Advertisement

The Journal of Supercomputing

, Volume 10, Issue 4, pp 349–369 | Cite as

Multiprocessing ocean circulation: Modeling, implementation, and performance on the Intel Paragon

  • Ishfaq Ahmad
  • Ka-Cheong Leung
  • Hsiao-Ming Hsu
Article
  • 43 Downloads

Abstract

In this paper we present the modeling and implementation of a grand challenge problem in the field of scientific computation: the primitive-equation numerical ocean circulation model. We present the mathematical formulation of the model and propose a scheme for its parallel implementation. Optimizations are made through collective communications and various partitioning schemes. In our experiments, which use up to 100 processors on the Intel Paragon parallel computer, the proposed strategy yields an encouraging speedup and exhibits a sustained scalability with increasing problem and machine sizes. We consider barotropic continental shelf waves in a periodic channel as a test problem. The model has numerous applications in environmental studies and ocean sciences.

Keywords

High-performance computing parallel applications partial differential equations partitioning numerical algorithms numerical ocean circulation scientific computation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. Beckmann and D. B. Haidvogel. Numerical simulation of flow around a tall isolated seamount. Part I: Problem formulation and model accuracy. Journal of Physical Oceanography, 23:1736–1753, 1993.Google Scholar
  2. 2.
    R. Berrendorf, H. C. Burg, U. Detert, R. Esser, M. Gerndt, and R. Knecht. Intel Paragon XP/S—Architecture, software environment, and performance. Technical Report KFA-ZAM-IB-9409, Central Institute for Applied Mathematics, May 1994.Google Scholar
  3. 3.
    R. Bleck and D. B. Boudra. Wind-driven spin-up in eddy-resolving ocean models formulated in isopycnic and isobaric coordinates. Journal of Geophysical Research, 91:7611–7621, 1986.Google Scholar
  4. 4.
    R. Bleck, H. P. Hanson, D. Hu, and E. B. Kraus. Mixed layer-thermocline interaction in a threedimensional isopycnic coordinate model. Journal of Physical Oceanography, 19:1417–1439, 1989.Google Scholar
  5. 5.
    K. A. Bryan. numerical investigation of a nonlinear model of a wind-driven ocean. Journal of Atmospheric Sciences, 20:594–606, 1963.Google Scholar
  6. 6.
    D. C. Chapman and D. B. Haidvogel. Formation of Taylor caps over a tall isolated seamount in a stratified ocean. Geophysical and Astrophysical Fluid Dynamics, 64:31–65, 1992.Google Scholar
  7. 7.
    D. C. Chapman and D. B. Haidvogel. Generation of internal lee waves trapped over a tall seamount. Geophysical and Astrophysical Fluid Dynamics, 69:33–54, 1993.Google Scholar
  8. 8.
    D. C. Chapman and S. Lentz. Trapping of a coastal density front by the bottom boundary layer. Journal of Physical Oceanography, 24:1464–1479, 1994.Google Scholar
  9. 9.
    T. L. Freeman and C. Phillips. Parallel Numerical Algorithms. Prentice-Hall, 1992.Google Scholar
  10. 10.
    G. Gawarkiewicz. Steady wind forcing of a density front over a circular bank. Journal of Marine Research, 51:109–134, 1993.Google Scholar
  11. 11.
    G. Gawarkiewicz and D. C. Chapman. Formation and maintenance of shelfbreak fronts in an unstratified flow. Journal of Physical Oceanography, 21:1225–1239, 1991.Google Scholar
  12. 12.
    G. Gawarkiewicz and D. C. Chapman. The role of stratification in the formation and maintenance of a shelf-break front. Journal of Physical Oceanography, 22:753–772, 1992.Google Scholar
  13. 13.
    D. B. A. Haidvogel, Beckmann, D. C. Chapman, and R. Q. Lin. Numerical simulation of flow around a tall isolated seamount. Part II: Resonant generation of trapped waves. Journal of Physical Oceanography, 23:2373–2391, 1993.Google Scholar
  14. 14.
    D. B. Haidvogel, A. Beckmann, and K. S. Hedstrom. Dynamical simulations of filament formation and evolution in the coastal transition zone. Journal of Geophysical Research, 96:15017–15040, 1991.Google Scholar
  15. 15.
    D. B. Haidvogel, J. L. Wilkin, and R. Young. A semi-spectral primitive equation ocean circulation model using vertical sigma and orthogonal curvilinear horizontal coordinates. Journal of Computational Physics, 94:151–185, 1991.Google Scholar
  16. 16.
    N. Heaps, ed. Three-Dimensional Coastal Ocean Models. American Geophysical Union, 208, 1987.Google Scholar
  17. 17.
    A. J. Hermann and H. Hsu. A vertical coordinate mapping technique for semispectral primitive equation models of oceanic circulation. Journal of Atmospheric and Oceanic Technology, 10:381–396, 1993.Google Scholar
  18. 18.
    E. E. Hoffmann, K. S. Hedstrom, J. R. Moisan, D. B. Haidvogel, and D. L. Mackas. Use of simulated drifter tracks to investigate general transport patterns and residence times in the coastal transition zone. Journal of Geophysical Research, 96:15041–15052, 1991.Google Scholar
  19. 19.
    H. Hsu. Numerical studies of Arctic shelf circulation for wintertime dense water and summertime light water (invited). In Arctic System Science (ARCSS)—Ocean Atmosphere Ice Interaction (OAII) Modelling Workshop, Monterey, Calif., July 13–14, 1992.Google Scholar
  20. 20.
    H. Hsu, R. C. Beardsley, and J. F. Price. Including surface and bottom boundary layer physics in a primitive-equation model to study coastal circulations. In AGU 1992 Ocean Science Meeting, New Orleans, January 27–31, 1992.Google Scholar
  21. 21.
    Intel Supercomputer Systems Division. Paragon User's Guide. Intel Corporation, June 1994.Google Scholar
  22. 22.
    J. L. Wilkin and D. C. Chapman. Scattering of coastal-trapped waves by irregularities in coastline and topography. Journal of Physical Oceanography, 20:396–421, 1990.Google Scholar

Copyright information

© Kluwer Academic Publishers 1996

Authors and Affiliations

  • Ishfaq Ahmad
    • 1
  • Ka-Cheong Leung
    • 1
  • Hsiao-Ming Hsu
    • 2
  1. 1.Department of Computer ScienceThe Hong Kong University of Science and TechnologyKowloonHong Kong
  2. 2.National Center for Atmospheric ResearchBoulderUSA

Personalised recommendations