Visualisation Architecture for REuse (VARE)

VARE is a client-server architecture. The server contains repositories and processes. On the client side, the user manages the activities associated with creating and viewing a visualisation. The component repository interface lets the user select a component from the repository to create a component set. Once this is created, the user can select an engine type from the engine repository to control the test driving of these components. The engine component is synonymous to the program component in the PMV model.

The engine generates a program trace/test drive trace as output, which is stored in the test drive report repository. A program trace contains all the information required to describe a program execution such as the order of object creation, method invocations, field accesses and field modifications. A program trace is then used as input to a transformer, which is synonymous to the mapping component from the PMV model. The transformer repository interface lets the user select the transformer to use and the program trace to use with it. The transformer then transforms the program trace into an appropriate visualisation.

Finally the finished visualisation is stored in the visualisation repository. The visualisations contain information such as a description of the components they are associated with, who created them, and notes that help the understanding of the visualisation. The visualisation interface lets a user choose a particular visualisation and control its presentation.



The VARE architecture is based on a client/server model, with the server being split into repositories and processes. Dashed lines represent test drive or visualization input/output, while solid lines represent control, queries or responses.


Topic attachments
I Attachment Action Size Date Who Comment
vare.jpgjpg vare.jpg manage 31 K 13 May 2009 - 14:57 Main.craig