A Strategy for Point-to-point Node Communication Performance Prediction in Cluster Environments

Abstract

In this paper, we present and apply a methodology for parallel programming, along with MPI performance measurement and prediction in a class of distributed computing environments, named networks of workstations (NOW). Our approach is based on a two-level model where, at the top, a new parallel version of timing graph representation is used to explicit the parallel communication and code segments of a given parallel program. At the bottom level, analytical models are developed to represent execution behavior of parallel communications and code segments. Obtained execution time, together with problem size and number of nodes are input to the model, which allows us to predict the performance of similar cluster computing systems with a different number of nodes. We validate our analytical model by performing experiments over homogeneous cluster of workstations. Results show that our approach produces accurate predictions, within 6% of actual results.