Generalized framework for global communication optimization

Mahmut Kandemir, P. Banerjee, A. Choudhary, J. Ramanujam, N. Shenoy

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

3 Scopus citations

Abstract

In distributed-memory message-passing architectures reducing communication cost is extremely important. In this paper, we present a technique to optimize communication globally. Our approach is based on a combination of linear algebra framework and dataflow analysis, and can take arbitrary control flow into account. The distinctive features of the algorithm are its accuracy in keeping communication set information and its support for general alignments and distributions including block-cyclic distributions. The method is currently being implemented in the PARADIGM compiler. The preliminary results show that the technique is effective in reducing both number as well as volume of the communication.

Original languageEnglish (US)
Title of host publicationProceedings of the International Parallel Processing Symposium, IPPS
Editors Anon
Pages69-73
Number of pages5
DOIs
StatePublished - Jan 1 1998
EventProceedings of the 1998 12th International Parallel Processing Symposium and 9th Symposium on Parallel and Distributed Processing - Orlando, FL, USA
Duration: Mar 30 1998Apr 3 1998

Publication series

NameProceedings of the International Parallel Processing Symposium, IPPS
ISSN (Print)1063-7133

Other

OtherProceedings of the 1998 12th International Parallel Processing Symposium and 9th Symposium on Parallel and Distributed Processing
CityOrlando, FL, USA
Period3/30/984/3/98

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Generalized framework for global communication optimization'. Together they form a unique fingerprint.

  • Cite this

    Kandemir, M., Banerjee, P., Choudhary, A., Ramanujam, J., & Shenoy, N. (1998). Generalized framework for global communication optimization. In Anon (Ed.), Proceedings of the International Parallel Processing Symposium, IPPS (pp. 69-73). (Proceedings of the International Parallel Processing Symposium, IPPS). https://doi.org/10.1109/IPPS.1998.669892