Package Diagram | UML 2 For Dummies

Your systems or software applications start out as one big design problem. Like all good engineers, you take that problem and break it up into smaller problems. If you can solve each of the little problems, you solve the big problem. You can use the package diagram to help break a big problem into smaller problems—and show that decomposition.

We use the UML package notation to show our application as a large package. Inside the large package, we show smaller packages—one for each subsystem. After our subsystems are set, we convert the packages into subsystems. Then we draw dependency lines to indicate which subsystems must rely on other subsystem(s)—and which ones they rely on. (For more details on this process, see Chapter 20.)

Tip At the start of your systems design phase, draw a simple package diagram showing the decomposition of your system into subsystems. Then consider all your system design issues such as layering, subsystem interfaces, database access, and networking. As you make your strategic design decisions, modify the package diagram to reflect those decisions. Finally, give a copy of this big picture system diagram to each of the subsystem design teams. By using the package diagram, they can understand how their subsystem fits into the overall system.