Your Favorite Parallel Algorithms Might Not Be as Fast as You Think

Authors:
D. C. Fisher
Affiliations:
-
Venue:
IEEE Transactions on Computers
Year:
1988

Citing 2
Cited 4

Spacetime representations of computational structures

Computing
Computational Aspects of VLSI

Computational Aspects of VLSI

A parallel computer model supporting procedure-based communication

Proceedings of the 1990 ACM/IEEE conference on Supercomputing
Spatial machines: a more realistic approach to parallel computation

Communications of the ACM
Effective GIS Mobile Query System

FGIT '09 Proceedings of the 1st International Conference on Future Generation Information Technology
The effect of multiprocessor radius on scaling

Parallel Computing

Quantified Score

Hi-index	15.00

Visualization

Abstract

A problem that requires I inputs, K outputs and I computations is to be solved on a d-dimensional parallel processing machine (usually dor=3). Finite transmission speed and other real-world conditions are assumed. It is proved that the time needed to solve the problem is t= Omega /sub max/ (I/sup 1/d/, K/sup 1/d/, T/sup 1/(d+1)/). This result is demonstrated for the standard algorithm for multiplying two n*n matrices.