Parallel Performance Studies for a Three-Species Application Problem on maya 2013

Abstract

High performance parallel computing depends on the interaction of a number of factors including the processors, the architecture of the compute nodes, their interconnect network, and the numerical method, and its implication. In this note, we present performance and scalability studies on the newest portion of the cluster maya, referred to as maya 2013, using an existing parallel code for a three-species application problem. This application problem requires long-time simulations on a fine mesh, thus posing a very computationally intensive problem. The speedup of run times afforded by parallel computing makes the difference between simply unacceptably long runs to obtain the results (e.g., several days) and practically feasible studies (e.g., hours). The results also support the scheduling policy implemented, since they con rm that it is beneficial to use all sixteen cores of the two eight-core processors on each node simultaneously.