Towards a more realistic comparative analysis of multicomputer networks

H. Sarbazi-Azad, M. Ould-Khaoua, L. M. Mackenzie

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Several studies have examined the relative performance merits of the torus and hypercube taking into account the channel bandwidth constraints imposed by implementation technology. While the torus has been shown to outperform the hypercube under the constant wiring density constraint, the opposite conclusion has been reached when the constant pin-out constraint is considered. However, these studies have assumed a pure uniform traffic pattern and deterministic routing. The 'uniform traffic' assumption is not always justifiable in practice as there are many real-world parallel applications that exhibit non-uniform traffic patterns, which can create unbalanced traffic such as hotspots in the network. This paper re-examines the performance merits of the torus and hypercube in the presence of hotspot traffic. The comparative analysis is based on fully adaptive routing as this has been gaining popularity in recent practical multicomputer. Moreover, it uses a new cost model that takes into account the implementation cost of the network and its routers. The results reveal that for moderate and large system sizes, lower dimensional k-ary n-cubes (e.g. 2D torus) always outperform their higher dimensional counterparts even under the pin-out constraint.

Original languageEnglish
Pages (from-to)1271-1289
Number of pages19
JournalConcurrency Computation Practice and Experience
Volume16
Issue number13
DOIs
Publication statusPublished - Nov 2004

Fingerprint

Multicomputers
Comparative Analysis
Traffic
Torus
Hypercube
Electric wiring
Routers
Costs
Hot Spot
Bandwidth
K-ary N-cubes
Adaptive Routing
Cost Model
Parallel Applications
Real-world Applications
Router
Routing
High-dimensional

Keywords

  • Fully-adaptive routing
  • Hotspot
  • Hypercube
  • Interconnection networks
  • K-ary n-cube
  • Performance comparison
  • Torus
  • Wormhole routing

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

Towards a more realistic comparative analysis of multicomputer networks. / Sarbazi-Azad, H.; Ould-Khaoua, M.; Mackenzie, L. M.

In: Concurrency Computation Practice and Experience, Vol. 16, No. 13, 11.2004, p. 1271-1289.

Research output: Contribution to journalArticle

@article{c1993189ca814dfebc805b206573f207,
title = "Towards a more realistic comparative analysis of multicomputer networks",
abstract = "Several studies have examined the relative performance merits of the torus and hypercube taking into account the channel bandwidth constraints imposed by implementation technology. While the torus has been shown to outperform the hypercube under the constant wiring density constraint, the opposite conclusion has been reached when the constant pin-out constraint is considered. However, these studies have assumed a pure uniform traffic pattern and deterministic routing. The 'uniform traffic' assumption is not always justifiable in practice as there are many real-world parallel applications that exhibit non-uniform traffic patterns, which can create unbalanced traffic such as hotspots in the network. This paper re-examines the performance merits of the torus and hypercube in the presence of hotspot traffic. The comparative analysis is based on fully adaptive routing as this has been gaining popularity in recent practical multicomputer. Moreover, it uses a new cost model that takes into account the implementation cost of the network and its routers. The results reveal that for moderate and large system sizes, lower dimensional k-ary n-cubes (e.g. 2D torus) always outperform their higher dimensional counterparts even under the pin-out constraint.",
keywords = "Fully-adaptive routing, Hotspot, Hypercube, Interconnection networks, K-ary n-cube, Performance comparison, Torus, Wormhole routing",
author = "H. Sarbazi-Azad and M. Ould-Khaoua and Mackenzie, {L. M.}",
year = "2004",
month = "11",
doi = "10.1002/cpe.755",
language = "English",
volume = "16",
pages = "1271--1289",
journal = "Concurrency Computation Practice and Experience",
issn = "1532-0626",
publisher = "John Wiley and Sons Ltd",
number = "13",

}

TY - JOUR

T1 - Towards a more realistic comparative analysis of multicomputer networks

AU - Sarbazi-Azad, H.

AU - Ould-Khaoua, M.

AU - Mackenzie, L. M.

PY - 2004/11

Y1 - 2004/11

N2 - Several studies have examined the relative performance merits of the torus and hypercube taking into account the channel bandwidth constraints imposed by implementation technology. While the torus has been shown to outperform the hypercube under the constant wiring density constraint, the opposite conclusion has been reached when the constant pin-out constraint is considered. However, these studies have assumed a pure uniform traffic pattern and deterministic routing. The 'uniform traffic' assumption is not always justifiable in practice as there are many real-world parallel applications that exhibit non-uniform traffic patterns, which can create unbalanced traffic such as hotspots in the network. This paper re-examines the performance merits of the torus and hypercube in the presence of hotspot traffic. The comparative analysis is based on fully adaptive routing as this has been gaining popularity in recent practical multicomputer. Moreover, it uses a new cost model that takes into account the implementation cost of the network and its routers. The results reveal that for moderate and large system sizes, lower dimensional k-ary n-cubes (e.g. 2D torus) always outperform their higher dimensional counterparts even under the pin-out constraint.

AB - Several studies have examined the relative performance merits of the torus and hypercube taking into account the channel bandwidth constraints imposed by implementation technology. While the torus has been shown to outperform the hypercube under the constant wiring density constraint, the opposite conclusion has been reached when the constant pin-out constraint is considered. However, these studies have assumed a pure uniform traffic pattern and deterministic routing. The 'uniform traffic' assumption is not always justifiable in practice as there are many real-world parallel applications that exhibit non-uniform traffic patterns, which can create unbalanced traffic such as hotspots in the network. This paper re-examines the performance merits of the torus and hypercube in the presence of hotspot traffic. The comparative analysis is based on fully adaptive routing as this has been gaining popularity in recent practical multicomputer. Moreover, it uses a new cost model that takes into account the implementation cost of the network and its routers. The results reveal that for moderate and large system sizes, lower dimensional k-ary n-cubes (e.g. 2D torus) always outperform their higher dimensional counterparts even under the pin-out constraint.

KW - Fully-adaptive routing

KW - Hotspot

KW - Hypercube

KW - Interconnection networks

KW - K-ary n-cube

KW - Performance comparison

KW - Torus

KW - Wormhole routing

UR - http://www.scopus.com/inward/record.url?scp=8344266804&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=8344266804&partnerID=8YFLogxK

U2 - 10.1002/cpe.755

DO - 10.1002/cpe.755

M3 - Article

VL - 16

SP - 1271

EP - 1289

JO - Concurrency Computation Practice and Experience

JF - Concurrency Computation Practice and Experience

SN - 1532-0626

IS - 13

ER -