MIMD Supercomputers for Numerical Applications

  • Ph. Berger
  • D. Comte
  • Ch. Fraboul


The numerical simulation becomes increasingly important day by day in domains which are more and more diverse: aerodynamics, hydrodynamics, structural analysis, nuclear weapons, weather prediction, etc. [1]. The needs in computation power are unmeasurable and far from the possibilities available today [2][3]. The super­computers which are on the market at the present time are the result of technological progress. Although this is important in itself, we have not yet been able to bridge the gap between the needs and the possibilities offered. The supercomputer designers have realized that the solution to this problem lies with the coming of the multiprocessors.


Parallel Application Computation Task System Task Memory Hierarchy Array Processor 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    D. Comte, J.C. Syre, Les supercalculateurs, Article de la Recherche, vol. 14, n°147 (sept. 1983 ).Google Scholar
  2. [2]
    P.D. Lax, Report of the Panel on Large Scale Computing in Science and Engineering, NSF, Washington (Dec. 26, 1982 ).Google Scholar
  3. [3]
    A major international conference on supercomputers, April 1984.Google Scholar
  4. [4]
    D. Kuck, Structure of computers and computation, Vol. 1, J. Wiley & Sons.Google Scholar
  5. [5]
    Ph. Treleaven, Future computers: logics, data flow,... control flow?, Computer (March 1984).Google Scholar
  6. [6]
    D. Comte, N. Hifdi, J.C. Syre, The Data driven LAU multiprocessor system: results and perspectives, IFIP, Melbourne (1980).Google Scholar
  7. [7]
    C. Rechbach, Numerical calculation of three-dimensional unsteady flows with vortex sheets, Congrés AIAA sur les Sciences Aéronautiques, Huntville, U.S.A., (January 1978).Google Scholar
  8. [8]
    S.G. Hotovy, L.J. Dickson, Evaluation of a vectorizable 2D transonic finite difference algorithm, 17th Aerospace Sciences Meeting, New Orleans (January 1979) .Google Scholar
  9. [9]
    Ph. Berger, P. Brouaye, J.C. Syre, A mesh coloring method for efficient MIMD processing in finite element problems, Int. Conf. on Parallel Processing, Bellaire, Michigan (August 1982)Google Scholar
  10. [10]
    A. George, W.G. Poole, R.G. Voigt, Incomplete nested dissection for solving n by n grid problems. SIAM Num. Analysis, Vol. 15, n°4, (August 1978).Google Scholar
  11. [11]
    A. Lichnewski, Sur la résolution de systèmes lineaires issus de la méthode des éléments finis par une machine multiprocesseur, Rapport Technique INRIA, n°119 (February 1982).Google Scholar
  12. [12]
    G.M. Baudet, Asynchronous iterative methods for multi-processors, Rapport Technique Carnegie-Mellon, (1976) .Google Scholar
  13. [13]
    J.C. Miellou, Algorithmes de relaxation chaotique à retards, RAIRO, Ri (1975).Google Scholar
  14. [14]
    Ph. Berger, Etude, expression et évaluation du parallelisme dans la résolution d’équations aux dérivées partielles sur une architecture MIMD à mémoires locales, Thése Doc.-Ing., ENSAE (December 1982).Google Scholar
  15. [15]
    C.P. Arnold, M.I. Parr, M.B. Dewe, An efficient parallel algorithm for the solution of large sparse linear matrix equations, IEEE computers, Vol. C-32, n°3, (March 1983).Google Scholar
  16. [16]
    M. Adelantado, D. Comte, P. Siron, J.C. Syre, A MIMD supercomputer system for large scale numerical applications, IFIP, Paris (1983).Google Scholar
  17. [17]
    M. Adelantado, D. Comte, S. Donnet, P. Siron, Description du langage XANADU, Rapport final n°3/3210/DERI, Onera-Cert, 2 avenue E. Belin, 31055 Toulouse Cedex, France.Google Scholar
  18. [18]
    Ph. Berger, Parallélisation d’algorithmes numériques, Rapports n°1/3191/DERI et n°3/3191/DERI, Onera-Cert, Toulouse.Google Scholar
  19. [19]
    J.P. Boisseau, M. Enselme, D. Guiraud, P. Leca, Potentiality assessment of a parallel architecture for the solution of partial differential equations, La Recherche Aérospatiale 1982–1.Google Scholar
  20. [20]
    Ch. Fraboul, N. Hifdi, LESTAP: a language for expressing and synchronization of task on a multiarray-processor, Minipaper 1983 FPS Users Meeting, Monterey, Canada (April 24–27, 1983 ).Google Scholar
  21. [21]
    Ch. Fraboul, N. Hifdi, LESTAP: Expression et mise en oeuvre d’applications paralléles, Rapport n°1/3192, Onera-Cert, Toulouse.Google Scholar
  22. [22]
    Ch. Fraboul, D. Guiraud, N. Hifdi, Implementing parallel applications on a multi-array-processor system, Proc. 1st Int. Symp. on a Vector and Parallel Computing in Scientific Applications, Paris (March 1983).Google Scholar
  23. [23]
    J.P. Boisseau, A. Cosnuau, M. Enselme, Ch. Fraboul, D. Guiraud, N. Hifdi, Le langage LESTAP et son application à la parallélisation d’applications d’aérodynamique, Rapport n°11/3479CY Onera.Google Scholar
  24. [24]
    P. Leca, Ph. Roy, Numerical simulation of turbulence on minisystems with attached processors, La Recherche Aérospatiale 1983–4.Google Scholar
  25. [25]
    F. Bourdel, Pré-étude algorithmique de la méthode des sous-domaines de Schwarz, Rapport n°1/3612 Onera-Cert GAN.Google Scholar
  26. [26].
    G.S. Patterson, Large scale scientific computing: future directions, Comp. Phys. Com. 26 (1982).Google Scholar
  27. [27]
    M. Enselme, Ch. Fraboul, P. Leca, A MIMD architecture system for PDE numerical simulation, 5th IMACS Int. Symp. on computer methods for partial differential equations, Bethlehem, Pennsylvania, U.S.A., (June 19–22, 1984 ).Google Scholar
  28. [28]
    Ch. Fraboul, Evaluation d’évolutions matérielles et logicielles du système parallèle SEL32/multiAP120B, Rapport n°2/3215, Onera-Cert.Google Scholar

Copyright information

© Plenum Press, New York 1985

Authors and Affiliations

  • Ph. Berger
    • 1
  • D. Comte
    • 1
  • Ch. Fraboul
    • 1
  1. 1.Département d’InformatiqueONERA-CERTToulouse CedexFrance

Personalised recommendations