Incorporating memory resource considerations into the workload distribution of software DSM systems

Yen Tso Liu, Tyng Yeu Liang, Jyh Biau Chang, Ce-Kuen Shieh

Research output: Contribution to journalArticle

Abstract

Conventional workload distribution schemes for software distributed shared memory (DSM) systems simply distribute the program threads in accordance with the CPU power of the individual processors or the data-sharing characteristics of the application. Although these schemes aim to minimize the program execution time by reducing the computation and communication costs, memory access costs also have a major influence on the overall program performance. If a processor has insufficient physical memory space to cache all of the data required by its local working threads, it must perform a series of page replacements if it is to complete its thread executions. Although these page replacements enable the threads to complete their tasks, thread execution is inevitably delayed by the latency of the page swapping operations. Consequently, the current study proposes a novel workload distribution scheme for DSM systems which considers not only the CPU power and data-sharing characteristics, but also the physical memory capabilities of the individual processors. The present results confirm the importance of considering memory resources when establishing an appropriate workload distribution for DSM systems and indicate that the proposed scheme is more effective than schemes which consider only CPU resources or memory resources, respectively.

Original languageEnglish
Pages (from-to)243-257
Number of pages15
JournalJournal of Information Science and Engineering
Volume23
Issue number1
Publication statusPublished - 2007 Jan 1

Fingerprint

workload
Computer systems
Data storage equipment
resources
Program processors
software
costs
Costs
communication
Communication
performance

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Hardware and Architecture
  • Library and Information Sciences
  • Computational Theory and Mathematics

Cite this

@article{2e149efcc4054325b8161f4fa6497888,
title = "Incorporating memory resource considerations into the workload distribution of software DSM systems",
abstract = "Conventional workload distribution schemes for software distributed shared memory (DSM) systems simply distribute the program threads in accordance with the CPU power of the individual processors or the data-sharing characteristics of the application. Although these schemes aim to minimize the program execution time by reducing the computation and communication costs, memory access costs also have a major influence on the overall program performance. If a processor has insufficient physical memory space to cache all of the data required by its local working threads, it must perform a series of page replacements if it is to complete its thread executions. Although these page replacements enable the threads to complete their tasks, thread execution is inevitably delayed by the latency of the page swapping operations. Consequently, the current study proposes a novel workload distribution scheme for DSM systems which considers not only the CPU power and data-sharing characteristics, but also the physical memory capabilities of the individual processors. The present results confirm the importance of considering memory resources when establishing an appropriate workload distribution for DSM systems and indicate that the proposed scheme is more effective than schemes which consider only CPU resources or memory resources, respectively.",
author = "Liu, {Yen Tso} and Liang, {Tyng Yeu} and Chang, {Jyh Biau} and Ce-Kuen Shieh",
year = "2007",
month = "1",
day = "1",
language = "English",
volume = "23",
pages = "243--257",
journal = "Journal of Information Science and Engineering",
issn = "1016-2364",
publisher = "Institute of Information Science",
number = "1",

}

Incorporating memory resource considerations into the workload distribution of software DSM systems. / Liu, Yen Tso; Liang, Tyng Yeu; Chang, Jyh Biau; Shieh, Ce-Kuen.

In: Journal of Information Science and Engineering, Vol. 23, No. 1, 01.01.2007, p. 243-257.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Incorporating memory resource considerations into the workload distribution of software DSM systems

AU - Liu, Yen Tso

AU - Liang, Tyng Yeu

AU - Chang, Jyh Biau

AU - Shieh, Ce-Kuen

PY - 2007/1/1

Y1 - 2007/1/1

N2 - Conventional workload distribution schemes for software distributed shared memory (DSM) systems simply distribute the program threads in accordance with the CPU power of the individual processors or the data-sharing characteristics of the application. Although these schemes aim to minimize the program execution time by reducing the computation and communication costs, memory access costs also have a major influence on the overall program performance. If a processor has insufficient physical memory space to cache all of the data required by its local working threads, it must perform a series of page replacements if it is to complete its thread executions. Although these page replacements enable the threads to complete their tasks, thread execution is inevitably delayed by the latency of the page swapping operations. Consequently, the current study proposes a novel workload distribution scheme for DSM systems which considers not only the CPU power and data-sharing characteristics, but also the physical memory capabilities of the individual processors. The present results confirm the importance of considering memory resources when establishing an appropriate workload distribution for DSM systems and indicate that the proposed scheme is more effective than schemes which consider only CPU resources or memory resources, respectively.

AB - Conventional workload distribution schemes for software distributed shared memory (DSM) systems simply distribute the program threads in accordance with the CPU power of the individual processors or the data-sharing characteristics of the application. Although these schemes aim to minimize the program execution time by reducing the computation and communication costs, memory access costs also have a major influence on the overall program performance. If a processor has insufficient physical memory space to cache all of the data required by its local working threads, it must perform a series of page replacements if it is to complete its thread executions. Although these page replacements enable the threads to complete their tasks, thread execution is inevitably delayed by the latency of the page swapping operations. Consequently, the current study proposes a novel workload distribution scheme for DSM systems which considers not only the CPU power and data-sharing characteristics, but also the physical memory capabilities of the individual processors. The present results confirm the importance of considering memory resources when establishing an appropriate workload distribution for DSM systems and indicate that the proposed scheme is more effective than schemes which consider only CPU resources or memory resources, respectively.

UR - http://www.scopus.com/inward/record.url?scp=33846707590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846707590&partnerID=8YFLogxK

M3 - Article

VL - 23

SP - 243

EP - 257

JO - Journal of Information Science and Engineering

JF - Journal of Information Science and Engineering

SN - 1016-2364

IS - 1

ER -