OMPCUDA: OpenMP execution framework for CUDA based on omni OpenMP compiler

Satoshi Ohshima, Shoichi Hirasawa, Hiroki Honda

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

9 Citations (Scopus)

Abstract

Arithmetic performance with GPGPU attracts attention. However, the difficulty of the programming poses a problem. We have proposed GPGPU programming which used the existing parallel programming technique. We are now developing OpenMP framework for GPU as a concrete of our proposal. The framework is based on Omni OpenMP Compiler and named "OMPCUDA". In this paper we describe a design and an implementation of OMPCUDA. We evaluated using test programs, and validated that parallel improvement in the speed could be easily carried out in the same code as the existing OpenMP.

Original languageEnglish
Title of host publicationBeyond Loop Level Parallelism in OpenMP
Subtitle of host publicationAccelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings
Pages161-173
Number of pages13
DOIs
Publication statusPublished - Aug 13 2010
Event6th International Workshop on OpenMP, IWOMP 2010 - Tsukuba, Japan
Duration: Jun 14 2010Jun 16 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6132 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other6th International Workshop on OpenMP, IWOMP 2010
CountryJapan
CityTsukuba
Period6/14/106/16/10

Fingerprint

Parallel programming
OpenMP
Computer programming
Compiler
GPGPU
Concretes
Programming
Parallel Programming
Framework
Graphics processing unit

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Ohshima, S., Hirasawa, S., & Honda, H. (2010). OMPCUDA: OpenMP execution framework for CUDA based on omni OpenMP compiler. In Beyond Loop Level Parallelism in OpenMP: Accelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings (pp. 161-173). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6132 LNCS). https://doi.org/10.1007/978-3-642-13217-9_13

OMPCUDA : OpenMP execution framework for CUDA based on omni OpenMP compiler. / Ohshima, Satoshi; Hirasawa, Shoichi; Honda, Hiroki.

Beyond Loop Level Parallelism in OpenMP: Accelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings. 2010. p. 161-173 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6132 LNCS).

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

Ohshima, S, Hirasawa, S & Honda, H 2010, OMPCUDA: OpenMP execution framework for CUDA based on omni OpenMP compiler. in Beyond Loop Level Parallelism in OpenMP: Accelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6132 LNCS, pp. 161-173, 6th International Workshop on OpenMP, IWOMP 2010, Tsukuba, Japan, 6/14/10. https://doi.org/10.1007/978-3-642-13217-9_13
Ohshima S, Hirasawa S, Honda H. OMPCUDA: OpenMP execution framework for CUDA based on omni OpenMP compiler. In Beyond Loop Level Parallelism in OpenMP: Accelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings. 2010. p. 161-173. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-13217-9_13
Ohshima, Satoshi ; Hirasawa, Shoichi ; Honda, Hiroki. / OMPCUDA : OpenMP execution framework for CUDA based on omni OpenMP compiler. Beyond Loop Level Parallelism in OpenMP: Accelerators, Tasking and More - 6th International Workshop on OpenMP, IWOMP 2010, Proceedings. 2010. pp. 161-173 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{87508c0a81034fa0b92602895e87a090,
title = "OMPCUDA: OpenMP execution framework for CUDA based on omni OpenMP compiler",
abstract = "Arithmetic performance with GPGPU attracts attention. However, the difficulty of the programming poses a problem. We have proposed GPGPU programming which used the existing parallel programming technique. We are now developing OpenMP framework for GPU as a concrete of our proposal. The framework is based on Omni OpenMP Compiler and named {"}OMPCUDA{"}. In this paper we describe a design and an implementation of OMPCUDA. We evaluated using test programs, and validated that parallel improvement in the speed could be easily carried out in the same code as the existing OpenMP.",
author = "Satoshi Ohshima and Shoichi Hirasawa and Hiroki Honda",
year = "2010",
month = "8",
day = "13",
doi = "10.1007/978-3-642-13217-9_13",
language = "English",
isbn = "3642132162",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "161--173",
booktitle = "Beyond Loop Level Parallelism in OpenMP",

}

TY - GEN

T1 - OMPCUDA

T2 - OpenMP execution framework for CUDA based on omni OpenMP compiler

AU - Ohshima, Satoshi

AU - Hirasawa, Shoichi

AU - Honda, Hiroki

PY - 2010/8/13

Y1 - 2010/8/13

N2 - Arithmetic performance with GPGPU attracts attention. However, the difficulty of the programming poses a problem. We have proposed GPGPU programming which used the existing parallel programming technique. We are now developing OpenMP framework for GPU as a concrete of our proposal. The framework is based on Omni OpenMP Compiler and named "OMPCUDA". In this paper we describe a design and an implementation of OMPCUDA. We evaluated using test programs, and validated that parallel improvement in the speed could be easily carried out in the same code as the existing OpenMP.

AB - Arithmetic performance with GPGPU attracts attention. However, the difficulty of the programming poses a problem. We have proposed GPGPU programming which used the existing parallel programming technique. We are now developing OpenMP framework for GPU as a concrete of our proposal. The framework is based on Omni OpenMP Compiler and named "OMPCUDA". In this paper we describe a design and an implementation of OMPCUDA. We evaluated using test programs, and validated that parallel improvement in the speed could be easily carried out in the same code as the existing OpenMP.

UR - http://www.scopus.com/inward/record.url?scp=77955354458&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955354458&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-13217-9_13

DO - 10.1007/978-3-642-13217-9_13

M3 - Conference contribution

AN - SCOPUS:77955354458

SN - 3642132162

SN - 9783642132162

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

SP - 161

EP - 173

BT - Beyond Loop Level Parallelism in OpenMP

ER -