Database Reference
In-Depth Information
Real-world
Information
FEASIBILITY
STUDY
REQUIREMENTS
DEFINITION
PLANNING
LOGICAL DESIGN
CUSTOMER
ORDER
CUSTOMER
ORDER
ORDER
DETAIL
ORDER
DETAIL
Conventional Data Model
Semantic Data Model
Figure 11-2
Logical design in DDLC.
taking the semantic data model and transforming it into the physical design. You
will be designing files and records based on the features of the hardware. Your issues
and concerns will be centered more on hardware features than on the information
requirements. As hardware changes occur, your design will have to be modified.
Let us say that you are designing a relational database system by adopting this
plan of bypassing the logical design. You need to define the files and records for
data elements determined in your requirements definition phase. You are not going
to think in terms of relational tables, but only about designing files and records. How
are you going to ensure that your files will be free from data redundancy and that
they will conform to the relational rules? How are you going to ensure that your
files will be free from update, deletion, and addition anomalies? Remember, the
relational model rests on the understanding that data may be perceived as two-
dimensional, relational tables. Relational DBMSs work on this premise.
This logical design phase is essential for the following reasons:
The output of the logical design phase represents data as perceived as tables,
segments, or nodes—the way data are considered in the relational, hierarchi-
cal, and network models.
Users do not understand the output of the physical design phase in the form
of hardware files, records, and so on. Tables, segments, or nodes are better
understood.
Logical design enables data independence, a crucial advantage of the database
approach.
The design effort becomes easier and more manageable when you separate it
into logical and physical design phases.
Search WWH ::




Custom Search