In this paper we describe how our computational model can be used f o r the problems of processor allocation and task mapping on a network of workstations. The intended applications f o r this model include the dynamic mapping problems of shrinking or spreading a n existing mapping when the available pool of workstations changes during execution of the problem. The concept of problem edge c l a s s and other features of our model are developed to realistically and eficiently support task partitioning and merging for static and dynamic mapping. Algorithms f o r shrinking and spreading are presented and execution times o n a network of workstations are used to illustrate the utility of our model.

No abstract