A static mapping of sparse Cholesky factorization onto distributed memory architectures

Thumbnail Image

Date

1995

Journal Title

Journal ISSN

Volume Title

Publisher

Texas A&M University

Abstract

The results of a new static mapping of the sparse Cholesky factorization problem onto two distributed memory architectures are presented within this report. This mapping technique is based on the ideas of both a critical path, and the fact that not all tasks have the same computational workload. The proposed mapping technique makes use of the elimination tree which is derived from the column dependencies in the sparse definite positive input matrix A. The mapping solution presented will be compared with both simple methods which do not rely on the elimination tree, and against other well known mapping techniques. Performance results favor the proposed weighted critical path (WCP) mapping.

Description

Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to [email protected], referencing the URI of the item.
Includes bibliographical references.
Issued also on microfiche from Lange Micrographics.

Keywords

computer science., Major computer science.

Citation