Adaptive load balancing for MPI programs

Milind Bhandarkar, L. V. Kalé, Eric de Sturler, Jay Hoeflinger

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


Parallel Computational Science and Engineering (CSE) ap- plications often exhibit irregular structure and dynamic load patterns. Many such applications have been developed using MPI. Incorporating dynamic load balancing techniques at the application-level involves significant changes to the design and structure of applications. On the other hand, traditional run-time systems for MPI do not support dynamic load balancing. Object-based parallel programming languages, such as Charm++ support efficient dynamic load balancing using object migration. However, converting legacy MPI applications to such object-based paradigms is cumbersome. This paper describes an implementation of MPI, called Adaptive MPI (AMPI) that supports dynamic load balancing for MPI applications. Conversion from MPI to this platform is straightforward even for large legacy codes. We describe our positive experience in converting the component codes ROCFLO and ROCSOLID of a Rocket Simulation application to AMPI.

Original languageEnglish (US)
Title of host publicationComputational Science – ICCS 2001 - International Conference, Proceedings
EditorsVassil N. Alexandrov, Jack J. Dongarra, Benjoe A. Juliano, Rene S. Renner, C.J. Kenneth Tan
Number of pages10
ISBN (Print)3540422331, 9783540422334
StatePublished - 2001
EventInternational Conference on Computational Science, ICCS 2001 - San Francisco, United States
Duration: May 28 2001May 30 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


OtherInternational Conference on Computational Science, ICCS 2001
Country/TerritoryUnited States
CitySan Francisco

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'Adaptive load balancing for MPI programs'. Together they form a unique fingerprint.

Cite this