PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS.

Kemal Oflazer

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

24 Citations (Scopus)

Abstract

The results of an analysis of production level parallelism in OPS5 production system programs are presented. The results indicate that, contrary to most expectations, the effective production level parallelism in this class of production systems is very low compared to the number of productions in these systems. Hence, significant speed-ups in executing such systems would be obtained by combining the limited parallelism with fast hardware and overlapped processing, rather than by massively parallel approaches using simple processors. The number of processors in a parallel processing system is also presented. The goal of partitioning is to improve the speed-up provided by the limited parallelism by finding assignments of productions to processors that achieve a more balanced load for each processor.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Parallel Processing
EditorsRobert M. Keller
Place of PublicationNew York, NY, USA
PublisherIEEE
Pages92-100
Number of pages9
ISBN (Print)081860560X
Publication statusPublished - 1984
Externally publishedYes

Fingerprint

Processing
Parallel processing systems
Hardware

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Oflazer, K. (1984). PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS. In R. M. Keller (Ed.), Proceedings of the International Conference on Parallel Processing (pp. 92-100). New York, NY, USA: IEEE.

PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS. / Oflazer, Kemal.

Proceedings of the International Conference on Parallel Processing. ed. / Robert M. Keller. New York, NY, USA : IEEE, 1984. p. 92-100.

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

Oflazer, K 1984, PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS. in RM Keller (ed.), Proceedings of the International Conference on Parallel Processing. IEEE, New York, NY, USA, pp. 92-100.
Oflazer K. PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS. In Keller RM, editor, Proceedings of the International Conference on Parallel Processing. New York, NY, USA: IEEE. 1984. p. 92-100
Oflazer, Kemal. / PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS. Proceedings of the International Conference on Parallel Processing. editor / Robert M. Keller. New York, NY, USA : IEEE, 1984. pp. 92-100
@inproceedings{829bd006a0dd4ee6a6b1c1cd9be3b8a6,
title = "PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS.",
abstract = "The results of an analysis of production level parallelism in OPS5 production system programs are presented. The results indicate that, contrary to most expectations, the effective production level parallelism in this class of production systems is very low compared to the number of productions in these systems. Hence, significant speed-ups in executing such systems would be obtained by combining the limited parallelism with fast hardware and overlapped processing, rather than by massively parallel approaches using simple processors. The number of processors in a parallel processing system is also presented. The goal of partitioning is to improve the speed-up provided by the limited parallelism by finding assignments of productions to processors that achieve a more balanced load for each processor.",
author = "Kemal Oflazer",
year = "1984",
language = "English",
isbn = "081860560X",
pages = "92--100",
editor = "Keller, {Robert M.}",
booktitle = "Proceedings of the International Conference on Parallel Processing",
publisher = "IEEE",

}

TY - GEN

T1 - PARTITIONING IN PARALLEL PROCESSING OF PRODUCTION SYSTEMS.

AU - Oflazer, Kemal

PY - 1984

Y1 - 1984

N2 - The results of an analysis of production level parallelism in OPS5 production system programs are presented. The results indicate that, contrary to most expectations, the effective production level parallelism in this class of production systems is very low compared to the number of productions in these systems. Hence, significant speed-ups in executing such systems would be obtained by combining the limited parallelism with fast hardware and overlapped processing, rather than by massively parallel approaches using simple processors. The number of processors in a parallel processing system is also presented. The goal of partitioning is to improve the speed-up provided by the limited parallelism by finding assignments of productions to processors that achieve a more balanced load for each processor.

AB - The results of an analysis of production level parallelism in OPS5 production system programs are presented. The results indicate that, contrary to most expectations, the effective production level parallelism in this class of production systems is very low compared to the number of productions in these systems. Hence, significant speed-ups in executing such systems would be obtained by combining the limited parallelism with fast hardware and overlapped processing, rather than by massively parallel approaches using simple processors. The number of processors in a parallel processing system is also presented. The goal of partitioning is to improve the speed-up provided by the limited parallelism by finding assignments of productions to processors that achieve a more balanced load for each processor.

UR - http://www.scopus.com/inward/record.url?scp=0021542157&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0021542157&partnerID=8YFLogxK

M3 - Conference contribution

SN - 081860560X

SP - 92

EP - 100

BT - Proceedings of the International Conference on Parallel Processing

A2 - Keller, Robert M.

PB - IEEE

CY - New York, NY, USA

ER -