Advertisement

Parallel GENESIS for Large-Scale Modeling

  • Nigel H. Goddard
  • Greg Hood

Abstract

Simulations of computational models are limited in size and speed by the power and capacity of the computational platform. We have parallelized the GENESIS object-oriented neural simulator [1] for networked workstations, multiprocessors and massively parallel supercomputers. These can provide two orders of magnitude increase in the size of the models that can be effectively simulated. As larger models are partitioned across many processors, interprocessor communication can limit the effective speedup obtainable. This suggests two classes of problems that may benefit most from parallel simulation: parameter searching and network models.

Keywords

Source Node Destination Node Parallel Simulation Interprocessor Communication Parallel Platform 
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. [1]
    J.M. Bower and D. Beeman. The Book of Genesis. Springer-Verlag, Santa Clara, CA, 1994.Google Scholar
  2. [2]
    A. Geist, A. Beguelin, J. Dongarra, W. Jiang, R. Manchek, and V.S. Sunderam. Parallel Virtual Machine. MIT Press, Cambridge, MA, 1994.Google Scholar
  3. [3]
    M. Vanier and D. Beeman. Constructing neural circuits and networks. In J.M. Bower and D. Beeman, editors, The Book of Genesis, chapter 17. Springer-Verlag, Santa Clara, CA, 1994.Google Scholar
  4. [4]
    M. Vanier and J.M. Bower. A comparison of automated parameter-searching methods for neural models. In J.M. Bower, editor. Computational Neuroscience: Trends in Research 1995, Monterey, CA, 1995. Academic Press.Google Scholar

Copyright information

© Springer Science+Business Media New York 1997

Authors and Affiliations

  • Nigel H. Goddard
    • 1
  • Greg Hood
    • 1
  1. 1.Pittsburgh Supercomputing CenterPittsburghUSA

Personalised recommendations