DB2

Introduction to DB2
Topics

• Introduction to DB2
o DBMS concepts
o Relational model
o E-R Diagrams
o Data normalization techniques
Introduction to DB2
Comparing databases with files

• Some limitations of file based data storage

• Sharing of data within the file concurrently with many people
• Security of data
• Data redundancy
• Why a database?
o To solve some of the problem, found in conventional file processing
o Database concept solves the problem of
• Duplication of Data and effort
• Programmers need to know the exact file structure
• Programs needs to be changed
o Databases are used to meet four main data-related goals
• Increasing data independence
• Reducing data redundancy
• Increasing security
• Maintaining data integrity
Database (contd.)

• Information & Data

o When information is entered and stored in a computer, it is referred as
data.
o After processing the data (such as formatting or printing), output data can
again be perceived as information
o Information is something that can be understood with a context whereas
data is just characters.
• Database
o Collection of data that is organized in a way, so that it can easily be
accessed, managed, and modified.
• DBMS
o A database management system (DBMS) is a software application that
interacts with the user, other applications, and the database itself to
capture and analyze data.
o A general-purpose DBMS is designed to allow the definition, creation,
querying, update, and administration of databases.
Database

• Data independence
o If the format of the data is changed or re-organized, the program dependent
on that data must also be changed
o for example, suppose you want to add city and state to an address field, then
the program must be modified even though it may not need the newly added
fields.
o In a database management system,
• the data and program are independent
• a change in data organization will not usually affect the user programs
• users do not have to worry about how the data is actually stored
• users can focus on the particular part of the data they want to access with an
application program
• user programs manipulate the data only, not the organization of the data
Database (contd.)

• Data redundancy
o In the conventional file system, the same data may be stored in many files to
meet needs of different applications.
• Therefore, some of the data is redundant.
o For example, with traditional file processing it is possible to find a customer’s
address repeated in many files.
• Customer Record System
• Purchase Order System
• Account Receivable System
o This redundancy in conventional file processing system could cause problem
when the address changes.
• address should be modified in all these files leads to duplication of efforts and
more opportunity for update anomaly
o In a database system, different applications share the same centralized data.
• A database system is able to reduce data redundancy by storing each element in
only one location.
Database (contd.)

• Data Security
o Databases often uses database security manager for protection.
o There are restrictions on how an user can access the data itself
o Most databases have security features that authorize access to the data and
places restrictions on the operations that may be performed on the data.
• Such features and restrictions enable the database to meet another of its goals –
to increase data security.
• Data Integrity
o Every one who uses the database is responsible for the integrity of the data.
o Maintaining integrity of the data is very important in a database system
because the database is shared by many users.
o Without integrity the data could become invalid or unreliable.
Database (contd.)

• Characteristics of a database
o Program-Data Independence
• Immunity of applications to change in the storage structure and access strategy of
data
o Database should be a collection of data which has no redundancy
o Data Integrity
• No inconsistency while maintaining data
o Data protection
o Dynamic definition
• e.g. add a column
Database (contd.)

• In a DBMS environment, there are two functionalities

o Design, develop and maintain the database
o Application program that uses the database
• DBMS functions
o Define data
o Read/Write data
o Write audit trail
o Provide recovery and backup
o Provide data security, data independence, multiple user access
o Avoid data redundancy
• Application user functions
o Logically access and process data
Database (contd.)

• Hierarchical model
o Top down structure
o Parent child relationship
o Used even today in mainframe applications
• example – IMS
• Network model
o No concept of parent/child
o Any record type can be associated with any number of arbitrary record
types
o Not a very successful model
• example – IDMS
• Relational model
o Data stored in the form of tables
• Consists of multiple rows and columns.
• examples – Oracle, DB2 (UDB), mySQL, SQL Server
Database (contd.)

• Problems with non-relational systems

o Programmer must know the database structure
o Limited flexibility for change
o Connections maintained by external pointers
o Many-to-many relationship problem
o Structures can be complex, making programming complex
Database (contd.)

• RDBMS
o A relational database is a system that is structured in sets of two dimensional
tables
o DB2’s simplicity is automatic navigation.
• Navigation is to simply determine the path to the data
o In some non-relational systems, when you want to retrieve data you must
specify
• What data you want to retrieve
• How to find it
o In relational database user can concentrate their business functions, instead
of wrestling with data processing problems, making them more productive
o User can easily interact with system with queries.
o Ease of use and power of SQL.
• Concern only for WHAT data is needed, not HOW do to get the data.
Database (contd.)

• Advantages of RDBMS
o Data independence (logical/physical)
o Avoiding data redundancy
o Provide multiple views
o Allow sharing of data
o Provide facilities to search quickly
o Provide good security & control over data
o Provide methods for easier backup & recovery
o Provide maximum concurrency without compromising on data integrity
Database (contd.)

• Relational terms
o Candidate key
• Some attribute (or a set of attributes) that may uniquely identify each row(tuple)
in the relation(table)
o Primary key
• The candidate key that is chosen for primary attributes to uniquely identify each
row.
o Alternate key
• Candidate keys that were not chosen as primary key can be alternate key.
o Foreign key
• An attribute of one relation that might be a primary key of another relation
Database (contd.)

• Data structure (Relational)

o All data in the relational model is organized into two dimensional TABLES
called RELATION.
o The rows of such tables are referred to as tuples.
o Columns are usually referred to as attributes of a relation.
• All entries in a column must be of the same kind (domain)
• All columns must be assigned with distinct values by name.
• Ordering of columns is not significant (for SQL processing).
o Each row in a relation must be distinct
• i.e. duplicate rows are not allowed in a RELATION.
• NULL (No value) is not supported for primary key
o Each column/row intersection(cell) in a relation should contain a single value
• i.e. values in a relation must be atomic.
o Sequence of rows is insignificant
Database (contd.)

• Relation
o In general tables are referred to as RELATIONS
o The relational approach to data is based on the realization that, files that
obey certain constraints may be considered as mathematical relations.
• Hence the term RELATION.
o A relational database is a collection of tables.
• Relational model components
o Data structure
o Data integrity
o Data manipulation
Database (contd.)

• Data integrity
o Refers to accuracy and validity of data in database.
o DB2’s security features ensure that data integrity is maintained and that
access to the data is authorized.
o Entity Integrity
• Primary key
o Declarative Integrity
• Declarations of columns
o Referential Integrity
• Foreign Key
Database (contd.)

• Entity Integrity
o A logical consequence of the definition of a primary key is that it guarantees
the uniqueness of each occurrence.
o No attribute participating in the primary key of a base relation is allowed to
accept NULLS.
o An entity with a primary key defined to it satisfies the entity integrity rules.
• Declarative integrity (Also called as Domain integrity)
o User defined constraints for a column value that is enforced by the DBMS
Database (contd.)

• Referential constraint
o A rule which enforces the relationship between a parent table and a child
table such that a non-null value of the foreign key must match a value of the
associated primary key.
• Referential Integrity
o The automatic enforcement of the referential constraint is called the
referential integrity.
o Referential integrity is used to define what will happen when the integrity of
the references defined for a particular database is encountered by a
DELETE, INSERT or UPDATE operations
Data Modeling
Data Modeling (contd.)

• Primary key
o It identifies its own row
o It must be unique in the entire table.
o Cannot be NULL
o Only one Primary Key can be defined in a table
• Related tables
o A row in one table carry KEY of another table
o It identifies a row of data.
• Foreign key
o It identifies the row of related data either in the same table or another table.
• But referenced column should be primary key
• Examples
o EMPT,DEPT
o CUSTOMER, ORDER, ORDER DETAILS, PRODUCT
Data Modeling (contd.)

• Table types
o The table containing the primary key which is related to a foreign key of a
dependent table is called parent table
o The table containing the foreign key is called dependent table
o The dependent of a dependent table is called descendant table
• Entity
o A business entity is something that is fundamental to the organization and
an individual instance or occurrence of this thing can be uniquely
identified.
o The relation between business entities can be
• One to One
• One to Many
• Many to Many
Data Modeling

• Business Modeling
o Business model is a formal representation of business information – its
object, the object's properties or attributes and the relationship between one
another
o Business modeling does not depend upon the underlying database
structure or a specific application.
• Instead the business model can span an entire enterprise or a division of the
organization
o It requires a solid understanding of the business and can serve as a
verification of the user's view of the business before the database is even
designed
Data Modeling (contd.)

• E-R Model
o Converting Business entities to data entities consists of turning business
entities into data entities and then determining the attributes or properties of
each entities.
o E-R model is a logical representation of data for a business area
• Business entity becomes data entity
• Represented as entities, relationship between entities and attributes of both
relationships and entities.
• E-R models are conceptual data models expressed in the form of an E-R
Diagram
o Steps in modeling
• Identify Primary and Foreign Keys
• Resolve many to many relationships
• Normalize the data design
Data Modeling (contd.)

• Example:
o CUSTOMER places ORDERS
o ORDERS have PRODUCTS
• Each order relates to only one customer (one-to-one)
• A customer can place any number of orders (one-to-many)
• Many orders can contain many products (many-to-many)
o A product can be a part of many orders
o An order can have many products
• Customer, Order & Product are called ENTITIES.
• An Entity may transform into one or more tables
• The unique identity for information stored in an ENTITY is called a
PRIMARY KEY.
o e.g... CUSTOMER-ID uniquely identifies each customer
Data Modeling (contd.)

• Relationships
o For e.g.. CUSTOMER is related to ORDERS through ‘ORDER_NO’ which is
the foreign-key in CUSTOMER and primary key in another ORDER table.
o Relationships transform into foreign keys.
o As per the relational integrity the primary key, ORDER_NO, for the table
‘ORDER’ can never be NULL, while it can be NULL in the table
‘CUSTOMER’.
• Since ORDER_NO is not a primary key in the CUSTOMER Table, it can be
allowed to have nulls, though practically it is rare
Data Modeling (contd.)

• E-R Diagrams
o There are many parts in a business model and there is no formal definition
of what part must exist to be a valid model.
o It can be as simple as just naming the major business entities or objects
and their relationship to each other.
o This is commonly manifested in an Entity Relationship Diagram
o Entity Relationship diagram is a pictorial representation of the user's view of
the business representation
Data Modeling (contd.)

• One to many relationship

o One Employee should work in only one department

EMP WORKDEPT

EMP WORKDEPT DEPT

EMP WORKDEPT
Data Modeling (contd.)

• Many to one relationship

o One department can have many employees

EMP WORKDEPT

EMP WORKDEPT DEPT

EMP WORKDEPT
Data Modeling (contd.)

• Many to many relationship

o Many employees work in many projects

MANY

EMP PROJ

EMP PROJ

EMP PROJ

MANY
Data Modeling (contd.)

• An E-R Diagram

1 Department M
DEPT has many LOCATION
locations

M M M
Many M
Employees
EMPL works on PROJECT
many projects
M M

SKILLS
Normalization
Normalization

• Data normalization
o Data Normalization is a step-by-step technique of analyzing data into its
constituent entities and attributes.
o Removes functional dependencies amongst entities.
o Eliminates redundancy.
o Prevents update anomalies and data inconsistencies.
o Necessary for transactional data
• De-normalization
o Reverse of normalization
o Useful for 'read-only' data with heavy SQL SELECT access for reports
Normalization (contd.)

• There are many levels in Normalization

o Un normalized entity
o First normal form
• Remove repeating groups
o Second normal form
• Remove partial key dependencies
o Third normal form
• Remove indirect or transitive dependencies
o Higher levels of normalization leads to more complexity
• Trade off between redundancy and easier access
Normalization (contd.)

• Consider an entity PURCHASE-ORDER represented as follows

• PURCHASE-ORDER
o (order-no, cust-no, cust-address, order date, delivery-date, {book-no, book-
title, book-price, qty-ordered, price}, total-price)
o The attribute type order-no is the entity identifier.
o The inner set of brackets signifies that the attribute types enclosed in them
form a repeating group.
Normalization (contd.)

• Un-normalized data

Order Number : 1101

Customer Number : 1234
Customer Address : 6, Smith Road, Chennai, 600 002
Order Date : 20/03/2003
Delivery Date : 20/04/2004

Book No Book Title Price Qty Total

7100 COBOL Handbook 155.00 2 310.00

3990 LOTUS 1-2-3 80.00 5 400.00
3667 dBASE-IV 100.00 4 400.00

1110.00
Normalization (contd.)

• First normal form

o Remove any repeating groups of attribute types.
o Rewrite them as new entities.
o In order to link the derived entities to the original entity the identifier of the
original entity must be part of the derived entities.
o Remove repeating groups
Here The repeating group is
{ order-no,book-no, book-title, book-price, Qty-ordered, price}

The 1NF of the entity PURCHASE-ORDER now becomes :

o PURCHASE-ORDER (order-no,cust-no,cust-address,order-date, delivery-

date )

PURCHASE-ITEM (order-no,book-no, book-title, book-price, Qty-ordered,

price)
Normalization (contd.)

• First normal form

PURCHASE – ORDER
ORDER-NO CUST-NO CUST-ADDRESS ORDER-DATE DELIVERY-DATE TOTAL-PRICE
1101 1234 6,Smith Road 20/03/2003 20/04/2004 1100.00
1103 1254 7,ITC 12/12/2004 18/12/2004 550.00
1104 1789 10/A UK Council 01/04/2004 08/04/2004 100.00

PURCHASE-ITEM
ORDER-NO BOOK-NO BOOK-TITLE BOOK-PRICE QTY PRICE
1101 7100 COBOL Hand Book 55.00 2 310.00
1101 3990 LOTUS 1-2-3 80.00 5 400.00
1101 3667 DBASE - IV 100.00 4 400.00
1103 7100 COBOL Hand Book 155.00 2 310.00
1103 3900 LOTUS 1-2-3 80.00 3 240.00
1104 3667 DBASE – IV 100.00 1 100.00
Normalization (contd.)

• Second normal form

o For an entity type to be in 2NF, it must already be in 1NF and every non-
identifying attribute type must be fully functionally dependent on the
identifier.
o Consider the purchase table.
• Order-no and Book-no constitute the composite primary key.
• But the book title and book-price does not depend on Order-no.
• The Book-title and Book-price primarily depend on Book-no.
• So it has to be eliminated
o PURCHASE-ORDER has only one attribute type as identifier and is
therefore already in 2NF.
o In PURCHASE-ITEM, Book-title & Book-price are partially dependent on the
full identifier and therefore not in 2NF.
Normalization (contd.)

• Second normal form

o PURCHASE-ORDER
• (order-no, cust-no, cust-address, order-date, delivery-date, total-price)
o PURCHASE-ITEM
• (order-no, book-no, qty-ordered, price)
o BOOK
• (book-no, book-title, book-price)
Normalization (contd.)
PURCHASE – ORDER

ORDER-NO CUST-NO CUST-ADDRESS ORDER-DATE DELIVERY-DATE TOTAL-PRICE

1101 1234 6,Smith Road 20/03/2003 20/04/2004 1100.00
1103 1254 7,ITC 12/12/2004 18/12/2004 550.00
1104 1789 10/A UK Council 01/04/2004 08/04/2004 100.00

PURCHASE-ITEM BOOK TABLE

ORDER-NO BOOK NO QTY PRICE BOOK-NO. BOOK-TITLE BOOK-PRICE

1101 7100 2 310.00 7100 COBOL HAND BOOK 155.00
1101 3990 5 400.00
1101 3667 4 400.00 3990 Lotus 1-2-3 80.00
3667 dBase – IV 100.00
1103 7100 2 310.00
8980 DB2 300.00
1103 3900 3 240.00
9854 CICS 400.00
1104 3667 1 100.00
2786 JCL Ranade 450.00
Normalization (contd.)

• Third normal form

o Determine whether any transitive dependencies, i.e. dependencies
between non-identifying attributes, exist.
o An entity type is in 3NF if it is already in 2NF and no transitive
dependencies exist.
o In the purchase-order table the customer-address not directly depend on
order-no. It is dependent on customer-no only.
• So in third normal form indirect dependency should be eliminated
Normalization (contd.)

• Third normal form

o CUSTOMER
• (cust-no, cust-address)

o PURCHASE- ORDER
• (order-no, cust-no, order-date, delivery- date, total-price)

o PURCHASE-ITEM
• (order-no, book-no, qty-ordered, price)

o BOOK
• (book-no, book-title, book-price)
Normalization (contd.)

• Third normal form

PURCHASE-ITEM TABLE BOOK-TABLE
ORDER-NO BOOK NO QTY PRICE BOOK-NO. BOOK-TITLE BOOK-PRICE
1101 7100 2 310.00 7100 COBOL HAND BOOK 155.00

1101 3990 5 400.00 3990 Lotus 1-2-3 80.00

1101 3667 4 400.00 3667 dBase – IV 100.00
1103 7100 2 310.00
8980 DB2 300.00
1103 3900 3 240.00
9854 CICS 400.00
1104 3667 1 100.00
2786 JCL Ranade 450.00

CUSTOMER TABLE PURCHASE-ORDER TABLE

CUSTOMER-NO CUSTOMER-ADDRESS ORDER-NO CUST-NO ORDER-DATE DELIVERY-DATE TOTAL-PRICE
1234 6,SMITH ROAD 1101 1234 20/03/2003 20/04/2004 1100.00
1254 7,ITC 1103 1254 12/12/2004 18/12/2004 550.00
1789 10/A UK
COUNCIL 1104 1789 01/04/2004 08/04/2004 100.00
Questions?

Thank You !