LAPACK: A portable linear algebra library for high-performance computers

E Angerson,A Greenbaum,J Demmel,C Bischof,D Sorensen,S Hammarling,A Mckenney,J Du Croz,J Dongarra,Z Bai

doi:10.1109/superc.1990.129995

LAPACK: A portable linear algebra library for high-performance computers

E Angerson, A Greenbaum + Show 8 more

Open Access

https://doi.org/10.1109/superc.1990.129995

Copy DOI

Publication Date: Jan 1, 1990

Citations: 291

Affiliation: University of Tennessee at Knoxville, New York University, University of California, Berkeley, Argonne National Laboratory, Rice University, Numerical Algorithms Group (United Kingdom)

#Linear Least-squares Problems #Matrix-matrix Multiplication + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

The goal of the LAPACK project is to design and implement a portable linear algebra library for efficient use on a variety of high-performance computers. The library is based on the widely used LINPACK and EISPACK packages for solving linear equations, eigenvalue problems, and linear least-squares problems, but extends their functionality in a number of ways. The major methodology for making the algorithms run faster is to restructure them to perform block matrix operations (e.g., matrix-matrix multiplication) in their inner loops. These block operations may be optimized to exploit the memory hierarchy of a specific architecture. The LAPACK project is also working on new algorithms that yield higher relative accuracy for a variety of linear algebra problems. >

Full Text