Databases Reference
In-Depth Information
•
Facilitating a strategic overview of the business's information needs.
The
data modeling process, minus some of its more detailed steps, can be
employed to build a strategic data model or high-level representation
of all the business's information requirements. A strategic data model
can be thought of as a summarized version of the business conceptual
schema. The strategic data model can assist in analyzing information
interrelationships throughout the business. It therefore can contrib-
ute to more effective planning of database and application implemen-
tations. For example, it can aid in planning and relating detailed
operational systems (those required to run the daily business) and
summarized, executive decision-support systems.
•
Migrating data from one technology to another.
Databases can more eas-
ily migrate from one technology to a newer or more appropriate tech-
nology when a current data model exists. The model can be translated
into a new database implementation. The data model can also be used
to design extract procedures. If a data model does not exist, one
should be built by treating the existing database as a user view and by
working with users to discard unnecessary information requirements
as well as add new ones.
SUMMARY
Data modeling is the first phase of an effective database design process.
It is a technique for understanding and capturing business information
requirements that must be met through a database implementation.
The data modeling process begins with the definition of a user view or
representation of information requirements for one business function,
user, or user group. It culminates in the integration of all user views into
one composite data model. Multiple data models are related through their
mappings to a business conceptual schema, an integrated data model rep-
resenting all the business's data requirements at their most detailed level.
Development of data models is a critical component of a data-driven
design philosophy. Success stories of database designs that were founded
on data models and business conceptual schemas are increasing. The
emergence of computer-aided software engineering (CASE) tools has made
it possible to automate portions of the logical data modeling process. In
addition, the dramatic surge and acceptance of commercial relational
DBMS products have provided opportunities for designing databases that
closely resemble data models. For all of these reasons, the benefits of a sta-
ble data model are perhaps more widely understood and more easily
accomplished than ever before.
Search WWH ::
Custom Search