Database Reference
In-Depth Information
Note the data models at the different levels from the semantic data model to the
physical model. All of these are data models. Why? Remember that a data model
is a representation of real-world information and not the actual real world itself.
Each of the models indicated in the figure is a representation of real-world infor-
mation at a particular level, but they are all representations; they are all data models.
Let us describe the distinguishing features of each pattern. Pay special attention to
understanding where each pattern fits in, why it is necessary, and how it contributes
to the goal of implementing the proposed database.
High-Level Data Model
Let us begin at a level where none of the specifics of predefined arrangement of
data structures comes into consideration. This is a level at which the data model is
as generic as possible. In other words, once you have a data model at this high level,
then making the transition to any type of conventional model is easy and straight-
forward. From a generic model, you can branch out to any of the conventional
models, namely, hierarchical, network, or relational. That is the advantage of creat-
ing the generic, nonspecific model first.
Some database authors refer to the data model at the highest level as a seman-
tic data model. A semantic model conveys the essential meaning of the data to be
stored in the proposed database. This data model is created with commonly
accepted building blocks or components. The following is a list of the essential fea-
tures of the semantic data model:
•
Contains relevant, selected details
•
Correctly reflects the information requirements by excluding unnecessary
details
•
Truly conveys the meanings of all relevant “things” in reality
•
Exactly represents the associations between “things” in real-world situations
•
Correct data model of a real-world system
•
Proper mapping of real-world information requirements
•
Convenient means for communicating proposed information content to users
Object-Based Modeling Technique
Enough has been said about the semantic, generic data model. You have noted its
nature, its characteristics, and its place in the database development cycle. You also
understand that the semantic model is built with certain components. Now let us
briefly turn our attention to the major techniques for building the semantic model.
Two techniques are commonly used: the object-based modeling technique and the
entity-relationship modeling technique. When you use either of the techniques, the
result is a semantic data model of the proposed database. The techniques are similar
and have many common features. They even share common notations or symbols
to denote the building blocks or components. Chapters 6 and 7 cover the principles
and components used in these two techniques. At this stage, we will just introduce
the two techniques and broadly describe them.
Search WWH ::
Custom Search