Matrix Multiplication on Heterogeneous Platforms

Abstract

ÐIn this paper, we address the issue of implementing matrix multiplication on heterogeneous platforms. We target two different classes of heterogeneous computing resources: heterogeneous networks of workstations and collections of heterogeneous clusters. Intuitively, the problem is to load balance the work with different speed resources while minimizing the communication volume. We formally state this problem in a geometric framework and prove its NP-completeness. Next, we introduce a (polynomial) column-based heuristic, which turns out to be very satisfactory: We derive a theoretical performance guarantee for the heuristic and we assess its practical usefulness through MPI experiments. Index TermsÐParallel algorithms, load balancing, communication volume, matrix multiplication, numerical linear algebra libraries, heterogeneous platforms, cluster computing, metacomputing.

DOI: 10.1109/71.963416

Extracted Key Phrases

31 Figures and Tables

Statistics

01020'03'05'07'09'11'13'15'17
Citations per Year

140 Citations

Semantic Scholar estimates that this publication has 140 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{Beaumont2001MatrixMO, title={Matrix Multiplication on Heterogeneous Platforms}, author={Olivier Beaumont and Vincent Boudet and Fabrice Rastello and Yves Robert}, journal={IEEE Trans. Parallel Distrib. Syst.}, year={2001}, volume={12}, pages={1033-1051} }