Advertisement

Theoretical Chemistry Accounts

, Volume 99, Issue 3, pp 197–206 | Cite as

A scalable divide-and-conquer algorithm combining coarse and fine-grain parallelization

  • Sor Koon Goh
  • Carlos P. Sosa
  • Alain St-Amant
Regular article

Abstract.

We describe an efficient algorithm for carrying out a “divide-and-conquer” fit of a molecule's electronic density on massively parallel computers. Near linear speedups are achieved with up to 48 processors on a Cray T3E, and our results indicate that similar efficiencies could be attained on an even greater number of processors. To achieve optimum efficiency, the algorithm combines coarse and fine-grain parallelization and adapts itself to the existing ratio of processors to subsystems. The subsystems employed in our divide-and-conquer approach can also be made smaller or bigger, depending on the number of processors available. This allows us to further reduce the wallclock time and improve the method's overall efficiency. The strategies implemented in this paper can be extended to any other divide-and-conquer method used within an ab initio, density functional, or semi-empirical quantum mechanical program.

Key words: Scalable algorithms Supercomputing Quantum mechanics Divide-and-conquer Large molecules 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Sor Koon Goh
    • 1
  • Carlos P. Sosa
    • 2
  • Alain St-Amant
    • 1
  1. 1.Department of Chemistry, University of Ottawa, 10 Marie Curie, Ottawa, Ontario, Canada K1N 6N5CA
  2. 2.Silicon Graphics, Inc./Cray Research, Inc., 655E Lone Oak Drive, Eagan, MN 55121, USAUS

Personalised recommendations