Scientific Computing on Bulk Synchronous Parallel Architectures

Abstract

Bulk synchronous parallel architectures offer the prospect of achieving both scalable parallel performance and architecture independent parallel software. They provide a robust model on which to base the future development of general purpose parallel computing systems. In this paper we theoretically and experimentally analyse the efficiency with which a wide range of important scientific computations can be performed on bulk synchronous architectures. The computations considered include the iterative solution of sparse linear systems, molecular dynamics, linear programming, and the solution of partial differential equations on a multidimensional discrete grid. These computations are analysed in a uniform manner by formulating their basic procedures as sparse matrix-vector multiplications. 1 Introduction Bulk synchronous parallel (BSP) architectures [30] offer the prospect of achieving both scalable parallel performance and architecture independent parallel software. They provide a ...