Improving Performance of Multi-Dimensional Array Redistribution on Distributed Memory Machines

Array redistribution is required very often in programs on distributed memory parallel computers. It is essential to use e cient algorithms for redistribution, otherwise the performance of the programs may degrade considerably. In this paper, we focus on automatic generation of communication routines for multidimensional redistribution. The principal advantage of this work is to gain the ability to handle redistribution between arbitrary source and destination processor sets and between arbitrary source and destination distribution schemes. We have implemented these algorithms using Parallelware communication library. Some optimization techniques for our algorithms are also proposed in this paper. Experimental results show the e ciency and exibility of our techniques compared to the other redistribution works. keyword Parallelizing compiler, HPF, Array redistribution, Automatic data distribution, Automatic parallelization and optimization.