Some Experiments with a Basic Linear Algebra Routine on Distributed Memory Parallel Systems
Hongsu Byun, Elias N. Houstis, E. A. Vavalis · Purdue e-Pubs (Purdue University System) · 1993
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.