Heuristic model for task allocation in a heterogeneous distributed computing system

In heterogeneous distributed computing systems, partitioning of the application software into modules and the proper allocation of these modules among dissimilar processors are important factors which determine the efficient utilization of resources. This paper presents a new heuristic model which performs static allocation of such program modules in a heterogeneous distributed computing system in a manner that is designed to minimize the interprocessor communication time, while keeping the processing load balanced. The new methodology expands the augmented Maximally Linked Module with Simulated Annealing (MLM/SA) method to allow it to take into account the specific capabilities of the heterogeneous system. The execution time of the algorithm and the quality of the allocation produced is shown to be superior to that of the pure simulated annealing when it was applied to randomly-generated systems and structures based upon real-world problems.