Performance of Parallel Sparse Supports for Fortran 90 on PC-based Networks of Clusters

摘要

PC-based networks of clusters have become an important research testbed recently, as their competitive computing powers, inexpensive costs, scalability, and fast interconnections among CPUs are considered. In order to evaluate the computing power of PC Clusters and maximize their performance, it is very interesting to that observe the performance results of numerical applications on sparse data sets on PC Clusters. Our previous work has designed and implemented a parallel sparse library supporting array intrinsic operations of Fortran 90 on IBM SP2. In this paper, in addition to supporting sparse array intrinsics of Fortran 90, we present a flexible framework to extend the support of our sparse library towards direct numerical solvers such as SuperLU and Cholesky factorization. Experiments are performed on 8-node PC Cluster.