Accelerating the reduction to upper Hessenberg, tridiagonal, and bidiagonal forms through hybrid GPU-based computing

We present a Hessenberg reduction (HR) algorithm for hybrid systems of homogeneous multicore with GPU accelerators that can exceed 25× the performance of the corresponding LAPACK algorithm running on current homogeneous multicores. This enormous acceleration is due to proper matching of algorithmic...

Full description

Saved in:

Bibliographic Details
Published in:	Parallel computing Vol. 36; no. 12; pp. 645 - 654
Main Authors:	Tomov, Stanimire, Nath, Rajib, Dongarra, Jack
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 01.12.2010
Subjects:	Accelerators Algorithms Bidiagonalization Computation Computer programs Dense linear algebra Eigenvalues GPUs Hessenberg reduction Hybrid computing Hybrid systems Reduction Running Tridiagonalization Two-sided factorizations Hybrid computing Bidiagonalization GPUs Tridiagonalization Two-sided factorizations Hessenberg reduction Dense linear algebra
ISSN:	0167-8191, 1872-7336
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!