Limited Combining Strategies for Large-Scale Shared-Memory Multiprocessors

Kyung Ho Lee, Byung Chang Kang

Research output: Contribution to journalArticle

Abstract

In a large-scale shared-memory multiprocessor, there is a possibility of serious contention due to many requests issued concurrently for the same memory location. A multistage combining network, in which each switch is enhanced with combining so that multiple requests directed to the same memory location can form a single request, significantly reduces the amount of the contention. However, employing combining in every switch of a multistage interconnection network tends to increase the cost and to slow down the network. In this paper, assuming a single-job environment, we investigate some simple strategies that allow only a limited portion of a network to have a combining capability. We show that for situations with a limited number of hot spot locations, these simple strategies can provide performance comparable to a complete combining network in which every switch is enhanced with combining

Original languageEnglish
Pages (from-to)109-119
Number of pages11
JournalJournal of Parallel and Distributed Computing
Volume52
Issue number2
DOIs
Publication statusPublished - 1998 Aug 1
Externally publishedYes

Fingerprint

Shared-memory multiprocessors
Switches
Data storage equipment
Switch
Contention
Multistage Interconnection Networks
Hot Spot
Tend
Strategy
Costs

Keywords

  • Combining network, hot spots, multiprocessors, multistage interconnection networks, parallel processing, performance analysis

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Limited Combining Strategies for Large-Scale Shared-Memory Multiprocessors. / Lee, Kyung Ho; Kang, Byung Chang.

In: Journal of Parallel and Distributed Computing, Vol. 52, No. 2, 01.08.1998, p. 109-119.

Research output: Contribution to journalArticle

@article{892161f8b33447a19475e0ba9635ecd9,
title = "Limited Combining Strategies for Large-Scale Shared-Memory Multiprocessors",
abstract = "In a large-scale shared-memory multiprocessor, there is a possibility of serious contention due to many requests issued concurrently for the same memory location. A multistage combining network, in which each switch is enhanced with combining so that multiple requests directed to the same memory location can form a single request, significantly reduces the amount of the contention. However, employing combining in every switch of a multistage interconnection network tends to increase the cost and to slow down the network. In this paper, assuming a single-job environment, we investigate some simple strategies that allow only a limited portion of a network to have a combining capability. We show that for situations with a limited number of hot spot locations, these simple strategies can provide performance comparable to a complete combining network in which every switch is enhanced with combining",
keywords = "Combining network, hot spots, multiprocessors, multistage interconnection networks, parallel processing, performance analysis",
author = "Lee, {Kyung Ho} and Kang, {Byung Chang}",
year = "1998",
month = "8",
day = "1",
doi = "10.1006/jpdc.1998.1460",
language = "English",
volume = "52",
pages = "109--119",
journal = "Journal of Parallel and Distributed Computing",
issn = "0743-7315",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Limited Combining Strategies for Large-Scale Shared-Memory Multiprocessors

AU - Lee, Kyung Ho

AU - Kang, Byung Chang

PY - 1998/8/1

Y1 - 1998/8/1

N2 - In a large-scale shared-memory multiprocessor, there is a possibility of serious contention due to many requests issued concurrently for the same memory location. A multistage combining network, in which each switch is enhanced with combining so that multiple requests directed to the same memory location can form a single request, significantly reduces the amount of the contention. However, employing combining in every switch of a multistage interconnection network tends to increase the cost and to slow down the network. In this paper, assuming a single-job environment, we investigate some simple strategies that allow only a limited portion of a network to have a combining capability. We show that for situations with a limited number of hot spot locations, these simple strategies can provide performance comparable to a complete combining network in which every switch is enhanced with combining

AB - In a large-scale shared-memory multiprocessor, there is a possibility of serious contention due to many requests issued concurrently for the same memory location. A multistage combining network, in which each switch is enhanced with combining so that multiple requests directed to the same memory location can form a single request, significantly reduces the amount of the contention. However, employing combining in every switch of a multistage interconnection network tends to increase the cost and to slow down the network. In this paper, assuming a single-job environment, we investigate some simple strategies that allow only a limited portion of a network to have a combining capability. We show that for situations with a limited number of hot spot locations, these simple strategies can provide performance comparable to a complete combining network in which every switch is enhanced with combining

KW - Combining network, hot spots, multiprocessors, multistage interconnection networks, parallel processing, performance analysis

UR - http://www.scopus.com/inward/record.url?scp=0038856908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038856908&partnerID=8YFLogxK

U2 - 10.1006/jpdc.1998.1460

DO - 10.1006/jpdc.1998.1460

M3 - Article

AN - SCOPUS:0038856908

VL - 52

SP - 109

EP - 119

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

SN - 0743-7315

IS - 2

ER -