A Shortest Tree Algorithm for Optimal Assignments Across Space and Time in a Distributed Processor System

Abstract
The problem of optimally assigning the modules of a program over the processors of an inhomogeneous distributed processor system is analyzed. The objective is to assign modules, wherever possible, to the processors on which they execute most rapidly while taking into account the overhead of interprocessor communication. Factors contributing to the cost of an assignment are 1) the amount of computation required by each module, 2) the amount of data transmitted between each pair of modules, 3) the speed of each processor, and 4) the speed of the communication link between each pair of processors.