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

E. Anderson; Z. Bai; J. Dongarra; A. Greenbaum; A. McKenney; J. Du Croz; S. Hammarling; J. Demmel; C. Bischof; D. Sorensen

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

E. Anderson, Z. Bai, J. Dongarra, A. Greenbaum, A. McKenney, J. Du Croz, S. Hammarling, J. Demmel, C. Bischof, D. Sorensen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

258 Scopus citations

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.

Original language	English
Title of host publication	Proc Supercomput 90
Publisher	Publ by IEEE
Pages	2-11
Number of pages	10
ISBN (Print)	0818620560
State	Published - 1990
Externally published	Yes
Event	Proceedings of Supercomputing '90 - Duration: Nov 12 1990 → Nov 16 1990

Publication series

Name	Proc Supercomput 90

Conference

Conference	Proceedings of Supercomputing '90
Period	11/12/90 → 11/16/90

Cite this

@inproceedings{5aac3c5241244e589f63a93e4e88642b,

title = "LAPACK: A portable linear algebra library for high-performance computers",

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.",

author = "E. Anderson and Z. Bai and J. Dongarra and A. Greenbaum and A. McKenney and {Du Croz}, J. and S. Hammarling and J. Demmel and C. Bischof and D. Sorensen",

year = "1990",

language = "English",

isbn = "0818620560",

series = "Proc Supercomput 90",

publisher = "Publ by IEEE",

pages = "2--11",

booktitle = "Proc Supercomput 90",

note = "Proceedings of Supercomputing '90 ; Conference date: 12-11-1990 Through 16-11-1990",

}

TY - GEN

T1 - LAPACK

T2 - Proceedings of Supercomputing '90

AU - Anderson, E.

AU - Bai, Z.

AU - Dongarra, J.

AU - Greenbaum, A.

AU - McKenney, A.

AU - Du Croz, J.

AU - Hammarling, S.

AU - Demmel, J.

AU - Bischof, C.

AU - Sorensen, D.

PY - 1990

Y1 - 1990

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=0025536635&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0025536635

SN - 0818620560

T3 - Proc Supercomput 90

SP - 2

EP - 11

BT - Proc Supercomput 90

PB - Publ by IEEE

Y2 - 12 November 1990 through 16 November 1990

ER -

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

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this