Dyn-MPI: Supporting MPI on non dedicated clusters

D. Brent Weatherly, David K. Lowenthal, Mario Nakazawa, Franklin Lowenthal

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

5 Scopus citations


Distributing data is a fundamental problem in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in environments where the participating nodes are not dedicated to a parallel application. We are investigating the data distribution problem in non dedicated environments in the context of explicit message-passing programs. To address this problem, we have designed and implemented an extension to MPI called Dynamic MPI (Dyn-MPI). The key component of Dyn-MPI is its run-time system, which efficiently and automatically redistributes data on the fly when there are changes in the application or the underlying environment. Dyn-MPI supports efficient memory allocation, precise measurement of system load and computation time, and node removal. Performance results show that programs that use Dyn-MPI execute efficiently in non dedicated environments, including up to almost a three-fold improvement compared to programs that do not redistribute data and a 25% improvement over standard adaptive load balancing techniques.

Original languageEnglish (US)
Title of host publicationProceedings of the 2003 ACM/IEEE Conference on Supercomputing, SC 2003
StatePublished - 2003
Event2003 ACM/IEEE Conference on Supercomputing, SC 2003 - Phoenix, AZ, United States
Duration: Nov 15 2003Nov 21 2003

Publication series

NameProceedings of the 2003 ACM/IEEE Conference on Supercomputing, SC 2003


Other2003 ACM/IEEE Conference on Supercomputing, SC 2003
Country/TerritoryUnited States
CityPhoenix, AZ

ASJC Scopus subject areas

  • Software


Dive into the research topics of 'Dyn-MPI: Supporting MPI on non dedicated clusters'. Together they form a unique fingerprint.

Cite this