skip to main content
poster

Cache friendly sparse matrix-vector multiplication

Published:28 January 2011Publication History
Skip Abstract Section

Abstract

Sparse matrix-vector multiplication or SpMxV is an important kernel in scientific computing. For example, in the conjugate gradient method, where SpMxV is the main computational step. Though the total number of arithmetic operations in SpMxV is fixed, reducing the probability of cache misses per operation is still a challenging area of research. In this work, we present a new column ordering algorithm for sparse matrices. We analyze the cache complexity of SpMxV when A is ordered by our technique. The numerical experiments, with very large test matrices, clearly demonstrate the performance gains rendered by our proposed technique.

References

  1. T. Davis, Uni. of Florida sparse matrix collection. http://www.cise.ufl.edu/research/sparse/Google ScholarGoogle Scholar
  2. M. Frigo, C. E. Leiserson, Prokop, Harald, Ramachandran, and Sridhar, Cache-Oblivious algorithms. FOCS '99: Proc. of the 40th Annual Symp. on Foundations of Comp. Sc., 1999. Google ScholarGoogle ScholarDigital LibraryDigital Library
  3. A. Pinar and M. Heath, Improving performance of sparse matrix-vector multiplication. In Supercomputing '99: Proc. of the 1999 ACM/IEEE, New York, USA, 1999. Google ScholarGoogle ScholarDigital LibraryDigital Library
  4. S. Toledo, Improving the memory-system performance of sparse-matrix vector multiplication, In IBM J. Res. Dev., vol. 41, num. 6, 1997. Google ScholarGoogle ScholarDigital LibraryDigital Library

Index Terms

  1. Cache friendly sparse matrix-vector multiplication

          Recommendations

          Comments

          Login options

          Check if you have access through your login credentials or your institution to get full access on this article.

          Sign in

          Full Access

          • Published in

            cover image ACM Communications in Computer Algebra
            ACM Communications in Computer Algebra  Volume 44, Issue 3/4
            September/December 2010
            145 pages
            ISSN:1932-2240
            DOI:10.1145/1940475
            Issue’s Table of Contents

            Copyright © 2011 Authors

            Publisher

            Association for Computing Machinery

            New York, NY, United States

            Publication History

            • Published: 28 January 2011

            Check for updates

            Qualifiers

            • poster
          • Article Metrics

            • Downloads (Last 12 months)2
            • Downloads (Last 6 weeks)0

            Other Metrics

          PDF Format

          View or Download as a PDF file.

          PDF

          eReader

          View online with eReader.

          eReader