Database Design Phases
THE FOUR PHASES TO DESIGN ANY DATA BASE SYSTEM ARE:
1. FORMULATION OF INFORMATION REQUIREMENT & ANALYSIS PHASE:
This phase is also calledFeasibility phase. In this phase through the interviews and reviewing all related documents and policies in the organization the following items are identified:
a. Clear and concise definition of the problem
b. Local dependency lists
c. local dependency diagrams
d. Local Schema
2. LOGICAL SCHEMA DESIGN PHASE:
In this phase the following items are performed:
a. Consolidation of dependency lists.
c. Consolidation of logical schema.
The output of this phase is alogical schema that is independent of all computer hardware and software systems.
3. IMPLEMENTATION DESIGNPHASE:
In this phase the logical schema which was designed in the Logical Design Phase is modified to fit the specific data model, hardware and software system that the designer wants to use.This new schema is called IMPLEMENTATION SCHEMA.
4. PHYSICAL DESIGN PHASE:
In this phase the Implementation Schema which was designed in the Implementation Phase is programmed using theDDL (Data Definition Language) or any other software language which is available for the programmer.
OVERALL THE DATA BASE DESIGN MEANS:
a. Identifying all the required files. In data base terminology files are called record types.
b. Identifying the fields of each of these record types.Fields in database terminology are called attributes. Note that attributes in semantic object oriented databases are called properties.
c. Identifying the primary key of each of these file. Note that the primary key is a field uniquely identifies a specific record in the file.
d. Identifying the relationship between record types.The designer must identify the manner in which information from different files are related and accessed. This is achieved by identifying the foreign and secondary keys. A foreign key is a field within a file which is the primary key of another file. The secondary key is a field identifies not a unique record but a group of records with a specific characteristic. For example credit limit can be a key to identify all record with a specific credit limit such as $500.