Performance prediction of wormhole switching in hypercubes with bursty traffic pattern

Geyong Min; Mohamed Ould-Khaoua; John Ferguson

Performance prediction of wormhole switching in hypercubes with bursty traffic pattern

Geyong Min^*, Mohamed Ould-Khaoua, John Ferguson

^*Corresponding author for this work

Electrical and Computer Engineering

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

1 Citation (Scopus)

Abstract

Wormhole switching has been extensively deployed in the current-generation of high-end parallel systems. The performance of wormhole switching in hypercubes has been primarily investigated under the assumption that traffic load follows a Poisson arrival process. However, many studies have shown that traffic in parallel computation environments can exhibit a high degree of burstiness and the Poisson arrival process is unable to model the behaviour of bursty traffic. This paper proposes a new analytical model for wormhole-switched hypercubes under bursty traffic. Simulation experiments demonstrate that the proposed model exhibits a good degree of accuracy under various operating conditions.

Original language	English
Title of host publication	Proceedings of the ACM Symposium on Applied Computing
Editors	G. Lamont
Pages	985-989
Number of pages	5
Publication status	Published - 2003
Event	Proceedings of the 2003 ACM Symposium on Applied Computing - Melbourne, FL, United States Duration: Mar 9 2003 → Mar 12 2003

Other

Other	Proceedings of the 2003 ACM Symposium on Applied Computing
Country/Territory	United States
City	Melbourne, FL
Period	3/9/03 → 3/12/03

Keywords

Bursty Traffic
Interconnection Networks
Parallel Systems
Performance Modelling

ASJC Scopus subject areas

Computer Science(all)

Cite this

@inproceedings{02510aa9aa5e4522abd67a4042d4a300,

title = "Performance prediction of wormhole switching in hypercubes with bursty traffic pattern",

abstract = "Wormhole switching has been extensively deployed in the current-generation of high-end parallel systems. The performance of wormhole switching in hypercubes has been primarily investigated under the assumption that traffic load follows a Poisson arrival process. However, many studies have shown that traffic in parallel computation environments can exhibit a high degree of burstiness and the Poisson arrival process is unable to model the behaviour of bursty traffic. This paper proposes a new analytical model for wormhole-switched hypercubes under bursty traffic. Simulation experiments demonstrate that the proposed model exhibits a good degree of accuracy under various operating conditions.",

keywords = "Bursty Traffic, Interconnection Networks, Parallel Systems, Performance Modelling",

author = "Geyong Min and Mohamed Ould-Khaoua and John Ferguson",

year = "2003",

language = "English",

pages = "985--989",

editor = "G. Lamont",

booktitle = "Proceedings of the ACM Symposium on Applied Computing",

note = "Proceedings of the 2003 ACM Symposium on Applied Computing ; Conference date: 09-03-2003 Through 12-03-2003",

}

TY - GEN

T1 - Performance prediction of wormhole switching in hypercubes with bursty traffic pattern

AU - Min, Geyong

AU - Ould-Khaoua, Mohamed

AU - Ferguson, John

PY - 2003

Y1 - 2003

N2 - Wormhole switching has been extensively deployed in the current-generation of high-end parallel systems. The performance of wormhole switching in hypercubes has been primarily investigated under the assumption that traffic load follows a Poisson arrival process. However, many studies have shown that traffic in parallel computation environments can exhibit a high degree of burstiness and the Poisson arrival process is unable to model the behaviour of bursty traffic. This paper proposes a new analytical model for wormhole-switched hypercubes under bursty traffic. Simulation experiments demonstrate that the proposed model exhibits a good degree of accuracy under various operating conditions.

AB - Wormhole switching has been extensively deployed in the current-generation of high-end parallel systems. The performance of wormhole switching in hypercubes has been primarily investigated under the assumption that traffic load follows a Poisson arrival process. However, many studies have shown that traffic in parallel computation environments can exhibit a high degree of burstiness and the Poisson arrival process is unable to model the behaviour of bursty traffic. This paper proposes a new analytical model for wormhole-switched hypercubes under bursty traffic. Simulation experiments demonstrate that the proposed model exhibits a good degree of accuracy under various operating conditions.

KW - Bursty Traffic

KW - Interconnection Networks

KW - Parallel Systems

KW - Performance Modelling

UR - http://www.scopus.com/inward/record.url?scp=0038310385&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038310385&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0038310385

SP - 985

EP - 989

BT - Proceedings of the ACM Symposium on Applied Computing

A2 - Lamont, G.

T2 - Proceedings of the 2003 ACM Symposium on Applied Computing

Y2 - 9 March 2003 through 12 March 2003

ER -

Performance prediction of wormhole switching in hypercubes with bursty traffic pattern

Abstract

Other

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this