Database Systems I

Database Design, ER Modelling

Lecture 2

ESOFT COMPUTER STUDIES - GALLE DB 1 1

Objectives
Purpose of Database Systems

Phases of database design

Relational Database design

Introduction to ER Modelling

ESOFT COMPUTER STUDIES - GALLE DB 1 2

Design process (6 steps)
1.Requirements Analysis
What does the user want?
2.Conceptual Database Design

Defining the entities and attributes, and the relationships between these
--> The ER model
3.Logical Database Design (Map ER to Relational Schema)
4.Schema Refinement (fine tune )
5.Physical Database Design
Implementation of the design using a Database Management System
6.Security Design
Implement Controls to ensure security and integrity

ESOFT COMPUTER STUDIES - GALLE DB 1 3

Requirements Analysis
What users want from the database?
What is going to be stored in the database?
What applications are going to be built on top the database?
What are the most frequently asked queries?

ESOFT COMPUTER STUDIES - GALLE DB 1 4

Requirement Analysis (Library database)
Data to be stored can be…
◦ Record of all books in the library
◦ Record of members of the library
◦ Students,
◦ Faculty
◦ Record borrowing information

ESOFT COMPUTER STUDIES - GALLE DB 1 5

Requirements Analysis
Some applications on top of the database can be…

◦ Renewal service (may be on-line)

◦ Borrowing-Lending service
◦ Resource reservation system (may be on-line)
◦ Resource request service (may be on-line)

ESOFT COMPUTER STUDIES - GALLE DB 1 6

Requirements Analysis
Think about Facebook…
•What is going to be stored in the database?
•What applications are going to be built on top the database?
•What are the most frequently asked queries?

ESOFT COMPUTER STUDIES - GALLE DB 1 7

Conceptual Database Design
The information gathered in the requirements analysis phase is used to create a:

high-level description of the data in a conceptual data model or Semantic Data

Model.

Eg. ER Model

ESOFT COMPUTER STUDIES - GALLE DB 1 8

Logical Database Design
◦ In this step, we determine the DBMS to implement the database & also the
data model.

◦ We utilize the conceptual schema created in the previous step and convert it
into a schema of a particular data model* (e.g. Relational Data Model )

◦ *we will cover this in the next two lectures.

ESOFT COMPUTER STUDIES - GALLE DB 1 9

 Schema Refinement
The schema created by the logical database design phase is further refined for
potential problems such as redundancies.

ESOFT COMPUTER STUDIES - GALLE DB 1 10

Physical Database Design
In this step, performance criteria are taken into consideration and further
enhancements to the schema & creation of indexes are considered.

ESOFT COMPUTER STUDIES - GALLE DB 1 11

Security Design
Different user groups and their roles are identified.
Appropriate levels of access are then provided to the data ensuring that users have
access to only the necessary data.
Eg. Bank
Customer – read access
Teller – read/update (limited)
Manager – read/update

ESOFT COMPUTER STUDIES - GALLE DB 1 12

Entity-Relationship (ER) Model
Requirements gathering

Conceptual database design

high level data model(ER)created

The 2 main constructs of the ER model:

entities , relationships.

different books use different graphical notation.

ESOFT COMPUTER STUDIES - GALLE DB 1 13

ER Diagrams

ESOFT COMPUTER STUDIES - GALLE DB 1 14

NOTATION FOR ER
Symbol Meaning

ENTITY

WEAK ENTITY

RELATIONSHIP

IDENTIFYING RELATIONSHIP

ATTRIBUTE

KEY ATTRIBUTE

MULTIVALUED ATTRIBUTE

COMPOSITE ATTRIBUTE

DERIVED ATTRIBUTE
E1 R E2
TOTAL PARTICIPATION OF E2 IN R
1 N
E1 R E2
CARDINALITY RATIO 1:N FOR E1:E2 IN R
(min,max)
R E
STRUCTURAL CONSTRAINT (min, max) ON PARTICIPATION OF E IN
R

ESOFT COMPUTER STUDIES - GALLE DB 1 15

Entity
•An entity is an object in the real world with an independent existence (place,
person, car, house, company, job, university, etc).

•A collection of similar entities is called an entity set.

Student
•Graphically, ENTITY

• First letter of each word in the entity name is uppercase

• E.g., Student

• Normally an entity is named using a ‘noun’ or ‘noun phrase’

ESOFT COMPUTER STUDIES - GALLE DB 1 16

 Attributes
• An entity is described using a set of attributes
• Graphically, Oval.
attribute

For example staff has 3 attributes name

• Name
• Position student address
• Salary

age

ESOFT COMPUTER STUDIES - GALLE DB 1 17

Entities & Attributes
1. ESOFT Student Registration
New students are required to produce their national ID card number,
name, address, age, date of birth and gender during the registration
process. they are also required to select the degree program of their
choice.
•Entity ?
•Attributes ?
2. Sampath Bank Account Opening System
Customers are required to provide NIC number , address , phone, type of
account (savings/current etc) and the opening balance at the time of
opening an Account..
• Entity ?
•Attributes ?

ESOFT COMPUTER STUDIES - GALLE DB 1 18

Entity Set
Example: name, id, age & salary are attributes in
Employee entity

salary
salary name
e1
age id
e2 age
e3
Employee
e4
e5 i
e6 d
…
name

Employee Entity set

ESOFT COMPUTER STUDIES - GALLE DB 1 19

Domain of an Attribute
The domain of an attribute specifies the set of possible values that the
attribute can have.

age : integer values from 0 - 120

name: 20 character string

ESOFT COMPUTER STUDIES - GALLE DB 1 20

Composite Attributes
Composite attributes can be divided into smaller
subparts.

Graphically, Employee
name

surname middle_name
first_name

ESOFT COMPUTER STUDIES - GALLE DB 1 21

Multi valued Attributes
•Attributes containing multiple values are called multivalued attributes.
Double lines

Student

E-mail
first_name
surname

gmail , yahoo

ESOFT COMPUTER STUDIES - GALLE DB 1 22

Derived Attributes
Some attributes can be derived - called derived attributes. Dotted line
For example, age is derived from date of birth & current date.

Employee

age
name

Derived attribute

ESOFT COMPUTER STUDIES - GALLE DB 1 23

Tick where applicable
Attribute simple single-valued composite multivalued derived

stuNum

stuName

stuMajor

stuDob

stuHrs

stuYr

stuGpa

stuAge

ESOFT COMPUTER STUDIES - GALLE DB 1 24

Key Attributes
A key attribute minimal set of attributes which uniquely identifies an
entity in the entity set.
Underlined

salary name
e1
age id
e2
e3
e4
Employee Entity set e5
e6
…

ESOFT COMPUTER STUDIES - GALLE DB 1 25

Primary Key
There can be multiple key attributes called candidate keys in a single entity.
A single candidate key is designated as the primary key.

Name

Employee

Candidate
eid nic
key

ESOFT COMPUTER STUDIES - GALLE DB 1 26

Composite Key
Sometimes, a group of attributes make up the key. This is called a
composite key.

Example :
Composite key = (First name + surname + tel)

Key = NIC

ESOFT COMPUTER STUDIES - GALLE DB 1 27

Students should enroll for courses. Student need to provide studentID number,
name, address , date of birth, gender, contact number when he/she is going to
register for a course. University keeps track of course number , course name ,
offering faculty ,duration and course fee of each course.
How do you represent this scenario ?

ESOFT COMPUTER STUDIES - GALLE DB 1 28

 name
address
phone

Student gender

studentID dob

courseNo

Course coursename
fee

duration faculty

ESOFT COMPUTER STUDIES - GALLE DB 1 29

Relationship
•Relationship is an association among two or more entities. Diamond
•Collection of similar relationships - relationship set .
•Shown as a line connecting the associated entities, labelled with the name of the
relationship.

• Normally a relationship is named using a ‘verb’ or ‘verb group’.

N 1
Student registers Course

relationship

ESOFT COMPUTER STUDIES - GALLE DB 1 30

 Descriptive Attributes
A relationships can also have descriptive attributes . Used to record information
about the relationship rather than any one of the participating entities.

startdate
since

N 1
Employee Works in Department

Nimal works in accounts dept since 2008

ESOFT COMPUTER STUDIES - GALLE DB 1 31

Descriptive Attributes
1/2/99

e011 1/2/09
HR
e022
1/5/01
Sales
e033 1/2/07
Marketing
e044
1/3/05

Departments
Employees
Work_in

ESOFT COMPUTER STUDIES - GALLE DB 1 32

Degree of Relationship type

Degree of a relationship is the number of participating entities in the

relationship.

Relationships can be classified based on their degree into

◦ Binary – relationship with two participants

◦ Ternary – relationship with three participants
◦ Quaternary – relationship with four participants

ESOFT COMPUTER STUDIES - GALLE DB 1 33

A Ternary Relationship
•A diamond is used to represent ternary and quaternary relationships

Ex : Arpico super centers are situated all over the country and each Arpico super center mainly
contains 5 departments such as furniture , rubber, plastic, supermarket items and maintenance.

Employee Works in Department

Location

ESOFT COMPUTER STUDIES - GALLE DB 1 34

Cardinality
The cardinality ratio for a binary relationship specifies the number of
relationship instances that an entity can participate in.

•There are three types of cardinality ratios for binary relationships.

• one-to-one (1: 1)
• one-to-many (1 : N)
• many-to-many. (N : M)

ESOFT COMPUTER STUDIES - GALLE DB 1 35

One-to-One relationship
An employee manages at most one department. A department can have only one
manager managing it.

1 1
Employee Manages Department

ESOFT COMPUTER STUDIES - GALLE DB 1 36

One-to-Many relationship
An employee works in at most one department.

N 1
Employee Works in Department

This is a one-to-many (or many-to-one) relationship

ESOFT COMPUTER STUDIES - GALLE DB 1 37

 Many-to-many relationship
Example: An employee can work on several projects. A project can have many
employees working on it.

M N
Employee Works on Project

ESOFT COMPUTER STUDIES - GALLE DB 1 38

ER Diagram

ESOFT COMPUTER STUDIES - GALLE DB 1 39

Participating Constraints
Participating constraint specifies whether the existence of an entity depends
on its being related to another entity via the relationship type.

ESOFT COMPUTER STUDIES - GALLE DB 1 40

Participation Constraints
For example, if we specify that an employee must always work for a department.
Then we say that the relationship “works in” is in total participation from
employee entity to department entity.

Employee Works in Department

(double lines)
Employee must work in a department total participation
Department may or may not have employees

ESOFT COMPUTER STUDIES - GALLE DB 1 41

E-R Model (contd.)
If the relationship is not in total participation, then it is known as in partial.

For example, from Department to Employee

Department may or may not have Employees

ESOFT COMPUTER STUDIES - GALLE DB 1 42

Participation Constraints

Participation Constraint
Student?

ESOFT COMPUTER STUDIES - GALLE DB 1 43

Weak Entity
Some entities cant exist on its own.
It is existence-dependent on another entity, i.e., it cannot exist without the
entity with which it has a relationship.

It inherits at least part of their primary key from the entity to which it is related.

Entity types without any key attributes is called weak entity types. Double
lined box

ESOFT COMPUTER STUDIES - GALLE DB 1 44

Weak Entity

Strong entity Weak entity

1 M
Employee covers Dependent

idid name age

age

Partial key

ESOFT COMPUTER STUDIES - GALLE DB 1 45

A weak entity is uniquely identified by considering some of its attributes
with the primary key of another entity called the identifying owner.
he attributes in the weak entity participating in the key are called partial
keys.

ESOFT COMPUTER STUDIES - GALLE DB 1 46

E-R Model (contd.)
• The owner entity and the weak entity participates in an identifying
relationship.

• The cardinality of the identifying relationship is either one-to-one or one-to-

many from owner entity to weak entity.

• The weak entity must have total participation in the identifying relationship.

ESOFT COMPUTER STUDIES - GALLE DB 1 47

E-R Model (contd.)
Identifying relationship Weak entity

Strong entity
1 M
Employee Policy Dependent

idid name age

age

Partial key

ESOFT COMPUTER STUDIES - GALLE DB 1 48

Recursive Relationship
Entities participating in a relationship need not be distinct. Such relationships
are called recursive relationships.

Each entity in the relationship play a role in the relationship. It is recommended

to state the role in recursive relationships.

ESOFT COMPUTER STUDIES - GALLE DB 1 49

E-R Model (contd.)
Student leads set of students in group assignments.

leads

Group leader member

Student

ESOFT COMPUTER STUDIES - GALLE DB 1 50

Exercise
Employees work in departments; each department is managed by an
employee; a child must be identified uniquely by name when the parent
(who is an employee; assume that only parent works for the company) is
known. We are not interested about a child once the parent leaves the
company.

ESOFT COMPUTER STUDIES - GALLE DB 1 51

Alternate Notation – Min..Max
Used to specify the number of possible occurrences of each participating
entity type in a relationship.
has two parts
◦ Min
◦ Max
◦ For example, (0,1)
◦ Min = 0
◦ Max = 1

Max of a multiplicity range denotes Cardinality

Min of the range denotes Participation constraint

ESOFT COMPUTER STUDIES - GALLE DB 1 52

 Enhanced ER Modelling
•ER modelling does not capture all the semantics of client’s domain, such as

• ‘ISA’ (‘is a’) relationship or specialization-generalization

• ‘Manager’ entity type ‘is a’ subentity of ‘Staff’ entity.

•Enhanced ER models represent the above relationships

• Therefore capture client’s domain more comprehensively.

ESOFT COMPUTER STUDIES - GALLE DB 1 53

Diagrammatic Representation of ‘ISA’ relationship

name
position
staffNo

Staff

Specialization/generalization indicator

ISA Both hourly employees &

contract employees
Belong to the type Staff
Hourly_Emp Contract_Emp

Hours worked
Contract_ duration
Hourly rate

ESOFT COMPUTER STUDIES - GALLE DB 1 54