An object-oriented framework for loop parallelization

Yoichi Omori, Kazuki Joe, Akira Fukuda

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

Abstract

The main goal of parallelizing compiler research is the ability to produce efficient parallel programs and the portability because of wide choices of current and future available parallel computer architectures. Since the design of paralletizing compilers tends to be more complicated than conventional compilers, it is extremely difficult to achieve both the efficiency and the portability. To meet this problem, we have investigated an application of object oriented design to parallelizing compilers. Our parallelizing compiler design is based on abstractions of intermediate representations of loops and class definitions for them. In this paper, we focus on loop parallelization and propose a framework where loop parallelization process is divided into three phases and the optimization of loops is performed via cyclic use of those three phases. The class of each phase is hierarchically derived from intermediate representations of loops. This increases the portability of resultant parallelizing compilers. Furthermore, one of the phases uses a reservation table of hardware resource to obtain practicM optimization of parallel programs for given hardware resource.

Original languageEnglish
Title of host publicationHigh Performance Computing - International Symposium, ISHPC 1997, Proceedings
EditorsKeijiro Araki, Makoto Amamiya, Constantine Polychronopoulos, Kazuki Joe
PublisherSpringer Verlag
Pages351-360
Number of pages10
ISBN (Print)3540637664, 9783540637660
DOIs
Publication statusPublished - Jan 1 1997
Externally publishedYes
EventInternational Symposium on High Performance Computing, ISHPC 1997 - Fukuoka, Japan
Duration: Nov 4 1997Nov 6 1997

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1336
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Symposium on High Performance Computing, ISHPC 1997
CountryJapan
CityFukuoka
Period11/4/9711/6/97

Fingerprint

Parallelizing Compilers
Parallelization
Object-oriented
Portability
Parallel Programs
Hardware
Computer architecture
Compiler
Object-oriented Design
Resources
Computer Architecture
Optimization
Parallel Architectures
Reservation
Parallel Computers
Table
Framework
Tend
Design
Class

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Omori, Y., Joe, K., & Fukuda, A. (1997). An object-oriented framework for loop parallelization. In K. Araki, M. Amamiya, C. Polychronopoulos, & K. Joe (Eds.), High Performance Computing - International Symposium, ISHPC 1997, Proceedings (pp. 351-360). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1336). Springer Verlag. https://doi.org/10.1007/BFb0024230

An object-oriented framework for loop parallelization. / Omori, Yoichi; Joe, Kazuki; Fukuda, Akira.

High Performance Computing - International Symposium, ISHPC 1997, Proceedings. ed. / Keijiro Araki; Makoto Amamiya; Constantine Polychronopoulos; Kazuki Joe. Springer Verlag, 1997. p. 351-360 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1336).

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

Omori, Y, Joe, K & Fukuda, A 1997, An object-oriented framework for loop parallelization. in K Araki, M Amamiya, C Polychronopoulos & K Joe (eds), High Performance Computing - International Symposium, ISHPC 1997, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1336, Springer Verlag, pp. 351-360, International Symposium on High Performance Computing, ISHPC 1997, Fukuoka, Japan, 11/4/97. https://doi.org/10.1007/BFb0024230
Omori Y, Joe K, Fukuda A. An object-oriented framework for loop parallelization. In Araki K, Amamiya M, Polychronopoulos C, Joe K, editors, High Performance Computing - International Symposium, ISHPC 1997, Proceedings. Springer Verlag. 1997. p. 351-360. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/BFb0024230
Omori, Yoichi ; Joe, Kazuki ; Fukuda, Akira. / An object-oriented framework for loop parallelization. High Performance Computing - International Symposium, ISHPC 1997, Proceedings. editor / Keijiro Araki ; Makoto Amamiya ; Constantine Polychronopoulos ; Kazuki Joe. Springer Verlag, 1997. pp. 351-360 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{2b2ec4202cd24adf86d9f418747aad9e,
title = "An object-oriented framework for loop parallelization",
abstract = "The main goal of parallelizing compiler research is the ability to produce efficient parallel programs and the portability because of wide choices of current and future available parallel computer architectures. Since the design of paralletizing compilers tends to be more complicated than conventional compilers, it is extremely difficult to achieve both the efficiency and the portability. To meet this problem, we have investigated an application of object oriented design to parallelizing compilers. Our parallelizing compiler design is based on abstractions of intermediate representations of loops and class definitions for them. In this paper, we focus on loop parallelization and propose a framework where loop parallelization process is divided into three phases and the optimization of loops is performed via cyclic use of those three phases. The class of each phase is hierarchically derived from intermediate representations of loops. This increases the portability of resultant parallelizing compilers. Furthermore, one of the phases uses a reservation table of hardware resource to obtain practicM optimization of parallel programs for given hardware resource.",
author = "Yoichi Omori and Kazuki Joe and Akira Fukuda",
year = "1997",
month = "1",
day = "1",
doi = "10.1007/BFb0024230",
language = "English",
isbn = "3540637664",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "351--360",
editor = "Keijiro Araki and Makoto Amamiya and Constantine Polychronopoulos and Kazuki Joe",
booktitle = "High Performance Computing - International Symposium, ISHPC 1997, Proceedings",
address = "Germany",

}

TY - GEN

T1 - An object-oriented framework for loop parallelization

AU - Omori, Yoichi

AU - Joe, Kazuki

AU - Fukuda, Akira

PY - 1997/1/1

Y1 - 1997/1/1

N2 - The main goal of parallelizing compiler research is the ability to produce efficient parallel programs and the portability because of wide choices of current and future available parallel computer architectures. Since the design of paralletizing compilers tends to be more complicated than conventional compilers, it is extremely difficult to achieve both the efficiency and the portability. To meet this problem, we have investigated an application of object oriented design to parallelizing compilers. Our parallelizing compiler design is based on abstractions of intermediate representations of loops and class definitions for them. In this paper, we focus on loop parallelization and propose a framework where loop parallelization process is divided into three phases and the optimization of loops is performed via cyclic use of those three phases. The class of each phase is hierarchically derived from intermediate representations of loops. This increases the portability of resultant parallelizing compilers. Furthermore, one of the phases uses a reservation table of hardware resource to obtain practicM optimization of parallel programs for given hardware resource.

AB - The main goal of parallelizing compiler research is the ability to produce efficient parallel programs and the portability because of wide choices of current and future available parallel computer architectures. Since the design of paralletizing compilers tends to be more complicated than conventional compilers, it is extremely difficult to achieve both the efficiency and the portability. To meet this problem, we have investigated an application of object oriented design to parallelizing compilers. Our parallelizing compiler design is based on abstractions of intermediate representations of loops and class definitions for them. In this paper, we focus on loop parallelization and propose a framework where loop parallelization process is divided into three phases and the optimization of loops is performed via cyclic use of those three phases. The class of each phase is hierarchically derived from intermediate representations of loops. This increases the portability of resultant parallelizing compilers. Furthermore, one of the phases uses a reservation table of hardware resource to obtain practicM optimization of parallel programs for given hardware resource.

UR - http://www.scopus.com/inward/record.url?scp=84958641450&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958641450&partnerID=8YFLogxK

U2 - 10.1007/BFb0024230

DO - 10.1007/BFb0024230

M3 - Conference contribution

SN - 3540637664

SN - 9783540637660

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

SP - 351

EP - 360

BT - High Performance Computing - International Symposium, ISHPC 1997, Proceedings

A2 - Araki, Keijiro

A2 - Amamiya, Makoto

A2 - Polychronopoulos, Constantine

A2 - Joe, Kazuki

PB - Springer Verlag

ER -