Quantifying the effectiveness of load balance algorithms

Olga Pearce, Todd Gamblin, Bronis R. De Supinski, Martin Schulz, Nancy M. Amato

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Load balance is critical for performance in large parallel applications. An imbalance on today's fastest supercomputers can force hundreds of thousands of cores to idle, and on future exascale machines this cost will increase by over a factor of a thousand. Improving load balance requires a detailed understanding of the amount of computational load per process and an application's simulated domain, but no existing metrics sufficiently account for both factors. Current load balance mechanisms are often integrated into applications and make implicit assumptions about the load. Some strategies place the burden of providing accurate load information, including the decision on when to balance, on the application. Existing application-independent mechanisms simply measure the application load without any knowledge of application elements, which limits them to identifying imbalance without correcting it. Our novel load model couples abstract application information with scalable measurements to derive accurate and actionable load metrics. Using these metrics, we develop a cost model for correcting load imbalance. Our model enables comparisons of the effectiveness of load balancing algorithms in any specific imbalance scenario. Our model correctly selects the algorithm that achieves the lowest runtime in up to 96% of the cases, and can achieve a 19% gain over selecting a single balancing algorithm for all cases.

Original languageEnglish (US)
Title of host publicationICS'12 - Proceedings of the 2012 ACM International Conference on Supercomputing
Pages185-194
Number of pages10
DOIs
StatePublished - 2012
Event26th ACM International Conference on Supercomputing, ICS'12 - San Servolo Island, Venice, Italy
Duration: Jun 25 2012Jun 29 2012

Publication series

NameProceedings of the International Conference on Supercomputing

Conference

Conference26th ACM International Conference on Supercomputing, ICS'12
CountryItaly
CitySan Servolo Island, Venice
Period6/25/126/29/12

Keywords

  • Framework
  • Load balance
  • Modeling
  • Performance
  • Simulation

ASJC Scopus subject areas

  • Computer Science(all)

Fingerprint Dive into the research topics of 'Quantifying the effectiveness of load balance algorithms'. Together they form a unique fingerprint.

  • Cite this

    Pearce, O., Gamblin, T., De Supinski, B. R., Schulz, M., & Amato, N. M. (2012). Quantifying the effectiveness of load balance algorithms. In ICS'12 - Proceedings of the 2012 ACM International Conference on Supercomputing (pp. 185-194). (Proceedings of the International Conference on Supercomputing). https://doi.org/10.1145/2304576.2304601