Memory resource considerations in the load balancing of software DSM systems

Yen Tso Liu, Tyng Yeu Liang, Chi Ting Huang, Ce-Kuen Shieh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Obtaining a load balance is essential if the program performance of distributed shared memory (DSM) systems is to be optimized. In achieving this load balance, most DSM systems simply distribute program threads in accordance with the CPU power of the individual processors within the network. However, memory access costs also play a significant role in determining the program performance. Although the threads will be able to complete their tasks, their execution will inevitably be delayed by the latency associated with executing the page replacements. With the rapid development of CPU chips, the relative influence of this memory access latency upon the overall program performance has become increasingly significant. Therefore, attempts to minimize the execution time of applications by establishing a load balance based purely upon CPU resource considerations will only achieve limited success. The current study proposes a new load balancing scheme for DSM systems which considers both CPU and memory resources. The present results confirm the importance of considering memory resources when addressing the load balancing of DSM systems. It is shown that the proposed method is more effective than previous schemes which considered only CPU resources or memory resources.

Original languageEnglish
Title of host publicationProceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003
EditorsChua-Huang Huang, J. Ramanujam
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages71-78
Number of pages8
ISBN (Electronic)0769520189
DOIs
Publication statusPublished - 2003 Jan 1
Event2003 International Conference on Parallel Processing Workshops, ICPPW 2003 - Kaohsiung, Taiwan
Duration: 2003 Oct 62003 Oct 9

Publication series

NameProceedings of the International Conference on Parallel Processing Workshops
Volume2003-January
ISSN (Print)1530-2016

Other

Other2003 International Conference on Parallel Processing Workshops, ICPPW 2003
CountryTaiwan
CityKaohsiung
Period03-10-0603-10-09

Fingerprint

Distributed Shared Memory
Load Balancing
Resource allocation
Computer systems
Load Balance
Data storage equipment
Resources
Program processors
Software
Thread
Latency
Execution Time
Replacement
Chip
Minimise
Computer networks
Costs

All Science Journal Classification (ASJC) codes

  • Software
  • Mathematics(all)
  • Hardware and Architecture

Cite this

Liu, Y. T., Liang, T. Y., Huang, C. T., & Shieh, C-K. (2003). Memory resource considerations in the load balancing of software DSM systems. In C-H. Huang, & J. Ramanujam (Eds.), Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003 (pp. 71-78). [1240356] (Proceedings of the International Conference on Parallel Processing Workshops; Vol. 2003-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPPW.2003.1240356
Liu, Yen Tso ; Liang, Tyng Yeu ; Huang, Chi Ting ; Shieh, Ce-Kuen. / Memory resource considerations in the load balancing of software DSM systems. Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003. editor / Chua-Huang Huang ; J. Ramanujam. Institute of Electrical and Electronics Engineers Inc., 2003. pp. 71-78 (Proceedings of the International Conference on Parallel Processing Workshops).
@inproceedings{e2195234025b4274a690609df223f76b,
title = "Memory resource considerations in the load balancing of software DSM systems",
abstract = "Obtaining a load balance is essential if the program performance of distributed shared memory (DSM) systems is to be optimized. In achieving this load balance, most DSM systems simply distribute program threads in accordance with the CPU power of the individual processors within the network. However, memory access costs also play a significant role in determining the program performance. Although the threads will be able to complete their tasks, their execution will inevitably be delayed by the latency associated with executing the page replacements. With the rapid development of CPU chips, the relative influence of this memory access latency upon the overall program performance has become increasingly significant. Therefore, attempts to minimize the execution time of applications by establishing a load balance based purely upon CPU resource considerations will only achieve limited success. The current study proposes a new load balancing scheme for DSM systems which considers both CPU and memory resources. The present results confirm the importance of considering memory resources when addressing the load balancing of DSM systems. It is shown that the proposed method is more effective than previous schemes which considered only CPU resources or memory resources.",
author = "Liu, {Yen Tso} and Liang, {Tyng Yeu} and Huang, {Chi Ting} and Ce-Kuen Shieh",
year = "2003",
month = "1",
day = "1",
doi = "10.1109/ICPPW.2003.1240356",
language = "English",
series = "Proceedings of the International Conference on Parallel Processing Workshops",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "71--78",
editor = "Chua-Huang Huang and J. Ramanujam",
booktitle = "Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003",
address = "United States",

}

Liu, YT, Liang, TY, Huang, CT & Shieh, C-K 2003, Memory resource considerations in the load balancing of software DSM systems. in C-H Huang & J Ramanujam (eds), Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003., 1240356, Proceedings of the International Conference on Parallel Processing Workshops, vol. 2003-January, Institute of Electrical and Electronics Engineers Inc., pp. 71-78, 2003 International Conference on Parallel Processing Workshops, ICPPW 2003, Kaohsiung, Taiwan, 03-10-06. https://doi.org/10.1109/ICPPW.2003.1240356

Memory resource considerations in the load balancing of software DSM systems. / Liu, Yen Tso; Liang, Tyng Yeu; Huang, Chi Ting; Shieh, Ce-Kuen.

Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003. ed. / Chua-Huang Huang; J. Ramanujam. Institute of Electrical and Electronics Engineers Inc., 2003. p. 71-78 1240356 (Proceedings of the International Conference on Parallel Processing Workshops; Vol. 2003-January).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Memory resource considerations in the load balancing of software DSM systems

AU - Liu, Yen Tso

AU - Liang, Tyng Yeu

AU - Huang, Chi Ting

AU - Shieh, Ce-Kuen

PY - 2003/1/1

Y1 - 2003/1/1

N2 - Obtaining a load balance is essential if the program performance of distributed shared memory (DSM) systems is to be optimized. In achieving this load balance, most DSM systems simply distribute program threads in accordance with the CPU power of the individual processors within the network. However, memory access costs also play a significant role in determining the program performance. Although the threads will be able to complete their tasks, their execution will inevitably be delayed by the latency associated with executing the page replacements. With the rapid development of CPU chips, the relative influence of this memory access latency upon the overall program performance has become increasingly significant. Therefore, attempts to minimize the execution time of applications by establishing a load balance based purely upon CPU resource considerations will only achieve limited success. The current study proposes a new load balancing scheme for DSM systems which considers both CPU and memory resources. The present results confirm the importance of considering memory resources when addressing the load balancing of DSM systems. It is shown that the proposed method is more effective than previous schemes which considered only CPU resources or memory resources.

AB - Obtaining a load balance is essential if the program performance of distributed shared memory (DSM) systems is to be optimized. In achieving this load balance, most DSM systems simply distribute program threads in accordance with the CPU power of the individual processors within the network. However, memory access costs also play a significant role in determining the program performance. Although the threads will be able to complete their tasks, their execution will inevitably be delayed by the latency associated with executing the page replacements. With the rapid development of CPU chips, the relative influence of this memory access latency upon the overall program performance has become increasingly significant. Therefore, attempts to minimize the execution time of applications by establishing a load balance based purely upon CPU resource considerations will only achieve limited success. The current study proposes a new load balancing scheme for DSM systems which considers both CPU and memory resources. The present results confirm the importance of considering memory resources when addressing the load balancing of DSM systems. It is shown that the proposed method is more effective than previous schemes which considered only CPU resources or memory resources.

UR - http://www.scopus.com/inward/record.url?scp=55749095286&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=55749095286&partnerID=8YFLogxK

U2 - 10.1109/ICPPW.2003.1240356

DO - 10.1109/ICPPW.2003.1240356

M3 - Conference contribution

T3 - Proceedings of the International Conference on Parallel Processing Workshops

SP - 71

EP - 78

BT - Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003

A2 - Huang, Chua-Huang

A2 - Ramanujam, J.

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Liu YT, Liang TY, Huang CT, Shieh C-K. Memory resource considerations in the load balancing of software DSM systems. In Huang C-H, Ramanujam J, editors, Proceedings - 2003 International Conference on Parallel Processing Workshops, ICPPW 2003. Institute of Electrical and Electronics Engineers Inc. 2003. p. 71-78. 1240356. (Proceedings of the International Conference on Parallel Processing Workshops). https://doi.org/10.1109/ICPPW.2003.1240356