Is SIMD Enough for Scientific and Engineering Applications on Massively Parallel Computers ?
S. J. Plimpton, S. S. Dosanjh, R. B. Krall, in Proc of 37th IEEE International Computer Conference, COMPCON '92, San Francisco, CA, February 1992. p 95-102.
Some basic issues involved in matching an application to a distributed memory parallel machine are addressed in this paper. In particular, data communication and processor synchronization as they relate to MIMD (multiple instruction - multiple data) and SIMD (single instruction - multiple data) architectures are discussed. To illustrate the differences, we describe the implementation and performance of several engineering and scientific applications that have been coded for both kinds of machines. We find that many problems are well suited to both architectures. However, when the natural parallelism in a simulation requires a loose synchronization between processors, the MIMD paradigm offers a greater programming flexibility than SIMD.
Return to Publications page