Skip to main navigation Skip to search Skip to main content

Performance of distributed sparse Cholesky factorization with pre-scheduling

  • Rutgers - The State University of New Jersey, New Brunswick
  • IBM

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

We propose a sparse Cholesky factorization scheme based on a static task and communication schedule generated by symbolic preprocessing of the intercolumn dependencies. This information is used to reduce the overheads of maintaining data structures and the communication costs incurred in message passing during the numerical factorization step. We introduce three primitives which encapsulate these optimizations and alleviate the user's programming effort. Performance results obtained on the iPSC/860 show that an implementation using these primitives results in 30% to 40% savings over one that does not use any static communication structure information.

Original languageEnglish
Title of host publicationProceedings of the 1992 ACM/IEEE conference on Supercomputing, Supercomputing 1992
EditorsRobert Werner
PublisherAssociation for Computing Machinery
Pages52-61
Number of pages10
ISBN (Electronic)0818626305
DOIs
StatePublished - Dec 1 1992
Event1992 ACM/IEEE conference on Supercomputing, Supercomputing 1992 - Minneapolis, United States
Duration: Nov 16 1992Nov 20 1992

Publication series

NameProceedings of the International Conference on Supercomputing
VolumePart F129723

Conference

Conference1992 ACM/IEEE conference on Supercomputing, Supercomputing 1992
Country/TerritoryUnited States
CityMinneapolis
Period11/16/9211/20/92

Fingerprint

Dive into the research topics of 'Performance of distributed sparse Cholesky factorization with pre-scheduling'. Together they form a unique fingerprint.

Cite this