A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines

Tong-Yee Lee, C. S. Raghavendra, H. Sivaraman

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

2 Citations (Scopus)

Abstract

Loops without dependences among iterations are a rich source of parallelism in many applications. Among these types of loops, non-uniform loops with variable ezecution times need eficient scheduling schemes to take advantages of the capabilities of parallel machines. In this paper, we present a global distributed control scheme (GDC) to schedule nonuniform loops on distributed memory parallel machines. GDC decentralizes scheduling controls among all processors with an attempt to keep heavily loaded processors being in charge of scheduling tasks. FOT comparative evaluation, GDC and other well-known scheduling schemes are implemented on a 512 processor Intel Delta parallel machine. Our experimental results show that GDC performs well on many applications with diflerent characteristics.

Original languageEnglish
Title of host publicationArchitecture and Software Technology
PublisherIEEE Computer Society
Pages243-250
Number of pages8
ISBN (Electronic)0818673249
DOIs
Publication statusPublished - 1996 Jan 1
Event29th Hawaii International Conference on System Sciences, HICSS 1996 - Wailea, United States
Duration: 1996 Jan 31996 Jan 6

Publication series

NameProceedings of the Annual Hawaii International Conference on System Sciences
Volume1
ISSN (Print)1530-1605

Conference

Conference29th Hawaii International Conference on System Sciences, HICSS 1996
CountryUnited States
CityWailea
Period96-01-0396-01-06

Fingerprint

Scheduling
Data storage equipment

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Lee, T-Y., Raghavendra, C. S., & Sivaraman, H. (1996). A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines. In Architecture and Software Technology (pp. 243-250). [495468] (Proceedings of the Annual Hawaii International Conference on System Sciences; Vol. 1). IEEE Computer Society. https://doi.org/10.1109/HICSS.1996.495468
Lee, Tong-Yee ; Raghavendra, C. S. ; Sivaraman, H. / A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines. Architecture and Software Technology. IEEE Computer Society, 1996. pp. 243-250 (Proceedings of the Annual Hawaii International Conference on System Sciences).
@inproceedings{50e06bebb6f042db8acde394a3ee8e60,
title = "A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines",
abstract = "Loops without dependences among iterations are a rich source of parallelism in many applications. Among these types of loops, non-uniform loops with variable ezecution times need eficient scheduling schemes to take advantages of the capabilities of parallel machines. In this paper, we present a global distributed control scheme (GDC) to schedule nonuniform loops on distributed memory parallel machines. GDC decentralizes scheduling controls among all processors with an attempt to keep heavily loaded processors being in charge of scheduling tasks. FOT comparative evaluation, GDC and other well-known scheduling schemes are implemented on a 512 processor Intel Delta parallel machine. Our experimental results show that GDC performs well on many applications with diflerent characteristics.",
author = "Tong-Yee Lee and Raghavendra, {C. S.} and H. Sivaraman",
year = "1996",
month = "1",
day = "1",
doi = "10.1109/HICSS.1996.495468",
language = "English",
series = "Proceedings of the Annual Hawaii International Conference on System Sciences",
publisher = "IEEE Computer Society",
pages = "243--250",
booktitle = "Architecture and Software Technology",
address = "United States",

}

Lee, T-Y, Raghavendra, CS & Sivaraman, H 1996, A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines. in Architecture and Software Technology., 495468, Proceedings of the Annual Hawaii International Conference on System Sciences, vol. 1, IEEE Computer Society, pp. 243-250, 29th Hawaii International Conference on System Sciences, HICSS 1996, Wailea, United States, 96-01-03. https://doi.org/10.1109/HICSS.1996.495468

A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines. / Lee, Tong-Yee; Raghavendra, C. S.; Sivaraman, H.

Architecture and Software Technology. IEEE Computer Society, 1996. p. 243-250 495468 (Proceedings of the Annual Hawaii International Conference on System Sciences; Vol. 1).

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

TY - GEN

T1 - A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines

AU - Lee, Tong-Yee

AU - Raghavendra, C. S.

AU - Sivaraman, H.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - Loops without dependences among iterations are a rich source of parallelism in many applications. Among these types of loops, non-uniform loops with variable ezecution times need eficient scheduling schemes to take advantages of the capabilities of parallel machines. In this paper, we present a global distributed control scheme (GDC) to schedule nonuniform loops on distributed memory parallel machines. GDC decentralizes scheduling controls among all processors with an attempt to keep heavily loaded processors being in charge of scheduling tasks. FOT comparative evaluation, GDC and other well-known scheduling schemes are implemented on a 512 processor Intel Delta parallel machine. Our experimental results show that GDC performs well on many applications with diflerent characteristics.

AB - Loops without dependences among iterations are a rich source of parallelism in many applications. Among these types of loops, non-uniform loops with variable ezecution times need eficient scheduling schemes to take advantages of the capabilities of parallel machines. In this paper, we present a global distributed control scheme (GDC) to schedule nonuniform loops on distributed memory parallel machines. GDC decentralizes scheduling controls among all processors with an attempt to keep heavily loaded processors being in charge of scheduling tasks. FOT comparative evaluation, GDC and other well-known scheduling schemes are implemented on a 512 processor Intel Delta parallel machine. Our experimental results show that GDC performs well on many applications with diflerent characteristics.

UR - http://www.scopus.com/inward/record.url?scp=0342745565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0342745565&partnerID=8YFLogxK

U2 - 10.1109/HICSS.1996.495468

DO - 10.1109/HICSS.1996.495468

M3 - Conference contribution

AN - SCOPUS:0342745565

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

SP - 243

EP - 250

BT - Architecture and Software Technology

PB - IEEE Computer Society

ER -

Lee T-Y, Raghavendra CS, Sivaraman H. A practical scheduling scheme for non-uniform parallel loops on distributed memory parallel machines. In Architecture and Software Technology. IEEE Computer Society. 1996. p. 243-250. 495468. (Proceedings of the Annual Hawaii International Conference on System Sciences). https://doi.org/10.1109/HICSS.1996.495468