Mapping asynchronous parallel simulation on a network of workstations

This paper examines the performance of mapping schemes used in parallel simulation on a network of workstations. A conservative parallel simulation of multistage interconnection networks is used as an example in our analytical model. Performance metrics such as elapsed time, speedup and simulation bandwidth associated with different schemes for partitioning/mapping parallel simulation onto physical processors are evaluated. Our studies show that a perfectly balanced workload distribution may not necessary translate into better performance. We also show in both analytical and implementation results that inter-processors communication overheads incurred in a balanced mapping may occasionally cause dominant aggravation to the program elapsed time.