Parallel gauss-seidel on a torus network-on-chip architecture

Mohammad H. Al-Towaiq, Khaled Day

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Network-on-chip multicore architectures with a large number of processing elements are becoming a reality with the recent developments in technology. In these modern systems the processing elements are interconnected with regular network-on-chip (NoC) topologies such as meshes and trees. In this paper we propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a torus NoC architecture. The proposed parallel algorithm is O(Nn2/k2) time complexity for solving a system with matrix of order n on a k × k torus NoC architecture with N iterations assuming n and N are large compared to k (i.e. for large linear systems that require a large number of iterations). We show that under these conditions the proposed parallel GS algorithm has near optimal speedup.

Original languageEnglish
Article number1250001
JournalJournal of Interconnection Networks
Volume13
Issue number1-2
DOIs
Publication statusPublished - Mar 2012

Fingerprint

Processing
Linear equations
Parallel algorithms
Technology
Linear systems
Topology
Network-on-chip

Keywords

  • Gauss-Seidel
  • iterative method
  • linear system of equations
  • Network-on-chip
  • torus, parallel algorithm

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Parallel gauss-seidel on a torus network-on-chip architecture. / Al-Towaiq, Mohammad H.; Day, Khaled.

In: Journal of Interconnection Networks, Vol. 13, No. 1-2, 1250001, 03.2012.

Research output: Contribution to journalArticle

@article{0acd4785d3374b85bbc38f65b8584c63,
title = "Parallel gauss-seidel on a torus network-on-chip architecture",
abstract = "Network-on-chip multicore architectures with a large number of processing elements are becoming a reality with the recent developments in technology. In these modern systems the processing elements are interconnected with regular network-on-chip (NoC) topologies such as meshes and trees. In this paper we propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a torus NoC architecture. The proposed parallel algorithm is O(Nn2/k2) time complexity for solving a system with matrix of order n on a k × k torus NoC architecture with N iterations assuming n and N are large compared to k (i.e. for large linear systems that require a large number of iterations). We show that under these conditions the proposed parallel GS algorithm has near optimal speedup.",
keywords = "Gauss-Seidel, iterative method, linear system of equations, Network-on-chip, torus, parallel algorithm",
author = "Al-Towaiq, {Mohammad H.} and Khaled Day",
year = "2012",
month = "3",
doi = "10.1142/S0219265912500016",
language = "English",
volume = "13",
journal = "Advanced healthcare materials",
issn = "2192-2640",
publisher = "John Wiley and Sons Ltd",
number = "1-2",

}

TY - JOUR

T1 - Parallel gauss-seidel on a torus network-on-chip architecture

AU - Al-Towaiq, Mohammad H.

AU - Day, Khaled

PY - 2012/3

Y1 - 2012/3

N2 - Network-on-chip multicore architectures with a large number of processing elements are becoming a reality with the recent developments in technology. In these modern systems the processing elements are interconnected with regular network-on-chip (NoC) topologies such as meshes and trees. In this paper we propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a torus NoC architecture. The proposed parallel algorithm is O(Nn2/k2) time complexity for solving a system with matrix of order n on a k × k torus NoC architecture with N iterations assuming n and N are large compared to k (i.e. for large linear systems that require a large number of iterations). We show that under these conditions the proposed parallel GS algorithm has near optimal speedup.

AB - Network-on-chip multicore architectures with a large number of processing elements are becoming a reality with the recent developments in technology. In these modern systems the processing elements are interconnected with regular network-on-chip (NoC) topologies such as meshes and trees. In this paper we propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a torus NoC architecture. The proposed parallel algorithm is O(Nn2/k2) time complexity for solving a system with matrix of order n on a k × k torus NoC architecture with N iterations assuming n and N are large compared to k (i.e. for large linear systems that require a large number of iterations). We show that under these conditions the proposed parallel GS algorithm has near optimal speedup.

KW - Gauss-Seidel

KW - iterative method

KW - linear system of equations

KW - Network-on-chip

KW - torus, parallel algorithm

UR - http://www.scopus.com/inward/record.url?scp=84874799178&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874799178&partnerID=8YFLogxK

U2 - 10.1142/S0219265912500016

DO - 10.1142/S0219265912500016

M3 - Article

AN - SCOPUS:84874799178

VL - 13

JO - Advanced healthcare materials

JF - Advanced healthcare materials

SN - 2192-2640

IS - 1-2

M1 - 1250001

ER -