The model can be thought of as a repository of semantic concepts and their interrelations, visualized by a set of
views. These views can be:
-
Structural (e.g. class diagram, structure diagram, package diagram, source code)
-
Behavioral (e.g. state machine diagram, activity diagram)
-
Interaction (e.g. sequence diagram, timing diagram)
-
Functional (e.g. information flow diagram, use case diagram)
-
Performance (e.g. constraints modifying elements present in the other views)
A model the is composed of a great many contained artifacts or work products. In addition, a model can contain a nested
model, such as a Platform-Independent Model (PIM) or Platform-Specific Model (PSM), which are themselves models within
the over all model.
A "logical" model can be split across multiple physical realization models for developer convenience; indeed this is
almost always the case for large models. A typical large-scale logical model consists of the following separate
physical models:
-
System engineering model (specifies the overall system requirements as use cases and the overall system
architecture)
-
Software Requirements model (specifies the use cases specific to the software development)
-
(Several) subsystem model (specifies the realization of each architectural unit)
-
Shared model (contains interfaces, classes and types shared among subsystem models)
|