On the Performance of MPI-OpenMP on a 12 nodes Multi-core Cluster

TL;DR

This paper evaluates the performance of MPI-OpenMP hybrid programming on a multi-core cluster with dual Quad-Core processors per node, highlighting the importance of optimizing both local and network communications for scalability.

Contribution

It provides an analysis of MPI-OpenMP performance on multi-core clusters with dual Quad-Core nodes, emphasizing communication optimization strategies.

Findings

MPI-OpenMP reduces communication overhead in multi-core clusters

Local and network communication optimization are crucial for scalability

Benchmark results demonstrate performance benefits of hybrid approach

Abstract

With the increasing number of Quad-Core-based clusters and the introduction of compute nodes designed with large memory capacity shared by multiple cores, new problems related to scalability arise. In this paper, we analyze the overall performance of a cluster built with nodes having a dual Quad-Core Processor on each node. Some benchmark results are presented and some observations are mentioned when handling such processors on a benchmark test. A Quad-Core-based cluster's complexity arises from the fact that both local communication and network communications between the running processes need to be addressed. The potentials of an MPI-OpenMP approach are pinpointed because of its reduced communication overhead. At the end, we come to a conclusion that an MPI-OpenMP solution should be considered in such clusters since optimizing network communications between nodes is as important as optimizing local communications between processors in a multi-core cluster.