Parallel computation of the singular value decomposition on tree architectures

Date

1993

Authors

Zhou, B. B.
Brent, Richard P.

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

We describe three new Jacobi orderings for parallel computation of SVD problems on tree architectures. The first ordering uses the high bandwidth of a perfect binary fat-tree to minimise global interprocessor communication costs. The second is a new ring ordering which may be implemented efficiently on an ordinary binary tree. By combining these two orderings, an efficient new ordering, well suited for implementation on the Connection Machine CM5, is obtained.

Description

Keywords

tree architectures, SVD, binary tree, binary fat-tree, fat-tree orderings, ring orderings, hybrid ordering, Jacobi ordering algorithms

Citation

Source

Type

Working/Technical Paper

Book Title

Entity type

Access Statement

License Rights

DOI

Restricted until

Downloads