A quantitative comparison of the BSP and LogP models of parallel computation is developed. We concentrate on a variant of LogP that disallows the so-called stalling behavior, although issues surrounding the stalling phenomenon are also explored. Very efficient cross simulations between the two models are derived, showing their substantial equivalence for algorithmic design guided by asymptotic analysis. It is also shown that the two models can be implemented with similar performance on most point-to-point networks. In conclusion, within the limits of our analysis that is mainly of an asymptotic nature, BSP and (stall-free) LogP can be viewed as closely related variants within the bandwidth-latency framework for modeling parallel computation. BSP seems somewhat preferable due to its greater simplicity and portability, and slightly greater power. LogP lends itself more naturally to multiuser mode.

data types and the development of data structures. Communications of the ACM, 20(6):396-404, June 1977. [82] A. Nico Habermann. Synchronization of communicating processes. Communications of the ACM, 15(3):171-176, March 1972. [83] J.Y. Halpern and Y. Moses. Knowledge and common knowledge in a distributed environment. Journal of the ACM, 37(3):549-587, 1990. A preliminary version appeared in Proc. 3rd ACM Symposium on Principles of Distributed Computing, 1984. [84] David Harel and Michal Politi. Modeling Reactive Systems with Statecharts. McGraw-Hill, 1998. [85] E.C.R. Hehner. Another look at communicating processes. Technical Report CSRG-134, University of Toronto, September 1981. [86] E.C.R. Hehner and C.A.R. Hoare. A more complete model of communicating processes. Theoretical Computer Science, 26, September 1983. [87] Eric C.R. Hehner. A Practical Theory of Programming. SpringerVerlag, 1993. [88] B. Heyd and P. Cregut. A modular coding of UNITY in COQ. In J. Grundy and J. Ha...

We study two parallel computing models – the Bulk Synchronous Parallel (BSP) and the Queued Shared Memory (QSM) – as alternatives to the PRAM model to provide more accurate performance predictions and analyses, and compares the two models in detail. As a case study, we consider a simple hashing problem, design the two versions – the message passing version and the shared memory version – of the algorithm, and compare their run time analytically. The message passing version of the algorithm is implemented and the experiments are performed to display the accuracy and the limitations of the predicted performance analysis. 1.

Abstract not found