Parallel Computation in Mechanics

  • Gary M. Johnson


The success of numerical simulation as an independent approach to the solution of engineering problems requires computing capability far exceeding that which is presently available. In this paper, the computing requirements posed by challenging problems in mechanics are examined and contrasted with contemporary supercomputer resources. Of the means available to help fill the gap between the demands of scientific computation and the performance level of present-generation supercomputer systems, parallel processing appears to have the greatest potential for near-term success. Typical parallel computer architectures are reviewed and categorized. Philosophies of parallel processing are distinguished by the number and size of the parallel tasks which they employ. Selected engineering problems are examined for parallelism inherent at the physical level. Typical algorithms and their mappings onto parallel architectures are discussed. Computational examples are presented to document the performance of scientific applications on present-generation parallel processors. Projections are made concerning future algorithms and machine architectures.


Parallel Processing Parallel Architecture Parallel Processor Processor Figure Synchronization Overhead 
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.
    Wilson, K. G.: “Grand Challenges to Computational Science,” Cornell University, May 1987.Google Scholar
  2. 2.
    Peterson, V. L., J. O. Arnold: “The Impact of Supercomputers on Experimentation: A View from a National Laboratory,” ASEE Annual Conference Proceedings, 1985.Google Scholar
  3. 3.
    Hasslacher, B.: “Discrete Fluids,” Los Alamos Science, No. 15, Special Issue, 1987, pp. 175–217.Google Scholar
  4. 4.
    Worlton, J.: “Toward a Science of Parallel Computation”,Computational Mechanics — Advances and Trends, AMD Vol. 75, ASME, New York, 1987, pp. 23–35.Google Scholar
  5. 5.
    Johnson, G. M. M., “Requirements for Parallel Processing in Scientific Computation,” Third International Conference on Supercomputing, Boston, Massachusetts, May 1988.Google Scholar
  6. 6.
    Weeks, C. L., “Concurrent Extensions to the Fortran Language,” First World Congress on Computational Mechanics, Austin, Texas, September 1986.Google Scholar
  7. 7.
    Dongarra, J. J., D. C. Sorenson, “SCHEDULE: An Environment for Developing Transportable Explicitly Parallel Codes in Fortran,” Third SIAM Conference on Parallel Processing in Scientific Computing, Los Angeles, California, December 1987.Google Scholar

Copyright information

© Plenum Press, New York 2011

Authors and Affiliations

  • Gary M. Johnson
    • 1
  1. 1.San Diego Supercomputer CenterSan DiegoUSA

Personalised recommendations