Towards Efficient and Portable Parallel Computing: Programming with the BSP Model

Abstract

Advances in parallel computing have long been hindered by the lack of a unifying model acceptable to both software and hardware designers. What we seek is essentially a parallel analogue to the von Neumann model, which has served as the basis for sustained improvements in sequential computing, and hence the success of the sequential computing industry. The Bulk-Synchronous Parallel (BSP) model has been proposed by Valiant as a unifying model for parallel computing. In this talk, we discuss recent experimental work that demonstrates the use of the BSP model to develop efficient and portable programs for a range of parallel machines and applications.

About the Speaker

Mark W. Goudreau received the B.S. degree in Electrical Engineering from Rutgers University, New Brunswick, New Jersey, in 1988. He received the M.A. degree in Electrical Engineering in 1990 and the Ph.D. degree in Electrical Engineering in 1993, both from Princeton University, Princeton, New Jersey. From 1990-1993, he was first an Intern and later a Visiting Scientist at the NEC Research Institute, Princeton, NJ. Since 1993, he has been an Assistant Professor in the School of Computer Science at the University of Central Florida, Orlando, Florida. His research interests include parallel computing, computer networks, and artificial neural networks.


Philip Chan, pkc@cs.fit.edu
Last modified: Tue Jan 6 13:29:04 EST 1998