Scientiic Computing on Bulk Synchronous Parallel Architectures

Bulk synchronous parallel architectures o er 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 e ciency with which a wide range of important scienti c 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 di erential equations on a multidimensional discrete grid. These computations are analysed in a uniform manner by formulating their basic procedures as sparse matrix-vector multiplications. 2