Skip to main content
SpringerLink
Log in

A vectorization technique for a family of finite difference formulae and its performance evaluation

  • Conference paper
  • First Online:
Applied Parallel Computing Industrial Computation and Optimization (PARA 1996)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 1184))

Included in the following conference series:

  • 163 Accesses

Abstract

In this contribution the vectorization technique on the current vector supercomputer are derived from the relation between the number of gridpoints in the x, y directions in three dimension. This technique is applied to vectorization of the SOR method. Moreover the actual efficiency on the vector supercomputer is examined, and it is shown that the SOR method vectorized by this technique has a high efficiency as more than 90% of the maximum speed of the vector supercomputer.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Adams, L.M., Jordan, H.F.: Is SOR color-blind ?. SIAM J. Sci. Stat. Comput. 7(1986) 490–506

    Google Scholar 

  2. Brand, C.W.: An incomplete-factorization preconditioning using repeated redblack ordering. Numer. Math. 61(1992) 433–454

    Google Scholar 

  3. Fujino, S., Tamura, T., and Kuwahara, K.: Multicolored Poisson Solver for Fluid Flow Problems. Proc. of the 8-th GAMM Conference on Numerical Methods in Fluid Mechanics The Netherlands (1989) 148–158

    Google Scholar 

  4. Hageman, L.A., Young, D.M.: Applied Iterative Methods Academic Press, New York 1981

    Google Scholar 

  5. Iwatsu, R., Hyun, J.M., and Kuwahara, K.: Driven Cavity Flow with Stabilizing Temperature Stratification. AIAA Paper 92-0713 Reno NV 1992

    Google Scholar 

  6. Jones, M.T., Plassmann, P.E.: The effect of many color orderings on the convergence of iterative methods. Proc. of Copper Mountain Conference on Iterative Methods s2 Colorado 1992

    Google Scholar 

  7. Poole, E., Ortega, J.M.: Multicolor ICCG Methods for Vector Computers. SIAM J. Numer. Anal. 24(1987) 1394–1418

    Google Scholar 

  8. Ramdas, M., Kincaid, D.R.: Parallelizing ITPACKV 2D for the CRAY Y-MP. Proc. of IMACS Int. Conference on Iterative Methods in Linear Algebra 1992 323–347

    Google Scholar 

  9. Young, D.M.: Iterative Solution of Large Linear Systems. Academic Press New York 1971

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Faculty of Information Sciences, Hiroshima City University, 151-5 Ozuka, Numata-cho, Asaminami-ku, 731-31, Hiroshima, Japan

    Seiji Fujino

  2. Nissan Research Center, Nissan Motor Co., Ltd., 1 Natsushima-cho, Yokosuka, 237, Kanagawa, Japan

    Ryutaro Himeno

Authors
  1. Seiji Fujino
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ryutaro Himeno
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Jerzy Waśniewski Jack Dongarra Kaj Madsen Dorte Olesen

Rights and permissions

Reprints and permissions

Copyright information

© 1996 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Fujino, S., Himeno, R. (1996). A vectorization technique for a family of finite difference formulae and its performance evaluation. In: Waśniewski, J., Dongarra, J., Madsen, K., Olesen, D. (eds) Applied Parallel Computing Industrial Computation and Optimization. PARA 1996. Lecture Notes in Computer Science, vol 1184. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-62095-8_30

Download citation

  • DOI: https://doi.org/10.1007/3-540-62095-8_30

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-62095-2

  • Online ISBN: 978-3-540-49643-4

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation