Advertisement

Parallel computational design of NJR global climate models

  • Yukio Tanaka
  • Nobuhisa Goto
  • Masayuki Kakei
  • Takahiro Inoue
  • Yonejiro Yamagishi
  • Masayuki Kanazawa
  • Hisashi Nakamura
VI Earth Simulator
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1615)

Abstract

An overview of a parallel computational design for global climate models, called NJR, are described. The models consist of two atmospheric general circulation models: the spectral and grid point atmospheric model, and an ocean general circulation model. The spectral atmospheric model descritizes the equations by using orthogonal spherical harmonic function. The computation domain is decomposed into subdomains along latitudes so that FFT is calculated in parallel without data communication. The complete-data-exchange type of communication is performed after the FFT and Legendre transformation is executed in parallel along longitude without data communication. The grid point atmospheric model and the ocean circulation model employ finite-difference method. The computation domain is decomposed into subdomains along latitude and the boundary data of each subdomain are exchanged between neighboring subdomains in each time step. The NJR global climate model is run efficiently on high performance parallel computers.

Keywords

parallel computers general circulation NJR 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Numguti, A., Takahashi, M., Nakajima, T., Sumi, A.: CGER's Supercomputer Monograph Report. Center for Global Environmental Research, National Institute for Environmental Studies. 3, (1997) 1–48Google Scholar
  2. 2.
    Blumberg, A.F., and Mellor, G.L.: A coastal ocean numerical model. in Mathematical Modeling of Estarine Physics, Proc. Int. Symp. Hamburg, Aug. 1978, edited by J. Sunderman and K.-P Holtz, pp. 203–214, Springer-Verlag, Berlin, 1980.Google Scholar
  3. 3.
    Orszag, S.A.: Transform method for the calculation of vector-coupled sums: Application to the spectral form of the vorticity equation. J. Atmos. Sci., 27, (1970) 890–895.CrossRefGoogle Scholar
  4. 4.
    Eliasen, E., Machenhauer, B., Rasmussen, E.: On a numerical method for integration of the hydrodynamic equations with a spectral representation of the horizontal fields, Institute for Teoretisk Meteorologi, Kobenhavns Universitet, Denmark, Rept. No. 2, (1970) 35pp.Google Scholar
  5. 5.
    Arakawa, A.: Computational design for long-term numerical integration of the equations of fluid motions. J. Comp. Physics. 1, (1966) 119–143.CrossRefGoogle Scholar
  6. 6.
    Matsuno, T.: Numerical integration of the primitive equations by a simulated backward difference method. J. Meteor. Soc. Japan, Ser 2,44, (1966) 76–84.Google Scholar
  7. 7.
    Simons, T.J.: Verification of numerical models of Lake Ontario. Part I, circulation in spring and early summer, J. Phys. Oceanogr. 4, (1974), 507–523.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Yukio Tanaka
    • 1
  • Nobuhisa Goto
    • 1
  • Masayuki Kakei
    • 1
  • Takahiro Inoue
    • 1
  • Yonejiro Yamagishi
    • 1
  • Masayuki Kanazawa
    • 1
  • Hisashi Nakamura
    • 1
  1. 1.Research Organization for Information Science & Technology (RIST)TokyoJapan

Personalised recommendations