Advertisement

Using PMD to Parallel-Solve Large-Scale Navier-Stokes Equations. Performance Analysis on SGI/CRAY-T3E Machine

  • Jalel Chergui
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1697)

Abstract

PMD (Parallel Multi-domain Decomposition) is an MPI based Fortran 90 module which objective is to parallel-solve positive definite linear elliptic second order equations. It has been used to solve unsteady Navier-Stokes equations in order to simulate an axisymmetric incom- pressible viscous fluid flow inside a centrifugal pump. In this paper, we will present a brief description of the implementation and will discuss the performance measurements obtained on the SGI/CRAY T3E parallel machine.

Keywords

Centrifugal Pump Helmholtz Problem Helmholtz Operator Total Elapse Time Processor Number 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. CHE 98.
    J. Chergui & al., PMD: A Parallel Fortran 90 Module to Solve Elliptic Linear Second Order Equations. Calculateurs Parallèles, Hermes publisher, 75004 Paris, France. Vol. 10, No 6, Dec. 1998.Google Scholar
  2. CHER 97.
    J. Chergui, Performance ratio analysis between CRAY T3E and T3D machines (in French). Calculateurs Paralléles, Hermes publisher, 75004 Paris, France. Vol. 9. No 4. December 1997.Google Scholar
  3. GRO 96.
    W. Gropp, S. Huss-Lederman, A. Lumsdaine, E. Lusk, B. Nitzberg, W. Safir, M. Snir, MPI: The Complete Reference. Volume 1, The MPI Core. The MIT Press Cambridge, Massachusetts London, England. 1998.Google Scholar
  4. JAC 98.
    R. Jacques, P. Le Qur, O. Daube, Parallel Numerical Simulation of Turbulent Flows in Closed Rotor-Stator Cavities. Notes on Numerical Fluid Mechanics, Vieweg Publisher. 1998Google Scholar
  5. LAS 87.
    P. Lascaux, R. ThÉodor, Analyse numéerique matricielle appliquée l’art de l’ingénieur. Tome2, éd. Masson, Paris, France. 1987.Google Scholar
  6. QUA 90.
    A. Quarteroni, Domain Decomposition Method for the Numerical Solution of Partial Di_erential Equations. UMSI 90/246. University of Minnesota Supercomputer Institute. Research report. December 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Jalel Chergui
    • 1
  1. 1.Institut du Dveloppement et des Ressources en Informatique ScientfiqueCNRS/IDRISOrsay cedexFrance

Personalised recommendations