A new version of a parallel production system machine, MANJI-II

Jun Miyazaki, Kenji Takeda, Hideharu Amano, Hideo Aiso

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

Abstract

Parallel systems for OPS5 have been developed. In such systems, parallel implementations of the Rete algorithm are adopted because the number of pattern matchings is minimized in the original algorithm. However, conventional approaches for parallel Rete algorithms require special hardware to cope with the dynamic process allocation and frequent communication. Dedicated machines are necessary for such methods. In this paper, a new parallel execution method for general purpose multiprocessors is proposed. In this method, nodes in the Rete network are statically mapped onto a small number of processors (PUs) before execution. Since processes are not migrated and are not allocated dynamically, data structures can be designed so as to minimize the communication between PUs. Also the synchronization and process control can be simplified. The key for the method is static mapping algorithms. Since the characteristics of the Rete network are completely different from the data dependency graph which is suitable for extracting inherent parallelism in general programming languages like FORTRAN, the conventional static mapping techniques cannot be applied. Some new mapping strategies are discussed and evaluated. Although the method is utilized for most types of multiprocessors, an optimized architecture which is called MANJI-II is designed and proposed. MANJI-II supports quick transfer of tokens in the Rete network, by using a simple multicast mechanism, and non-cached shared memory. Although MANJI-II provides almost the same performance as that of dedicated architecture including the previous version of MANJI, the total cost of MANJI-II is extremely low.

Original languageEnglish
Title of host publicationDatabase Machines - 6th International Workshop, IWDM 1989, Proceedings
PublisherSpringer Verlag
Pages317-330
Number of pages14
Volume368 LNCS
ISBN (Print)9783540513247
DOIs
Publication statusPublished - 1989
Event6th International Workshop on Database Machines, IWDM 1989 - Deauville, France
Duration: 1989 Jun 191989 Jun 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume368 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other6th International Workshop on Database Machines, IWDM 1989
CountryFrance
CityDeauville
Period89/6/1989/6/21

Fingerprint

Parallel Systems
Production Systems
FORTRAN (programming language)
Multiprocessor
D.3.2 [Programming Languages]: Language Classifications - Fortran
Pattern matching
Communication
Parallel algorithms
Computer programming languages
Computer hardware
Process control
Data structures
Synchronization
Dependency Graph
Data Dependency
Data storage equipment
Dynamic Process
Pattern Matching
Parallel Implementation
Shared Memory

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Miyazaki, J., Takeda, K., Amano, H., & Aiso, H. (1989). A new version of a parallel production system machine, MANJI-II. In Database Machines - 6th International Workshop, IWDM 1989, Proceedings (Vol. 368 LNCS, pp. 317-330). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 368 LNCS). Springer Verlag. https://doi.org/10.1007/3-540-51324-8_44

A new version of a parallel production system machine, MANJI-II. / Miyazaki, Jun; Takeda, Kenji; Amano, Hideharu; Aiso, Hideo.

Database Machines - 6th International Workshop, IWDM 1989, Proceedings. Vol. 368 LNCS Springer Verlag, 1989. p. 317-330 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 368 LNCS).

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

Miyazaki, J, Takeda, K, Amano, H & Aiso, H 1989, A new version of a parallel production system machine, MANJI-II. in Database Machines - 6th International Workshop, IWDM 1989, Proceedings. vol. 368 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 368 LNCS, Springer Verlag, pp. 317-330, 6th International Workshop on Database Machines, IWDM 1989, Deauville, France, 89/6/19. https://doi.org/10.1007/3-540-51324-8_44
Miyazaki J, Takeda K, Amano H, Aiso H. A new version of a parallel production system machine, MANJI-II. In Database Machines - 6th International Workshop, IWDM 1989, Proceedings. Vol. 368 LNCS. Springer Verlag. 1989. p. 317-330. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-51324-8_44
Miyazaki, Jun ; Takeda, Kenji ; Amano, Hideharu ; Aiso, Hideo. / A new version of a parallel production system machine, MANJI-II. Database Machines - 6th International Workshop, IWDM 1989, Proceedings. Vol. 368 LNCS Springer Verlag, 1989. pp. 317-330 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5baa89a453ed4ff581caeeae407e2cd8,
title = "A new version of a parallel production system machine, MANJI-II",
abstract = "Parallel systems for OPS5 have been developed. In such systems, parallel implementations of the Rete algorithm are adopted because the number of pattern matchings is minimized in the original algorithm. However, conventional approaches for parallel Rete algorithms require special hardware to cope with the dynamic process allocation and frequent communication. Dedicated machines are necessary for such methods. In this paper, a new parallel execution method for general purpose multiprocessors is proposed. In this method, nodes in the Rete network are statically mapped onto a small number of processors (PUs) before execution. Since processes are not migrated and are not allocated dynamically, data structures can be designed so as to minimize the communication between PUs. Also the synchronization and process control can be simplified. The key for the method is static mapping algorithms. Since the characteristics of the Rete network are completely different from the data dependency graph which is suitable for extracting inherent parallelism in general programming languages like FORTRAN, the conventional static mapping techniques cannot be applied. Some new mapping strategies are discussed and evaluated. Although the method is utilized for most types of multiprocessors, an optimized architecture which is called MANJI-II is designed and proposed. MANJI-II supports quick transfer of tokens in the Rete network, by using a simple multicast mechanism, and non-cached shared memory. Although MANJI-II provides almost the same performance as that of dedicated architecture including the previous version of MANJI, the total cost of MANJI-II is extremely low.",
author = "Jun Miyazaki and Kenji Takeda and Hideharu Amano and Hideo Aiso",
year = "1989",
doi = "10.1007/3-540-51324-8_44",
language = "English",
isbn = "9783540513247",
volume = "368 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "317--330",
booktitle = "Database Machines - 6th International Workshop, IWDM 1989, Proceedings",
address = "Germany",

}

TY - GEN

T1 - A new version of a parallel production system machine, MANJI-II

AU - Miyazaki, Jun

AU - Takeda, Kenji

AU - Amano, Hideharu

AU - Aiso, Hideo

PY - 1989

Y1 - 1989

N2 - Parallel systems for OPS5 have been developed. In such systems, parallel implementations of the Rete algorithm are adopted because the number of pattern matchings is minimized in the original algorithm. However, conventional approaches for parallel Rete algorithms require special hardware to cope with the dynamic process allocation and frequent communication. Dedicated machines are necessary for such methods. In this paper, a new parallel execution method for general purpose multiprocessors is proposed. In this method, nodes in the Rete network are statically mapped onto a small number of processors (PUs) before execution. Since processes are not migrated and are not allocated dynamically, data structures can be designed so as to minimize the communication between PUs. Also the synchronization and process control can be simplified. The key for the method is static mapping algorithms. Since the characteristics of the Rete network are completely different from the data dependency graph which is suitable for extracting inherent parallelism in general programming languages like FORTRAN, the conventional static mapping techniques cannot be applied. Some new mapping strategies are discussed and evaluated. Although the method is utilized for most types of multiprocessors, an optimized architecture which is called MANJI-II is designed and proposed. MANJI-II supports quick transfer of tokens in the Rete network, by using a simple multicast mechanism, and non-cached shared memory. Although MANJI-II provides almost the same performance as that of dedicated architecture including the previous version of MANJI, the total cost of MANJI-II is extremely low.

AB - Parallel systems for OPS5 have been developed. In such systems, parallel implementations of the Rete algorithm are adopted because the number of pattern matchings is minimized in the original algorithm. However, conventional approaches for parallel Rete algorithms require special hardware to cope with the dynamic process allocation and frequent communication. Dedicated machines are necessary for such methods. In this paper, a new parallel execution method for general purpose multiprocessors is proposed. In this method, nodes in the Rete network are statically mapped onto a small number of processors (PUs) before execution. Since processes are not migrated and are not allocated dynamically, data structures can be designed so as to minimize the communication between PUs. Also the synchronization and process control can be simplified. The key for the method is static mapping algorithms. Since the characteristics of the Rete network are completely different from the data dependency graph which is suitable for extracting inherent parallelism in general programming languages like FORTRAN, the conventional static mapping techniques cannot be applied. Some new mapping strategies are discussed and evaluated. Although the method is utilized for most types of multiprocessors, an optimized architecture which is called MANJI-II is designed and proposed. MANJI-II supports quick transfer of tokens in the Rete network, by using a simple multicast mechanism, and non-cached shared memory. Although MANJI-II provides almost the same performance as that of dedicated architecture including the previous version of MANJI, the total cost of MANJI-II is extremely low.

UR - http://www.scopus.com/inward/record.url?scp=85031910536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031910536&partnerID=8YFLogxK

U2 - 10.1007/3-540-51324-8_44

DO - 10.1007/3-540-51324-8_44

M3 - Conference contribution

AN - SCOPUS:85031910536

SN - 9783540513247

VL - 368 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 317

EP - 330

BT - Database Machines - 6th International Workshop, IWDM 1989, Proceedings

PB - Springer Verlag

ER -