Prediction of communication delay in torus networks under multiple time-scale correlated traffic

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The efficiency of a large-scale multicomputer is critically dependent on the performance of its interconnection network. Current multicomputers have widely employed the torus as their underlying network topology for efficient interprocessor communication. In order to ensure a successful exploitation of the computational power offered by multicomputers it is essential to obtain a clear understanding of the performance capabilities of their interconnection networks under various system configurations. Analytical modelling plays an important role in achieving this goal. This study proposes a concise performance model for computing communication delay in the torus network with circuit switching in the presence of multiple time-scale correlated traffic which is found in many real-world parallel computation environments and has strong impact on network performance. The tractability and reasonable accuracy of the analytical model demonstrated by extensive simulation experiments make it a practical and cost-effective evaluation tool to investigate network performance with various alternative design solutions and under different operating conditions.

Original languageEnglish
Pages (from-to)255-273
Number of pages19
JournalPerformance Evaluation
Volume60
Issue number1-4
DOIs
Publication statusPublished - May 2005

Fingerprint

Multiple Time Scales
Multicomputers
Communication Delay
Network performance
Torus
Interconnection Networks
Traffic
Network Performance
Switching circuits
Prediction
Communication
Circuit Switching
Interprocessor Communication
Analytical models
Analytical Modeling
Tractability
Topology
Parallel Computation
Performance Model
Network Topology

Keywords

  • Burstiness
  • Circuit switching
  • Correlated traffic
  • Multicomputers
  • Multiple time-scales
  • Performance modelling

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software
  • Modelling and Simulation
  • Statistics and Probability

Cite this

Prediction of communication delay in torus networks under multiple time-scale correlated traffic. / Min, Geyong; Ould-Khaoua, Mohamed.

In: Performance Evaluation, Vol. 60, No. 1-4, 05.2005, p. 255-273.

Research output: Contribution to journalArticle

@article{f46cd2be5b9a42268e2696b70b9485c2,
title = "Prediction of communication delay in torus networks under multiple time-scale correlated traffic",
abstract = "The efficiency of a large-scale multicomputer is critically dependent on the performance of its interconnection network. Current multicomputers have widely employed the torus as their underlying network topology for efficient interprocessor communication. In order to ensure a successful exploitation of the computational power offered by multicomputers it is essential to obtain a clear understanding of the performance capabilities of their interconnection networks under various system configurations. Analytical modelling plays an important role in achieving this goal. This study proposes a concise performance model for computing communication delay in the torus network with circuit switching in the presence of multiple time-scale correlated traffic which is found in many real-world parallel computation environments and has strong impact on network performance. The tractability and reasonable accuracy of the analytical model demonstrated by extensive simulation experiments make it a practical and cost-effective evaluation tool to investigate network performance with various alternative design solutions and under different operating conditions.",
keywords = "Burstiness, Circuit switching, Correlated traffic, Multicomputers, Multiple time-scales, Performance modelling",
author = "Geyong Min and Mohamed Ould-Khaoua",
year = "2005",
month = "5",
doi = "10.1016/j.peva.2004.10.010",
language = "English",
volume = "60",
pages = "255--273",
journal = "Performance Evaluation",
issn = "0166-5316",
publisher = "Elsevier",
number = "1-4",

}

TY - JOUR

T1 - Prediction of communication delay in torus networks under multiple time-scale correlated traffic

AU - Min, Geyong

AU - Ould-Khaoua, Mohamed

PY - 2005/5

Y1 - 2005/5

N2 - The efficiency of a large-scale multicomputer is critically dependent on the performance of its interconnection network. Current multicomputers have widely employed the torus as their underlying network topology for efficient interprocessor communication. In order to ensure a successful exploitation of the computational power offered by multicomputers it is essential to obtain a clear understanding of the performance capabilities of their interconnection networks under various system configurations. Analytical modelling plays an important role in achieving this goal. This study proposes a concise performance model for computing communication delay in the torus network with circuit switching in the presence of multiple time-scale correlated traffic which is found in many real-world parallel computation environments and has strong impact on network performance. The tractability and reasonable accuracy of the analytical model demonstrated by extensive simulation experiments make it a practical and cost-effective evaluation tool to investigate network performance with various alternative design solutions and under different operating conditions.

AB - The efficiency of a large-scale multicomputer is critically dependent on the performance of its interconnection network. Current multicomputers have widely employed the torus as their underlying network topology for efficient interprocessor communication. In order to ensure a successful exploitation of the computational power offered by multicomputers it is essential to obtain a clear understanding of the performance capabilities of their interconnection networks under various system configurations. Analytical modelling plays an important role in achieving this goal. This study proposes a concise performance model for computing communication delay in the torus network with circuit switching in the presence of multiple time-scale correlated traffic which is found in many real-world parallel computation environments and has strong impact on network performance. The tractability and reasonable accuracy of the analytical model demonstrated by extensive simulation experiments make it a practical and cost-effective evaluation tool to investigate network performance with various alternative design solutions and under different operating conditions.

KW - Burstiness

KW - Circuit switching

KW - Correlated traffic

KW - Multicomputers

KW - Multiple time-scales

KW - Performance modelling

UR - http://www.scopus.com/inward/record.url?scp=14544300535&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14544300535&partnerID=8YFLogxK

U2 - 10.1016/j.peva.2004.10.010

DO - 10.1016/j.peva.2004.10.010

M3 - Article

VL - 60

SP - 255

EP - 273

JO - Performance Evaluation

JF - Performance Evaluation

SN - 0166-5316

IS - 1-4

ER -