Some Experiments with a Basic Linear Algebra Routine on Distributed Memory Parallel Systems

In this paper we describe tbe algorithm used, discllss the iIllJlI~lI\entaLiOil and pres~lIt thf' performance of the Basic Linear Algebra. Subroutine (BLAS) sgemv for distribuLed-memory multiprocessors. The basic assumption is that the maLrix and the vectors are row distrihuted among processors. Performance data from nCUBE 11, iPSC/860 and iPSe DELTA machines are presented.