Experiments with a task partitioning model for heterogeneous computing

David J. Lilja

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

4 Scopus citations

Abstract

One potentially promising approach for exploiting the best features of a variety of different computer architectures is to partition an application program to simultaneously execute on two or more different machines interconnected with a high-speed network. A fundamental problem with this heterogeneous computing, however, is the difficulty of partitioning an application program across the machines. This paper presents a partitioning strategy that relates the relative performance of two heterogeneous machines to the communication cost of transferring partial results across their interconnection network. Experiments are described that use this strategy to partition two different application programs across the sequential front-end processor of a Connection Machine CM-200, and its parallel back-end array.

Original languageEnglish (US)
Title of host publicationProceedings - Workshop on Heterogeneous Processing, VVHP 1993, Held in conjunction with the 7th International Parallel Processing Symposium
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages29-35
Number of pages7
ISBN (Electronic)0818635312, 9780818635311
DOIs
StatePublished - 1993
Event1993 Workshop on Heterogeneous Processing, VVHP 1993, Held in conjunction with the 7th International Parallel Processing Symposium - Newport, United States
Duration: Apr 13 1993 → …

Publication series

NameProceedings - Workshop on Heterogeneous Processing, VVHP 1993, Held in conjunction with the 7th International Parallel Processing Symposium

Conference

Conference1993 Workshop on Heterogeneous Processing, VVHP 1993, Held in conjunction with the 7th International Parallel Processing Symposium
Country/TerritoryUnited States
CityNewport
Period4/13/93 → …

Bibliographical note

Funding Information:
This work was supported in part by Army Research Office contract number DAAL03-89-C-0038 with the University of Minnesota Army High Performance Computing Research Center. The experiments with the shallow program were performed by Jonathan Schmitt while supported by a Minnesota Supercomputer Institute Sum- mer Undergraduate Intemship.

Publisher Copyright:
© 1993 IEEE.

Fingerprint

Dive into the research topics of 'Experiments with a task partitioning model for heterogeneous computing'. Together they form a unique fingerprint.

Cite this