Parallel gauss-seidel on a torus NOC architecture

Khaled Day; Mohammad H. Al-Towaiq

doi:10.2316/P.2013.795-025

Parallel gauss-seidel on a torus NOC architecture

Khaled Day, Mohammad H. Al-Towaiq

Computer Science

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

Abstract

We propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a 2D torus network-con-chip (NoC) architecture. The proposed parallel algorithm is of O(Nn2/k2) time complexity for solving a system with matrix of order n on a k×k torus NoC architecture requiring 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 and efficiency.

Original language	English
Title of host publication	IASTED Multiconferences - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013
Pages	608-612
Number of pages	5
DOIs	https://doi.org/10.2316/P.2013.795-025
Publication status	Published - 2013
Event	11th IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013 - Innsbruck, Austria Duration: Feb 11 2013 → Feb 13 2013

Other

Other	11th IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013
Country	Austria
City	Innsbruck
Period	2/11/13 → 2/13/13

Fingerprint

Keywords

2D torus
Gauss-seidel
Linear system of equations
Network-on-Chip
Parallel processing

ASJC Scopus subject areas

Computer Networks and Communications
Software

Cite this

Day, K., & Al-Towaiq, M. H. (2013). Parallel gauss-seidel on a torus NOC architecture. In IASTED Multiconferences - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013 (pp. 608-612) https://doi.org/10.2316/P.2013.795-025

Day, K & Al-Towaiq, MH 2013, Parallel gauss-seidel on a torus NOC architecture. in IASTED Multiconferences - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013. pp. 608-612, 11th IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013, Innsbruck, Austria, 2/11/13. https://doi.org/10.2316/P.2013.795-025

@inproceedings{b826d50922cc40ec9e42a6a0cb85070e,

title = "Parallel gauss-seidel on a torus NOC architecture",

abstract = "We propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a 2D torus network-con-chip (NoC) architecture. The proposed parallel algorithm is of O(Nn2/k2) time complexity for solving a system with matrix of order n on a k×k torus NoC architecture requiring 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 and efficiency.",

keywords = "2D torus, Gauss-seidel, Linear system of equations, Network-on-Chip, Parallel processing",

author = "Khaled Day and Al-Towaiq, {Mohammad H.}",

year = "2013",

doi = "10.2316/P.2013.795-025",

language = "English",

isbn = "9780889869431",

pages = "608--612",

booktitle = "IASTED Multiconferences - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013",

note = "11th IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013 ; Conference date: 11-02-2013 Through 13-02-2013",

}

TY - GEN

T1 - Parallel gauss-seidel on a torus NOC architecture

AU - Day, Khaled

AU - Al-Towaiq, Mohammad H.

PY - 2013

Y1 - 2013

N2 - We propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a 2D torus network-con-chip (NoC) architecture. The proposed parallel algorithm is of O(Nn2/k2) time complexity for solving a system with matrix of order n on a k×k torus NoC architecture requiring 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 and efficiency.

AB - We propose a parallel Gauss-Seidel (GS) iterative algorithm for solving large systems of linear equations on a 2D torus network-con-chip (NoC) architecture. The proposed parallel algorithm is of O(Nn2/k2) time complexity for solving a system with matrix of order n on a k×k torus NoC architecture requiring 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 and efficiency.

KW - 2D torus

KW - Gauss-seidel

KW - Linear system of equations

KW - Network-on-Chip

KW - Parallel processing

UR - http://www.scopus.com/inward/record.url?scp=84875488254&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875488254&partnerID=8YFLogxK

U2 - 10.2316/P.2013.795-025

DO - 10.2316/P.2013.795-025

M3 - Conference contribution

AN - SCOPUS:84875488254

SN - 9780889869431

SP - 608

EP - 612

BT - IASTED Multiconferences - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013

T2 - 11th IASTED International Conference on Parallel and Distributed Computing and Networks, PDCN 2013

Y2 - 11 February 2013 through 13 February 2013

ER -

Access to Document

10.2316/P.2013.795-025