Estimation of Intermodule Communication (IMC) and Its Applications in Distributed Processing Systems

Communication among program modules plays an important role in the performance of distributed processing systems. In this paper, a model for estimating intermodule communication (IMC) is developed. The model derives communication volume based on module invocation rates and file access probabilities via the control-and-data-flow graph. The IMC model is validated by simulation experiments. Interprocessor communication (IPC) and system resources utilization can be estimated from the IMC. We show that IMC and IPC are useful in finding good module assignments in distributed processing systems.