Performance Analysis and Optimization of Parallel Scientific Applications on CMP Clusters

Chip MultiProcessors (CMP) are widely used for high performance computing. Further, these CMPs are being configured in a hierarchical manner to compose a node in a cluster system. A major challenge to be addressed is efficient use of such cluster systems for large-scale scientific applications. In this paper, the authors quantify the performance gap resulting from using different number of processors per node; this information is used to provide a baseline for the amount of optimization needed when using all processors per node on CMP clusters.