Torus data flow for parallel computation of missized matrix problems

R. E. Funderlic; George A. Geist

doi:10.1016/0024-3795(86)90166-7

Torus data flow for parallel computation of missized matrix problems

R. E. Funderlic, George A. Geist

National Center for Computational Scienc

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

A data-flow approach is used to solve dense symmetric systems of equations on a torus-connected 2-D mesh of processors. A torus mapping of the matrix onto this processor array allows the Cholesky decomposition to be completed in 3n - 2 time steps using only n²/4 processors (less than half the number needed in previously reported results). New definitions for missized problems and parallel algorithm performance are given along with various time-step, efficiency, and processor utilization plots.

Original language	English
Pages (from-to)	149-163
Number of pages	15
Journal	Linear Algebra and Its Applications
Volume	77
Issue number	C
DOIs	https://doi.org/10.1016/0024-3795(86)90166-7
State	Published - May 1986

Access to Document

10.1016/0024-3795(86)90166-7

Cite this

@article{72265392cf2f4de2ae56c57368172865,

title = "Torus data flow for parallel computation of missized matrix problems",

abstract = "A data-flow approach is used to solve dense symmetric systems of equations on a torus-connected 2-D mesh of processors. A torus mapping of the matrix onto this processor array allows the Cholesky decomposition to be completed in 3n - 2 time steps using only n2/4 processors (less than half the number needed in previously reported results). New definitions for missized problems and parallel algorithm performance are given along with various time-step, efficiency, and processor utilization plots.",

author = "Funderlic, {R. E.} and Geist, {George A.}",

year = "1986",

month = may,

doi = "10.1016/0024-3795(86)90166-7",

language = "English",

volume = "77",

pages = "149--163",

journal = "Linear Algebra and Its Applications",

issn = "0024-3795",

publisher = "Elsevier",

number = "C",

}

TY - JOUR

T1 - Torus data flow for parallel computation of missized matrix problems

AU - Funderlic, R. E.

AU - Geist, George A.

PY - 1986/5

Y1 - 1986/5

N2 - A data-flow approach is used to solve dense symmetric systems of equations on a torus-connected 2-D mesh of processors. A torus mapping of the matrix onto this processor array allows the Cholesky decomposition to be completed in 3n - 2 time steps using only n2/4 processors (less than half the number needed in previously reported results). New definitions for missized problems and parallel algorithm performance are given along with various time-step, efficiency, and processor utilization plots.

AB - A data-flow approach is used to solve dense symmetric systems of equations on a torus-connected 2-D mesh of processors. A torus mapping of the matrix onto this processor array allows the Cholesky decomposition to be completed in 3n - 2 time steps using only n2/4 processors (less than half the number needed in previously reported results). New definitions for missized problems and parallel algorithm performance are given along with various time-step, efficiency, and processor utilization plots.

UR - http://www.scopus.com/inward/record.url?scp=46149131021&partnerID=8YFLogxK

U2 - 10.1016/0024-3795(86)90166-7

DO - 10.1016/0024-3795(86)90166-7

M3 - Article

AN - SCOPUS:46149131021

SN - 0024-3795

VL - 77

SP - 149

EP - 163

JO - Linear Algebra and Its Applications

JF - Linear Algebra and Its Applications

IS - C

ER -

Torus data flow for parallel computation of missized matrix problems

Abstract

Access to Document

Other files and links

Fingerprint

Cite this