Performance study of linearly connected parallel machines through simulation

L. F. Bautista, Dilip Sarkar

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

Abstract

Recent advancements in hardware technology have increased the power of the conventional computer. However, a conventional computer can only perform one task at a given moment. Parallel computers consist of many processors working together simultaneously, which should make them faster and more powerful than the conventional computer. In this work, we study the performance of linearly connected parallel machines by examining several job scheduling algorithms for these computers. Our motivation for this study is the observation that the computation power of personal computers is poorly utilized, and if personal computers are connected in parallel, they may be used instead of expensive mainframe computers. Furthermore, we study the common problem of sorting large files using parallel computers and compare the performance with hat of sequential computers. These simulations demonstrate that Linearly connected personal computers can provide good computing power and speedup if the local computing load on these personal computers is not too high.

Original languageEnglish (US)
Title of host publicationProceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993
PublisherIEEE Computer Society
Pages308-317
Number of pages10
ISBN (Electronic)0818632305
DOIs
StatePublished - Jan 1 1993
Event26th Hawaii International Conference on System Sciences, HICSS 1993 - Wailea, United States
Duration: Jan 8 1993 → …

Publication series

NameProceedings of the Annual Hawaii International Conference on System Sciences
Volume2
ISSN (Print)1530-1605

Conference

Conference26th Hawaii International Conference on System Sciences, HICSS 1993
CountryUnited States
CityWailea
Period1/8/93 → …

Fingerprint

Personal computers
Scheduling algorithms
Sorting
Hardware

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Bautista, L. F., & Sarkar, D. (1993). Performance study of linearly connected parallel machines through simulation. In Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993 (pp. 308-317). [284097] (Proceedings of the Annual Hawaii International Conference on System Sciences; Vol. 2). IEEE Computer Society. https://doi.org/10.1109/HICSS.1993.284097

Performance study of linearly connected parallel machines through simulation. / Bautista, L. F.; Sarkar, Dilip.

Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993. IEEE Computer Society, 1993. p. 308-317 284097 (Proceedings of the Annual Hawaii International Conference on System Sciences; Vol. 2).

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

Bautista, LF & Sarkar, D 1993, Performance study of linearly connected parallel machines through simulation. in Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993., 284097, Proceedings of the Annual Hawaii International Conference on System Sciences, vol. 2, IEEE Computer Society, pp. 308-317, 26th Hawaii International Conference on System Sciences, HICSS 1993, Wailea, United States, 1/8/93. https://doi.org/10.1109/HICSS.1993.284097
Bautista LF, Sarkar D. Performance study of linearly connected parallel machines through simulation. In Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993. IEEE Computer Society. 1993. p. 308-317. 284097. (Proceedings of the Annual Hawaii International Conference on System Sciences). https://doi.org/10.1109/HICSS.1993.284097
Bautista, L. F. ; Sarkar, Dilip. / Performance study of linearly connected parallel machines through simulation. Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993. IEEE Computer Society, 1993. pp. 308-317 (Proceedings of the Annual Hawaii International Conference on System Sciences).
@inproceedings{250b9bc2a9334e32afcfea5e8b79c374,
title = "Performance study of linearly connected parallel machines through simulation",
abstract = "Recent advancements in hardware technology have increased the power of the conventional computer. However, a conventional computer can only perform one task at a given moment. Parallel computers consist of many processors working together simultaneously, which should make them faster and more powerful than the conventional computer. In this work, we study the performance of linearly connected parallel machines by examining several job scheduling algorithms for these computers. Our motivation for this study is the observation that the computation power of personal computers is poorly utilized, and if personal computers are connected in parallel, they may be used instead of expensive mainframe computers. Furthermore, we study the common problem of sorting large files using parallel computers and compare the performance with hat of sequential computers. These simulations demonstrate that Linearly connected personal computers can provide good computing power and speedup if the local computing load on these personal computers is not too high.",
author = "Bautista, {L. F.} and Dilip Sarkar",
year = "1993",
month = "1",
day = "1",
doi = "10.1109/HICSS.1993.284097",
language = "English (US)",
series = "Proceedings of the Annual Hawaii International Conference on System Sciences",
publisher = "IEEE Computer Society",
pages = "308--317",
booktitle = "Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993",

}

TY - GEN

T1 - Performance study of linearly connected parallel machines through simulation

AU - Bautista, L. F.

AU - Sarkar, Dilip

PY - 1993/1/1

Y1 - 1993/1/1

N2 - Recent advancements in hardware technology have increased the power of the conventional computer. However, a conventional computer can only perform one task at a given moment. Parallel computers consist of many processors working together simultaneously, which should make them faster and more powerful than the conventional computer. In this work, we study the performance of linearly connected parallel machines by examining several job scheduling algorithms for these computers. Our motivation for this study is the observation that the computation power of personal computers is poorly utilized, and if personal computers are connected in parallel, they may be used instead of expensive mainframe computers. Furthermore, we study the common problem of sorting large files using parallel computers and compare the performance with hat of sequential computers. These simulations demonstrate that Linearly connected personal computers can provide good computing power and speedup if the local computing load on these personal computers is not too high.

AB - Recent advancements in hardware technology have increased the power of the conventional computer. However, a conventional computer can only perform one task at a given moment. Parallel computers consist of many processors working together simultaneously, which should make them faster and more powerful than the conventional computer. In this work, we study the performance of linearly connected parallel machines by examining several job scheduling algorithms for these computers. Our motivation for this study is the observation that the computation power of personal computers is poorly utilized, and if personal computers are connected in parallel, they may be used instead of expensive mainframe computers. Furthermore, we study the common problem of sorting large files using parallel computers and compare the performance with hat of sequential computers. These simulations demonstrate that Linearly connected personal computers can provide good computing power and speedup if the local computing load on these personal computers is not too high.

UR - http://www.scopus.com/inward/record.url?scp=85063394170&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063394170&partnerID=8YFLogxK

U2 - 10.1109/HICSS.1993.284097

DO - 10.1109/HICSS.1993.284097

M3 - Conference contribution

T3 - Proceedings of the Annual Hawaii International Conference on System Sciences

SP - 308

EP - 317

BT - Proceedings of the 26th Hawaii International Conference on System Sciences, HICSS 1993

PB - IEEE Computer Society

ER -