Scalable Basic Linear Algebra Subprograms (sB_BLAS)

Abstract

We present straight forward techniques for a highly efficient, scalable implementation of common matrix-matrix operations generally known as the Level 3 Basic Linear Algebra Subprograms (BLAS). This work builds on our recent discovery of a parallel matrix-matrix multiplication implementation, which has yielded superior performance, and requires little work space. We show that the techniques used for the matrix-matrix multiplication naturally extend to all important level 3 BLAS and thus this approach becomes an enabling technology for efficient parallel implementation of these routines and libraries that use BLAS.

DateOfInformation

Wed Apr 9 09:16:21 1997

Domain

Numerical Programs and Routines!Linear algebra

Keyword

GAMS D1. Elementary vector and matrix operations

Name

Scalable Basic Linear Algebra Subprograms (sB_BLAS)

TargetEnvironment

Not Available

Webpage

http://www.cs.utexas.edu/users/rvdg/software.html

ContactIs

Robert A. van de Geijn