

reminiscence, and later, the real database is updated.

●● log-based healing

Amity Directorate of Distance & Online Education

 272 Introduction to Data Base Management Systems

●● log is a chain of data, which continues the statistics of movements accomplished

Notes

e
with the aid of a transaction. It’s miles critical that the logs are written previous to
the real change and saved on a solid garage media, which is failsafe.

in
●● log-based totally recovery works as follows −
●● the log report is kept on a solid storage media.
●● whilst a transaction enters the device and begins execution, it writes a log

nl
approximately it.
●● <tn, start>

O
●● whilst the transaction modifies an object x, it write logs as follows −
●● <tn, x, v1, v2>
●● it reads tn has modified the cost of x, from v1 to v2.

ty
●● when the transaction finishes, it logs −
●● <tn, commit>

si
●● the database may be changed the usage of two approaches −
●● deferred database modification − all logs are written on to the solid storage and
the database is updated whilst a transaction commits.
●● r
instant database amendment − every log follows an actual database amendment.
ve
That is, the database is modified without delay after every operation.
●● whilst more than one transaction are being performed in parallel, the logs are
interleaved. On the time of recovery, it might end up difficult for the restoration
system to back off all logs, and then begin recuperating. To ease this example,
ni

most modern-day DBMS use the concept of ‘checkpoints’.

●● preserving and maintaining logs in actual time and in actual surroundings may
additionally fill out all the reminiscence space available in the device. As time
U

passes, the log file may also develop too huge to be dealt with at all. Checkpoint
is a mechanism where all the previous logs are eliminated from the device and
stored completely in a garage disk. Checkpoint announces a point earlier than
ity

which the DBMS turned into in regular country, and all of the transactions had
been dedicated.
●● while a gadget with concurrent transactions crashes and recovers, it behaves
within the following manner −
m

●● the healing device reads the logs backwards from the end to the remaining
checkpoint.
●● it maintains two lists, an undo-list and a redo-listing.
)A

●● if the restoration device sees a log with <tn, start> and <tn, commit> or just <tn,
commit>, it puts the transaction within the redo-list.
●● if the recuperation system sees a log with <tn, start> but no dedicate or abort log
found, it puts the transaction in undo-listing.
(c

●● all the transactions inside the undo-listing are then undone and their logs are
removed. All the transactions in the redo-list and their previous logs are eliminated

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 273

after which redone before saving their logs.

Notes

e
●● the concept behind deferred replace is to defer or put off any actual updates to
the database itself until the transaction completes its execution correctly and

in
reaches its commit factor. At some stage in transaction execution, the updates
are recorded only within the log and inside the transaction workspace. After the
transaction reaches its devote factor and the log is pressure-written to disk, the

nl
updates are recorded in the database itself. If a transaction fails earlier than
reaching its devote factor, there is no need to undo any operations, due to the fact
the transaction has no longer affected the database in any manner.

O
●● the steps involved within the deferred update protocol are as follows:
●● whilst a transaction begins, write an entry start_transaction(t) to the log.
●● whilst any operation is achieved with the intention to alternate values in the

ty
database, write a log entry write_item(t, x, old_value, new_value).
●● when a transaction is about to devote, write a log document of the form
dedicate(t); write all log information to disk.

si
●● commit the transaction, the use of the log to put in writing the updates to the
database; the writing of records to disk want now not arise at once.
●● if the transaction aborts, forget about the log statistics and do no longer write the
adjustments to disk. r
ve
●● restoration based totally on deferred update in unmarried-user structures, in
which no concurrency manage is wanted, so that we are able to understand the
recovery system independently of any concurrency manipulate technique. In such
an environment, the restoration algorithm may be instead simple. It really works as
ni

follows.
●● use lists to preserve the transactions: the dedicated transactions listing, which
includes all the dedicated transactions for the reason that ultimate checkpoint,
U

and the active transactions listing (at maximum one transaction falls on this class,
because the gadget is a single-user one). Observe the redo operation to all of the
write_item operations of the committed transactions from the log in the order in
ity

which they were written to the log. Restart the energetic transactions.
●● for a multi-consumer gadget with concurrency control, the restoration process can
be extra complex, relying at the protocols used for concurrency manage. In many
instances, the concurrency manipulate and recovery techniques are interrelated.
In general, the greater the degree of concurrency we wish to acquire, the extra
m

hard the task of recovery turns into.

●● recall a machine wherein concurrency manage uses two-segment locking (primary
)A

2pl) and stops impasse by means of pre-assigning all locks to objects needed
with the aid of a transaction before the transaction begins execution. To mix the
deferred replace methods for healing with this concurrency control technique, we
can maintain all the locks on gadgets in impact until the transaction reaches its
devote factor. After that, the locks may be released. This guarantees strict and
(c

serialisable schedules. Assuming that checkpoint entries are blanketed inside the
log, a possible recovery set of rules for this situation is given underneath.

Amity Directorate of Distance & Online Education

 274 Introduction to Data Base Management Systems

●● use two lists of transactions maintained by the device: the committed transactions
Notes

e
listing which incorporates all committed transactions because the last checkpoint,
and the lively transactions list. Redo all of the write operations of the committed
transactions from the log, in the order in which they have been written into the

in
log. The transactions inside the lively listing which are active and did no longer
dedicate are successfully cancelled and ought to be resubmitted.

nl
●● the redo technique is the same as defined in advance in the deferred replace in
the single-person environment.
●● inside the on the spot replace techniques, the database can be updated by

O
means of the operations of a transaction right away, earlier than the transaction
reaches its devote factor. But, those operations are generally recorded in the go
browsing disk by means of pressure-writing earlier than they’re implemented to the
database, so that recovery is viable.

ty
●● while instant replace is authorized, provisions must be made for undoing the
effect of update operations at the database, because a transaction can fail after it
has implemented some updates to the database itself. Hence recovery schemes

si
primarily based on immediate update need to encompass the functionality to roll
lower back a transaction with the aid of undoing the impact of its write operations.
●● whilst a transaction starts, write an entry start_transaction(t) to the log;
●●
r
while any operation is completed so one can trade values in the database, write a
ve
log access write_item(t, x, old_value, new_value);
●● write the log to disk;
●● as soon as the log document is written, write the update to the database buffers;
ni

●● while convenient, write the database buffers to the disk;

●● when a transaction is about to devote, write a log report of the shape devote(t);
U

●● write the log to disk.

●● we first remember a single-consumer machine in order that we are able to
examine the recuperation procedure separately from concurrency control. If a
failure occurs in a single-consumer device, the executing transaction on the time
ity

of failure may additionally have recorded a few changes within the database. The
impact of all such operations must be undone as part of the recovery procedure.
Consequently, the recuperation algorithm desires an undo system, defined
subsequently, to undo the impact of positive write operations which have been
m

implemented to the database following exam of their device log entry. The healing
algorithm also makes use of the redo procedure described earlier. Healing takes
location within the following way.
)A

●● use two lists of transaction maintained by using the device: the devoted
transactions since the remaining checkpoint, and the energetic transactions (at
maximum one transaction will fall in this category, due to the fact the system is
unmarried consumer). Undo all of the write operations of the lively transaction
from the log, using the undo system described hereafter. Redo all of the write
(c

operations of the devoted transactions from the log, within the order wherein they
had been written inside the log, the usage of the redo system.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 275

●● the undo process is defined as follows:

Notes

e
●● undoing a write operation includes examining its log access write_item(t, x,
old_value, new_value) and placing the value of x in the database to old_value.

in
Undoing a number of such write operations from one or extra transactions from the
log ought to proceed in the reverse order from the order wherein the operations
have been written within the log.

nl
●● when concurrency execution is permitted, the recuperation procedure once more
relies upon at the protocols used for concurrency control. The process below
outlines a recovery technique for concurrent transactions with instantaneous

O
replace. Assume that the log consists of checkpoints and that the concurrency
control protocol produces strict schedules – as, as an instance, the stern 2pl
protocol does. Don’t forget that a strict agenda does not permit a transaction to
read or write an object unless the transaction that remaining wrote the item has

ty
dedicated. However, deadlocks can occur in strict 2pl, as a result requiring undo of
transactions.
●● use lists of transaction maintained by way of the machine: the committed

si
transactions for the reason that closing checkpoint, and the energetic transactions.
Undo all of the write operations of the lively (uncommitted) transaction from the
log, using the undo method. The operations need to be undone inside the reverse
r
of the order in which they were written into the log. Redo all of the write operations
of the devoted transactions from the log, within the order wherein they had been
ve
written inside the log, using the redo technique.
●● in the shadow page scheme, the database is not without delay changed however
a copy, saved on permanent garage (e. G. Disk), is manufactured from the part of
ni

the database to be changed and all changes are made to this copy. Meanwhile,
the old model of the database stays intact. As soon as the transaction commits,
the changed copy replaces the original in an atomic manner, i. E. The replacement
is accomplished in its entirety or in no way. If a machine crashes at this point, the
U

vintage model is still to be had for recuperation.

●● the paging scheme as used in the working device for digital memory management
is appropriate. The reminiscence that is accessed by means of a manner is
ity

referred to as digital memory. Virtual reminiscence is divided into pages that

are all of a positive length . The digital or logical pages are mapped onto bodily
reminiscence blocks of the equal length as the pages. The mapping is achieved
through consulting a page table (or directory). The page desk lists every logical
page identifier and the cope with of the physical blocks that absolutely stores the
m

logical web page. The gain of this scheme is that the consecutive logical pages
need now not be mapped onto consecutive bodily blocks
)A

●● in the shadow page scheme, page tables are used. The original web page desk
(shadow page desk) and the cutting-edge web page desk. To start with, each web
page tables factor to the identical blocks of physical garage. The modern web
page desk is the handiest course thru which a transaction can benefit get entry
to to the statistics saved on disk. That is, a transaction always makes use of the
(c

modern-day page table to retrieve the appropriate database blocks.

●● the advantage of shadow paging is that it makes undoing the impact of the

Amity Directorate of Distance & Online Education

 276 Introduction to Data Base Management Systems

executing transaction quite simple. There’s no want to undo or redo any

Notes

e
transaction operations. In a multi-user surroundings with concurrent transactions,
logs and checkpoints need to be integrated into the shadow paging method. One
downside of shadow paging is that the up to date database pages trade region

in
on disk. This makes it hard to hold related database pages near collectively on
disk with out complex garage control strategies. Furthermore, if the web page
table (directory) is large, the overhead of writing shadow web page tables to disk

nl
as transactions dedicate is significant. A further hassle is the way to take care
of rubbish series while a transaction commits. The vintage pages referenced
by means of the shadow page that has been updated need to be released and

O
delivered to a list of unfastened pages for destiny use. Those pages are not
needed after the transaction commits, and the modern-day web page table
replaces the shadow web page desk to turn out to be the legitimate page table.
Situations which can important a crash healing include:

ty
●● a energy failure at the system, causing the database manager and the database
partitions on it to move down.
●● a hardware failure inclusive of reminiscence, disk, cpu, or community failure.

si
●● a extreme working device mistakes that reasons the database instance to cease
abnormally.
●● r
in case you want crash restoration to be achieved robotically by means of
ve
the database supervisor, enable the automated restart (autorestart) database
configuration parameter via putting it to on. If you do now not want automatic
restart conduct, set the autorestart database configuration parameter to off. As
a end result, you have to issue the restart database command while a database
ni

failure happens. If the database i/o become suspended earlier than the crash
happened, you should specify the write resume option of the restart database
command so as for the crash recovery to retain.
U

●● there are two particular forms of crash recuperation to be aware about: member
crash recovery and organization crash recuperation. Member crash restoration
is the method of recovering a part of a database the usage of the log flow of a
unmarried member after a member failure. Member crash recuperation, which
ity

is generally initiated robotically as part of a member restart, is a web operation-

which means that different contributors can nevertheless get right of entry to the
database. Multiple individuals can be undergoing member crash recovery at the
identical time. Institution crash restoration is the procedure of getting better a
database the use of more than one participants’ log streams after a failure that
m

reasons no possible cluster caching facility to stay within the cluster. Group crash
recovery is likewise commonly initiated mechanically (as part of a group restart)
and the database is inaccessible whilst it’s miles in progress, as with db2 crash
)A

recovery operations outdoor of a database purescale surroundings.

●● if crash recuperation happens on a database that is enabled for
rollforwardrecovery , and an errors occurs for the duration of crash recuperation
that is attributable to an individual desk space, that desk area is taken offline, and
(c

can not be accessed until it is repaired. Crash recovery continues on different desk
areas. At the crowning glory of crash restoration, the other desk spaces inside
the database are on hand, and connections to the database can be set up. But,

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 277

if the table area this is taken offline is the table area that consists of the machine
Notes

e
catalogs, it have to be repaired before any connections are accepted. This conduct
does not follow to database purescale environments. If an blunders happens at
some stage in member crash restoration or organization crash restoration, the

in
crash restoration operation fails.
●● media recuperation requires a control record, information documents (normally

nl
restored from backup), and on line and archived redo log files containing
adjustments because the time the data files had been subsidized up. Media
recuperation is most often used to recover from media failure, consisting of the
lack of a report or disk, or a consumer mistakes, along with the deletion of the

O
contents of a table.
●● media recuperation may be a whole restoration or a point-in-time healing.
Complete recovery can observe to character datafiles, tablespaces, or the entire

ty
database. Factor-in-time restoration applies to the whole database
●● in a complete recovery, you restore backup facts files and apply all modifications
from the archived and online redo log documents to the information documents.

si
The database is lower back to its kingdom at the time of failure and may be
opened and not using a lack of records.
●● in a factor-in-time recuperation, you come back a database to its contents at a
r
consumer-decided on time inside the past. You restore a backup of records files
ve
created before the target time and a entire set of archived redo log documents
from backup advent through the goal time. Restoration applies changes among
the backup time and the goal time to the information documents. All adjustments
after the goal time are discarded.
ni

●● rman enables you to perform each a entire and a factor-in-time restoration of your
database. But, this documentation makes a speciality of whole restoration. An
insider risk is a protection hazard from one of the following 3 assets, each of which
U

has privileged way of entry to the database:

●● a malicious insider with ill-motive
●● a negligent person inside the organisation who exposes the database to assault
ity

through careless moves

●● an interloper who obtains credentials through social engineering or different
methods, or profits get admission to to the database’s credentials
●● an insider risk is one of the most regular causes of database safety breaches and
m

it regularly occurs because plenty of personnel were granted privileged person get
entry to.
●● vulnerable passwords, password sharing, unintentional erasure or corruption of
)A

statistics, and other unwanted person behaviors are nonetheless the reason of just
about 1/2 of data breaches reported.
●● attackers continuously attempt to isolate and target vulnerabilities in software
program, and database management software program is a fairly valuable target.
(c

New vulnerabilities are determined day by day, and all open supply database
management platforms and business database software vendors problem safety

Amity Directorate of Distance & Online Education

 278 Introduction to Data Base Management Systems

patches regularly. However, if you don’t use these patches quickly, your database
Notes

e
might be exposed to assault.
●● even if you do observe patches on time, there is continually the threat of 0-day

in
attacks, when attackers discover a vulnerability, but it has now not yet been found
and patched by using the database supplier.
●● a database-particular chance includes using arbitrary non-sq. And sq. Attack

nl
strings into database queries. Typically, these are queries created as an extension
of internet application bureaucracy, or acquired via http requests. Any database
device is susceptible to those assaults, if builders do no longer adhere to

O
relaxed coding practices, and if the organisation does now not perform ordinary
vulnerability testing.
●● buffer overflow takes location whilst a procedure attempts to put in writing a big
quantity of statistics to a fixed-length block of reminiscence, extra than it’s miles

ty
authorised to hold. Attackers might use the extra data, stored in adjacent memory
addresses, as the starting point from which to launch assaults.
●● in a denial of service (dos) attack, the cybercriminal overwhelms the goal carrier—

si
on this example the database server—the usage of a huge amount of faux
requests. The result is that the server can not perform real requests from actual
customers, and frequently crashes or becomes risky.
●●
r
in a disbursed denial of carrier assault (ddos), faux site visitors is generated by
ve
a large number of computer systems, collaborating in a botnet managed via the
attacker. This generates very huge site visitors volumes, which might be hard to
stop without a fantastically scalable defensive architecture. Cloud-based ddos
safety services can scale up dynamically to address very big ddos attacks.
ni

●● malware is software program written to take advantage of vulnerabilities or to

cause harm to a database. Malware may want to arrive thru any endpoint tool
related to the database’s network. Malware safety is essential on any endpoint, but
U

specifically so on database servers, because of their high price and sensitivity. The
evolving it surroundings is making databases extra at risk of threats. Right here
are traits that can cause new forms of attacks on databases, or may additionally
require new protecting measures:
ity

●● developing statistics volumes—storage, statistics seize, and processing is

developing exponentially throughout almost all companies. Any statistics security
practices or gear need to be highly scalable to deal with remote and close to-future
necessities.
m

●● dispensed infrastructure—community environments are increasing in complexity,

specifically as businesses transfer workloads to hybrid cloud or multi-cloud
architectures, making the deployment, management, and preference of protection
)A

answers greater tough.

●● increasingly tight regulatory requirements—the global regulatory compliance
landscape is growing in complexity, so following all mandates are becoming more
challenging.
(c

●● cybersecurity capabilities scarcity—there is a global shortage of skilled

cybersecurity professionals, and companies are finding it tough to fill security

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 279

roles. This may make it more difficult to guard crucial infrastructure, along with
Notes

e
databases. The following safety features are recommended:
●● robust passwords have to be enforced

in
●● password hashes need to be salted and stored encrypted
●● debts have to be locked following multiple login attempts

nl
●● money owed ought to be regularly reviewed and deactivated if staff move to
specific roles, go away the organisation, or not require the identical level of get
right of entry to

O
Questions & sporting events
1. How to comfortable your database server?
2. Provide an explanation for kinds of database security cyber threats.

ty
3. Give an explanation for shadow paging and page control.
4. What are the healing techniques primarily based on instant update?

si
5. What are the restoration strategies primarily based on deferred replace?

Glossary

crash recuperation
r
ve
DBMS is a particularly complicated system with hundreds of transactions being
achieved each 2nd. The sturdiness and robustness of a DBMS relies upon on its
complicated structure and its underlying hardware and system software. If it fails or
ni

crashes amid transactions, it’s miles anticipated that the machine might observe a few
sort of algorithm or techniques to recover lost statistics. Failure classification

to peer in which the problem has befell, we generalize a failure into numerous
U

classes, as follows −

transaction failure
a transaction has to abort while it fails to execute or while it reaches a factor from in
ity

which it is able to’t pass any similarly. This is known as transaction failure where only a
few transactions or procedures are hurt. Motives for a transaction failure could be −

●● logical mistakes − where a transaction can’t whole because it has some code
blunders or any inner errors condition.
m

●● machine errors − where the database system itself terminates a lively transaction
due to the fact the DBMS isn’t always capable of execute it, or it has to stop due to
)A

some gadget situation. As an example, in case of impasse or aid unavailability, the

device aborts an active transaction. System crash
there are problems − external to the gadget − that may purpose the device to
forestall and motive the machine to crash. For example, interruptions in power supply
may cause the failure of underlying hardware or software failure. Examples might also
(c

include running system errors. Disk failure

Amity Directorate of Distance & Online Education

 280 Introduction to Data Base Management Systems

in early days of technology evolution, it became a common trouble wherein tough-

Notes

e
disk drives or garage drives used to fail frequently. Disk screw ups consist of formation
of horrific sectors, unreachability to the disk, disk head crash or every other failure,
which destroys all or a part of disk storage. Storage structure

in
we’ve already defined the storage gadget. In brief, the storage shape can be
divided into categories −

nl
●● risky storage − as the name suggests, a volatile garage cannot survive system
crashes. Risky garage gadgets are placed very near the cpu; commonly they’re
embedded onto the chipset itself. As an instance, fundamental reminiscence and

O
cache memory are examples of risky storage. They may be speedy but can save
only a small amount of records.
●● non-risky storage − those memories are made to live on system crashes. They’re

ty
massive in records garage capacity, however slower in accessibility. Examples
may encompass hard-disks, magnetic tapes, flash reminiscence, and non-volatile
(battery backed up) ram. Log-primarily based restoration
log is a series of data, which continues the records of moves finished through a

si
transaction. It’s miles critical that the logs are written previous to the actual change and
stored on a stable storage media, which is failsafe. Log-based totally recovery works as
follows −

●●
r
the log record is saved on a strong garage media.
ve
●● while a transaction enters the gadget and starts off evolved execution, it writes a
log approximately it.
<tn, start>
ni

●● whilst the transaction modifies an object x, it write logs as follows −

<tn, x, v1, v2>
U

it reads tn has modified the fee of x, from v1 to v2.

●● when the transaction finishes, it logs −

<tn, commit>
ity

the database may be changed the usage of two approaches −

●● deferred database change − all logs are written directly to the solid storage and
the database is up to date whilst a transaction commits.
m

●● instant database modification − each log follows an real database change. That is,
the database is modified at once after every operation.
)A

●● Checkpoint retaining and keeping logs in actual time and in actual surroundings
may additionally fill out all the reminiscence space to be had in the machine.
As time passes, the log file might also develop too large to be handled at all.
Checkpoint is a mechanism wherein all of the preceding logs are removed from
the gadget and stored permanently in a storage disk. Checkpoint pronounces
(c

a point earlier than which the DBMS become in regular nation, and all the
transactions were committed. Healing the recovery gadget reads the logs
backwards from the end to the final checkpoint.

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 281

●● it maintains lists, an undo-listing and a redo-listing.

Notes

e
●● if the restoration system sees a log with <tn, start> and <tn, commit> or just <tn,
commit>, it puts the transaction inside the redo-list.

in
●● if the healing device sees a log with <tn, start> but no devote or abort log
discovered, it puts the transaction in undo-listing. All the transactions within the
undo-list are then undone and their logs are eliminated. All the transactions in the

nl
redo-list and their preceding logs are removed and then redone before saving their
logs. Deferred update
the idea in the back of deferred update is to defer or put off any actual updates

O
to the database itself until the transaction completes its execution efficaciously and
reaches its dedicate factor. At some point of transaction execution, the updates are
recorded simplest in the log and within the transaction workspace. After the transaction
reaches its devote point and the log is pressure-written to disk, the updates are

ty
recorded inside the database itself. If a transaction fails earlier than attaining its commit
factor, there may be no need to undo any operations, due to the fact the transaction has
not affected the database in any way. The steps concerned within the deferred update

si
protocol are as follows:

◌◌ While a transaction starts, write an entry start_transaction(t) to the log.

◌◌
r
When any operation is performed in an effort to trade values within the
database, write a log access write_item(t, x, old_value, new_value).
ve
◌◌ Whilst a transaction is ready to commit, write a log file of the form dedicate(t);
write all log records to disk.
◌◌ Devote the transaction, the usage of the log to put in writing the updates to the
database; the writing of statistics to disk need not occur straight away.
ni

◌◌ If the transaction aborts, forget about the log information and do now not write
the adjustments to disk. Instant replace
U

Within the instant replace strategies, the database can be up to date through the
operations of a transaction right away, earlier than the transaction reaches its dedicate
point. But, these operations are commonly recorded within the go browsing disk by
means of force-writing earlier than they may be applied to the database, in order that
ity

recovery is viable. When on the spot replace is allowed, provisions should be made for
undoing the effect of update operations at the database, because a transaction can fail
after it has carried out some updates to the database itself. Subsequently restoration
schemes primarily based on immediately replace ought to encompass the functionality
to roll again a transaction by undoing the impact of its write operations.
m

When a transaction begins, write an entry start_transaction(t) to the log;

◌◌ When any operation is achieved with a view to trade values in the database,
)A

write a log entry write_item(t, x, old_value, new_value);

◌◌ Write the log to disk;
◌◌ As soon as the log record is written, write the update to the database buffers;
◌◌ Whilst convenient, write the database buffers to the disk;
(c

◌◌ Whilst a transaction is ready to devote, write a log document of the form

commit(t);

Amity Directorate of Distance & Online Education

 282 Introduction to Data Base Management Systems

◌◌ Write the log to disk. Shadow paging

Notes

e
inside the shadow web page scheme, the database isn’t without delay changed but
a replica, saved on permanent garage (e. G. Disk), is manufactured from the part of

in
the database to be modified and all changes are made to this copy. In the meantime,
the vintage model of the database remains intact. As soon as the transaction commits,
the modified copy replaces the original in an atomic way, i. E. The substitute is done in
its entirety or in no way. If a device crashes at this point, the old version continues to

nl
be available for recuperation. Web page control the paging scheme as used inside the
working machine for virtual reminiscence control is suitable. The reminiscence this is
accessed by a technique (a software in execution is a process) is referred to as digital

O
memory. Digital memory is split into pages that are all of a certain length (commonly
4096 bytes or 4k).

The digital or logical pages are mapped onto physical reminiscence blocks (i. E.

ty
Disk bodily facts) of the identical size as the pages. The mapping is carried out by using
consulting a page desk (or listing). The page table lists each logical page identifier and
the deal with of the bodily blocks that absolutely shops the logical page. The benefit
of this scheme is that the consecutive logical pages want no longer be mapped onto

si
consecutive bodily blocks. Media recuperation an error can get up while trying to write
or study a document that is required to perform an oracle database. This incidence is
known as media failure due to the fact there’s a bodily hassle analyzing or writing to
r
files on the storage medium. In case you restore the archived redo log files and records
ve
files, then you definitely need to carry out media recuperation earlier than you may
open the database. Any database transactions within the archived redo log files now
not meditated inside the records files are applied to the data documents, bringing them
to a transaction-constant country before the database is opened. Media recuperation
ni

calls for a manipulate document, facts documents (commonly restored from backup),
and on line and archived redo log documents containing adjustments for the reason
that time the data documents had been subsidized up. Media healing is most often
U

used to get over media failure, consisting of the loss of a file or disk, or a person error,
which includes the deletion of the contents of a desk. Media recovery may be a whole
recuperation or a point-in-time recuperation.

Entire recovery can practice to individual datafiles, tablespaces, or the complete

ity

database. Factor-in-time healing applies to the complete database in a complete

recovery, you restore backup information documents and observe all adjustments from
the archived and on-line redo log documents to the data files. The database is again
to its state on the time of failure and may be opened and not using a loss of statistics.
m

In a point-in-time healing, you return a database to its contents at a person-decided

on time in the beyond. You repair a backup of statistics files created earlier than the
goal time and a complete set of archived redo log files from backup advent through
the target time. Recuperation applies modifications among the backup time and the
)A

goal time to the records files. All adjustments after the target time are discarded. Rman
permits you to carry out each a complete and a factor-in-time healing of your database.
But, this documentation focuses on complete restoration. Database safety database
protection consists of a selection of measures used to secure database management
(c

systems from malicious cyber-attacks and illegitimate use. Database safety programs
are designed to protect not handiest the statistics inside the database, but also the facts
control gadget itself, and each software that accesses it, from misuse, damage, and

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 283

intrusion. Database protection encompasses equipment, processes, and methodologies

Notes

e
which set up security inside a database surroundings. Database security threats many
software vulnerabilities, misconfigurations, or styles of misuse or carelessness ought to
result in breaches. Here are some of the most regarded causes and kinds of database

in
safety cyber threats. Insider threats an insider hazard is a protection hazard from one
of the following three sources, every of which has privileged method of access to the
database:

nl
◌◌ a malicious insider with unwell-motive
◌◌ a negligent character inside the business enterprise who exposes the

O
database to assault thru careless moves
◌◌ an outsider who obtains credentials thru social engineering or different
techniques, or profits get admission to the database’s credentials
an insider hazard is one of the maximum typical reasons of database safety

ty
breaches and it regularly happens due to the fact a whole lot of employees have been
granted privileged user get entry to. Human errors susceptible passwords, password
sharing, accidental erasure or corruption of statistics, and other undesirable person

si
behaviors are nonetheless the purpose of just about half of facts breaches pronounced.
Exploitation of database software program vulnerabilities attackers continuously attempt
to isolate and target vulnerabilities in software program, and database management
r
software program is a exceedingly treasured goal. New vulnerabilities are observed
each day, and all open supply database management structures and commercial
ve
database software program vendors trouble security patches regularly.

But, in case you don’t use these patches fast, your database might be uncovered
to attack. Even in case you do apply patches on time, there is usually the chance of
ni

zero-day assaults, when attackers discover a vulnerability, but it has not yet been
discovered and patched by means of the database vendor. Square/nosql injection
attacks a database-unique threat includes the usage of arbitrary non-square and
U

square assault strings into database queries. Typically, these are queries created as an
extension of web utility bureaucracy, or received through http requests. Any database
system is susceptible to those assaults, if builders do no longer adhere to relaxed
coding practices, and if the corporation does now not carry out regular vulnerability
ity

testing. Buffer overflow assaults buffer overflow takes location whilst a procedure
attempts to write down a large quantity of records to a set-period block of memory,
greater than it’s far accredited to preserve.

Attackers may use the extra information, stored in adjoining reminiscence

m

addresses, because the place to begin from which to launch assaults. Denial of
provider (dos/ddos) attacks in a denial of provider (dos) assault, the cybercriminal
overwhelms the goal service—in this instance the database server—using a massive
)A

amount of fake requests. The end result is that the server cannot carry out proper
requests from actual customers, and frequently crashes or becomes volatile. In a
disbursed denial of carrier attack (ddos), fake visitors are generated through a large
variety of computer systems, taking part in a botnet controlled by means of the attacker.
This generates very huge site visitors volumes, that are hard to prevent without a
(c

distinctly scalable shielding structure. Cloud-based totally ddos safety services can
scale up dynamically to deal with very massive ddos assaults. Malware malware is
software written to take advantage of vulnerabilities or to motive damage to a database.

Amity Directorate of Distance & Online Education

 284 Introduction to Data Base Management Systems

Malware could arrive via any endpoint tool related to the database’s network. Malware
Notes

e
safety is essential on any endpoint, but particularly so on database servers, because of
their high value and sensitivity. An evolving it surroundings the evolving it environment
is making databases extra vulnerable to threats. Here are developments that can lead

in
to new forms of assaults on databases, or can also require new protective measures:

◌◌ growing data volumes—garage, records capture, and processing is

nl
developing exponentially across nearly all companies. Any data protection
practices or equipment ought to be quite scalable to cope with remote and
close to-destiny necessities.

O
◌◌ distributed infrastructure—network environments are growing in complexity,
in particular as corporations switch workloads to hybrid cloud or multi-cloud
architectures, making the deployment, control, and choice of protection
answers greater difficult.

ty
◌◌ an increasing number of tight regulatory necessities—the global regulatory
compliance landscape is developing in complexity, so following all mandates
are getting extra difficult.

si
◌◌ cybersecurity talents scarcity—there may be a international shortage of skilled
cybersecurity professionals, and groups are finding it hard to fill security roles.
This could make it more difficult to protect vital infrastructure, which includes
r
databases. A way to comfortable your database server? A database server is
a physical or digital gadget going for walks the database. Securing a database
ve
server, additionally called “hardening”, is a procedure that includes bodily
security, network protection, and at ease running device configuration. Ensure
bodily database safety
ni

chorus from sharing a server for web programs and database programs, in
case your database incorporates sensitive statistics. Despite the fact that it could be
inexpensive, and simpler, to host your web page and database together on a hosting
company, you are putting the safety of your records in someone else’s fingers. In case
U

you do depend on an internet website hosting carrier to control your database, you
ought to make sure that it’s far a organisation with a strong security music document.
It’s miles satisfactory to live clean of loose website hosting offerings due to the feasible
lack of protection. If you manipulate your database in an on-premise records center,
ity

remember that your records center is likewise liable to attacks from outsiders or insider
threats. Make sure you have bodily security measures, which includes locks, cameras,
and safety employees to your bodily facility. Any get entry to to physical servers must be
logged and most effective granted to legal people.
m

In addition, do not leave database backups in locations that are publicly available,
consisting of brief walls, internet folders, or unsecured cloud storage buckets. Lock
down accounts and privileges it is crucial to make certain that every privileged account
)A

on a database server is configured with a strong, particular password. If accounts

are not wished, they ought to be expired and locked. For the closing accounts,
get admission to has to be constrained to the absolute minimum required. Every
account needs to handiest have get right of entry to to the tables and operations (as
an example, choose or insert) required via the user. Keep away from growing person
(c

debts with access to every desk inside the database. Regularly patch database servers
make sure that patches stay modern-day. Effective database patch control is a essential

Amity Directorate of Distance & Online Education

 Introduction to Data Base Management Systems 285

protection exercise because attackers are actively searching out new security flaws in
Notes

e
databases, and new viruses and malware seem on a each day foundation. A well-timed
deployment of updated variations of database provider packs, essential protection
hotfixes, and cumulative updates will improve the steadiness of database overall

in
performance. Disable public network get right of entry to groups shop their programs in
databases.

nl
In most real-world situations, the cease-user doesn’t require direct get right of entry
to to the database. Accordingly, you ought to block all public community get admission
to to database servers except you’re a web hosting company. Preferably, a company
should set up gateway servers (vpn or ssh tunnels) for faraway administrators. Encrypt

O
all files and backups Regardless of how stable your defences are, there is usually a
possibility that a hacker may additionally infiltrate your gadget. But attackers aren’t
the most effective danger to the safety of your database. Your employees may pose a

ty
threat to your commercial enterprise. There may be continually the opportunity that a
malicious or careless insider will gain access to a document they don’t have permission
to get admission to. Encrypting your statistics makes it unreadable to each attacker and
personnel. Without an encryption key, they cannot get right of entry to it, this presents a

si
closing line of protection towards unwelcome intrusions.

Encrypt all-essential software files, information documents, and backups so that

r
unauthorized users cannot read your essential information. Database security best
practices here are numerous quality practices you may use to enhance the safety of
ve
sensitive databases. Actively manipulate passwords and person get entry to when
you have a large employer, you must reflect on consideration on automating get
entry to management thru password control or access control software. This can offer
accredited customers with a quick-time period password with the rights they want on
ni

every occasion they need to gain get right of entry to a database. It also continues
song of the activities finished at some point of that point frame and stops administrators
from sharing passwords. While directors may also sense that sharing passwords is
U

convenient, however, doing so makes effective database responsibility and security

nearly impossible. The subsequent security measures are encouraged:

◌◌ sturdy passwords should be enforced

ity

◌◌ password hashes ought to be salted and stored encrypted

◌◌ bills need to be locked following a couple of login tries
◌◌ debts should be regularly reviewed and deactivated if workforce flow to
exclusive roles, leave the organization, or now not require the identical level of
m

access

Test your database protection

)A

As soon as you’ve got put in area your database security infrastructure, you
must test it against a real threat. Auditing or performing penetration assessments
against your very own database will assist you get into the mind-set of a cybercriminal
and isolate any vulnerabilities you could have omitted. To make certain the check is
comprehensive, contain moral hackers or identified penetration checking out services
(c

on your safety checking out. Penetration testers provide massive reports listing
database vulnerabilities, and it’s miles crucial to quick inspect and remediate these
vulnerabilities. Run a penetration test on a critical database gadget as a minimum once

Amity Directorate of Distance & Online Education

 286 Introduction to Data Base Management Systems

consistent with year. Use real-time database tracking always scanning your database
Notes

e
for breach attempts will increase your protection and lets you unexpectedly react to
possible assaults. In particular, document integrity monitoring (fim) permit you to log
all actions carried out on the database’s server and to alert you of ability breaches.

in
When fim detects a alternate to vital database files, make certain safety groups are
alerted and able to analyze and reply to the risk. Use internet application and database
firewalls you should use a firewall to defend your database server from database

nl
security threats. By means of default, a firewall does now not allow get admission to to
visitors.

It wishes to also forestall your database from beginning outbound connections

O
unless there is a particular reason for doing so. In addition to safeguarding the
database with a firewall, you have to deploy a web software firewall (waf). That is
because assaults aimed at web packages, inclusive of square injection, can be used

ty
to advantage illicit get entry to in your databases. A database firewall will no longer stop
maximum internet application attacks, because conventional firewalls function at the
community layer, at the same time as net application layers perform at the application
layer . A waf operates at layer 7 and is capable of discover malicious net application

si
visitors, consisting of sq. Injection assaults, and block it earlier than it can harm your
database.

Further Readings r
ve
1. Carlos Coronel and Steven Morris. (2018). Database Systems: Design,
Implementation and Management. Thirteenth Edition. Course Technology, Inc.
2. Hectar Garcia-Molina, Jeff Ullman and Jennifer Widom. (2009). Database
Systems: The Complete Book. Second Edition. Pearson Education Inc.
ni

3. Thomas Connolly and Carolyn Begg. (2014). Database Systems: A Practical

Approach to Design, Implementation and Management. Sixth Edition. Pearson
Education, Inc.
U

Check your Understanding

A. Fill in the Blanks
ity

1. volatile
2. Checkpoint
3. recovery
m

4. REDO
5. RMAN
B. True or False
)A

1. True
2. False
3. True
(c

4. False
5. True

Amity Directorate of Distance & Online Education