Transcript Developing Database Specifications

TRACK 3
Developing Database
Specifications
Learning Objectives
At the end of the session, the participants should be able to
• Identify the information requirements of the organisation,
resource centre/department, end-user.
• Identify areas wherein database can address the
requirements.
• Write a database requirement specifications.
• Identify a minimal database design specification.
Why make specifications?
Software Crisis





80-90%
80%
40%
< 40%
< 25%
objectives
 10-20%
-do not meet their performance goals
-delivered late and over budget
-developments fail or are abandoned
-fully address training and skills requirements
-properly integrate business and technology
-meets all their success criteria.
Reasons...
 Lack of a complete requirement specification.
 Lack of an appropriate development methodology.
 Poor decomposition of design into manageable
components.
DDLC
STEP 1 - DATABASE PLANNING
Database Planning
Feasibility Study
- time frame
- resource: budget, people
Integrate the database with the overall
Information Architecture of the organization
Systems Definition
Requirements Collection
and analysis
DBMS
Selection
Prototyping
Database Design
Application
Design
Implementation
Data Conversion and loading
Testing
Evaluation & Maintenance
DDLC
Database Planning
Systems Definition
Requirements Collection
and analysis
DBMS
Selection
Prototyping
Database Design
Application
Design
Implementation
Data Conversion and loading
Testing
Evaluation & Maintenance
DB Specs. Components
 Requirements Specification
• Database Design Specification
The WENDY Example - WENT Database Directory
Requirements Specs.
 Statement of the Problem Domain
 Objectives
• Data Requirements
• Transaction Requirements
DB Design Objectives
 represent the data and the relationships between data
required from the database application
 provide a data model that supports any transactions
required on the data
 specify a minimal design that is appropriately structured to
achieve the stated performance requirements for the
database application
Relational DB Design Phases
• Building the logical/conceptual data model
• Building the physical database design
Data Modelling
• Entity-Relationship Diagram (ERD) - a tool for
modelling data.
• 3 major concepts to remember
• Entity
• Attribute
• Relation
Other Concepts
• Entity type
Example:
FULLTIME FACULTY and PARTTIME FACULTY entities both
keep the same sets of attributes like name, address,
gender, department, position. The two entities can be
generalised as FACULTY, which is the entity type.
Other Concepts
• Attribute Domain
Example:
The attribute GENDER can contain only any one of the
following values: female, male
• Simple Attribute
Example:
Firstname is a simple attribute of the entity PERSON.
Other Concepts
• Composite Attribute
Example:
Name is a composite attribute that has three parts-- first
name, middle name, last name
• Multi-valued Attribute
Example:
Area of expertise attribute can hold one or more values for a
particular PERSON record.
Other Concepts
• Derived Attributes
Example:
The AGE is a derived attribute of the formula: BIRTHDAY –
CURRENT DATE
Integrity Constraints
 Required data
 Attribute domain constraints
 Entity integrity
 Referential integrity