Advertisement

Vector ISA Extension for Sparse Matrix-Vector Multiplication

  • Stamatis Vassiliadis
  • Sorin Cotofana
  • Pyrrhos Stathis
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1685)

Abstract

In this paper we introduce a vector ISA extension to facilitate sparse matrix manipulation on vector processors (VPs). First we introduce a new Block Based Compressed Storage (BBCS) format for sparse matrix representation and a Block-wise Sparse Matrix-Vector Multiplication approach. Additionally, we propose two vector instructions, Multiple Inner Product and Accumulate (MIPA) and LoaD Section (LDS), specially tuned to increase the VP performance when executing sparse matrix-vector multiplications.

References

  1. [1]
    H. Amano, T. Boku, T. Kudoh, and H. Aiso. (SM)2-II: A new version of the sparse matrix solving machine. In Proceedings of the 12th Annual International Symposium on Computer Architecture, pages 100–107, Boston, Massachusetts, June 17-19, 1985. IEEE Computer Society TCA and ACM SIGARCH.Google Scholar
  2. [2]
    W. Buchholz. The IBM System/370 vector architecture. IBM Systems Journal, 25(1):51–62, 1986.Google Scholar
  3. [3]
    R. Doallo, J.T. no, and F. Hermo. Sparse matrix operations in vector and parallel processorse. High Performance Computing, 3:43–52, 1997.Google Scholar
  4. [4]
    I.S. Duff. The use of vector and parallel computers in the solution of large sparse linear equations. In Large scale scientific computing. Progress in Scientific Computing Volume 7, pages 331–348, Boston, MA, USA, 1986. Birkhäuser.Google Scholar
  5. [5]
    I.S. Duff, A.M. Erisman, and J.K. Reid. Direct Methods for Sparse Matrices. Clarendon Press, Oxford, UK, 1986.Google Scholar
  6. [6]
    A W. et al. The white dwarf: A high-performance application-specific processor. In Proceedings of the 15th Annual International Symposium on Computer Architecture, pages 212–222, Honolulu, Hawaii, May-June 1988. IEEE Computer Society Press.Google Scholar
  7. [7]
    T.J.R. Hughes. The Finite Element Method. Prentice-Hall, Englewood Cliffs, NJ, 1987.Google Scholar
  8. [8]
    R.N. Ibbett, T.M. Hopkins, and K.I.M. McKinnon. Architectural mechanisms to support sparse vector processing. In Proceedings of the 16th ASCI, pages 64–71, Jerusalem, Israel, June 1989. IEEE Computer Society Press.Google Scholar
  9. [9]
    P.M. Kogge. The Architecture of Pipelined Computers. McGraw-Hill, New York, 1981.Google Scholar
  10. [10]
    A. Padegs, B.B. Moore, R.M. Smith, and W. Buchholz. The IBM System/370 vector architecture: Design considerations. IEEE Transactions on Computers, 37:509–520, 1988.Google Scholar
  11. [11]
    V.E. Taylor, A. Ranade, and D.G. Messerschitt. SPAR: A New Architecture for Large Finite Element Computations. IEEE Transactions on Computers, 44(4):531–545, April 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Stamatis Vassiliadis
    • 1
  • Sorin Cotofana
    • 1
  • Pyrrhos Stathis
    • 1
  1. 1.Electrical Engineering Dept.Delft University of TechnologyDelftThe Netherlands

Personalised recommendations