OpenMP® runtime instrumentation for optimization

Taru Doodi; Jonathan Peyton; Jim Cownie; Maria Garzaran; Rubasri Kalidas; Jeongnim Kim; Amrita Mathuriya; Terry Wilmarth; Gengbin Zheng

doi:10.1007/978-3-319-65578-9_19

OpenMP® runtime instrumentation for optimization

Taru Doodi, Jonathan Peyton, Jim Cownie, Maria Garzaran, Rubasri Kalidas, Jeongnim Kim, Amrita Mathuriya, Terry Wilmarth, Gengbin Zheng

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The OpenMP (The OpenMP name is a registered trademark of the OpenMP Architecture Review Board.) application programming interface provides a simple way for programmers to write parallel programs that are portable between machines and vendors. Programmers parallelize their programs to obtain higher performance, but, as the number of cores per processor increases, taking advantage of parallelism efficiently becomes more difficult. To facilitate efficient parallelization and avoid poor utilization of machine resources, programmers need to know where an application is spending time and what factors hinder scalability. In this paper, we present a Tool for Runtime Instrumentation of OpenMP programs (TRIO) that automatically collects statistics about an application’s use of the OpenMP runtime. TRIO provides statistics such as the total number of times an OpenMP construct is called, the time spent in each OpenMP construct, and the total time spent within the OpenMP runtime. TRIO helps to identify the runtime calls where a program spends most of the time and which constructs are called the most at runtime.

Original language	English (US)
Title of host publication	Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings
Editors	Bronis R. de Supinski, Barbara M. Chapman, Christian Terboven, Matthias S. Muller, Stephen L. Olivier
Publisher	Springer
Pages	281-295
Number of pages	15
ISBN (Print)	9783319655772
DOIs	https://doi.org/10.1007/978-3-319-65578-9_19
State	Published - Jan 1 2017
Externally published	Yes
Event	13th International Workshop on OpenMP, IWOMP 2017 - Stony Brook, United States Duration: Sep 20 2017 → Sep 22 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10468 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	13th International Workshop on OpenMP, IWOMP 2017
Country/Territory	United States
City	Stony Brook
Period	9/20/17 → 9/22/17

Keywords

OpenMP constructs
Runtime instrumentation

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-319-65578-9_19

Cite this

Doodi, T., Peyton, J., Cownie, J., Garzaran, M., Kalidas, R., Kim, J., Mathuriya, A., Wilmarth, T., & Zheng, G. (2017). OpenMP® runtime instrumentation for optimization. In B. R. de Supinski, B. M. Chapman, C. Terboven, M. S. Muller, & S. L. Olivier (Eds.), Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings (pp. 281-295). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10468 LNCS). Springer. https://doi.org/10.1007/978-3-319-65578-9_19

OpenMP® runtime instrumentation for optimization. / Doodi, Taru; Peyton, Jonathan; Cownie, Jim; Garzaran, Maria; Kalidas, Rubasri; Kim, Jeongnim; Mathuriya, Amrita; Wilmarth, Terry; Zheng, Gengbin.

Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings. ed. / Bronis R. de Supinski; Barbara M. Chapman; Christian Terboven; Matthias S. Muller; Stephen L. Olivier. Springer, 2017. p. 281-295 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10468 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Doodi, T, Peyton, J, Cownie, J, Garzaran, M, Kalidas, R, Kim, J, Mathuriya, A, Wilmarth, T & Zheng, G 2017, OpenMP® runtime instrumentation for optimization. in BR de Supinski, BM Chapman, C Terboven, MS Muller & SL Olivier (eds), Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10468 LNCS, Springer, pp. 281-295, 13th International Workshop on OpenMP, IWOMP 2017, Stony Brook, United States, 9/20/17. https://doi.org/10.1007/978-3-319-65578-9_19

Doodi T, Peyton J, Cownie J, Garzaran M, Kalidas R, Kim J et al. OpenMP® runtime instrumentation for optimization. In de Supinski BR, Chapman BM, Terboven C, Muller MS, Olivier SL, editors, Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings. Springer. 2017. p. 281-295. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-65578-9_19

Doodi, Taru ; Peyton, Jonathan ; Cownie, Jim ; Garzaran, Maria ; Kalidas, Rubasri ; Kim, Jeongnim ; Mathuriya, Amrita ; Wilmarth, Terry ; Zheng, Gengbin. / OpenMP® runtime instrumentation for optimization. Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings. editor / Bronis R. de Supinski ; Barbara M. Chapman ; Christian Terboven ; Matthias S. Muller ; Stephen L. Olivier. Springer, 2017. pp. 281-295 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{14523a9dbae54c9db2e8e640c14789fb,

title = "OpenMP{\textregistered} runtime instrumentation for optimization",

abstract = "The OpenMP (The OpenMP name is a registered trademark of the OpenMP Architecture Review Board.) application programming interface provides a simple way for programmers to write parallel programs that are portable between machines and vendors. Programmers parallelize their programs to obtain higher performance, but, as the number of cores per processor increases, taking advantage of parallelism efficiently becomes more difficult. To facilitate efficient parallelization and avoid poor utilization of machine resources, programmers need to know where an application is spending time and what factors hinder scalability. In this paper, we present a Tool for Runtime Instrumentation of OpenMP programs (TRIO) that automatically collects statistics about an application{\textquoteright}s use of the OpenMP runtime. TRIO provides statistics such as the total number of times an OpenMP construct is called, the time spent in each OpenMP construct, and the total time spent within the OpenMP runtime. TRIO helps to identify the runtime calls where a program spends most of the time and which constructs are called the most at runtime.",

keywords = "OpenMP constructs, Runtime instrumentation",

author = "Taru Doodi and Jonathan Peyton and Jim Cownie and Maria Garzaran and Rubasri Kalidas and Jeongnim Kim and Amrita Mathuriya and Terry Wilmarth and Gengbin Zheng",

year = "2017",

month = jan,

day = "1",

doi = "10.1007/978-3-319-65578-9_19",

language = "English (US)",

isbn = "9783319655772",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "281--295",

editor = "{de Supinski}, {Bronis R.} and Chapman, {Barbara M.} and Christian Terboven and Muller, {Matthias S.} and Olivier, {Stephen L.}",

booktitle = "Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings",

address = "Germany",

note = "13th International Workshop on OpenMP, IWOMP 2017 ; Conference date: 20-09-2017 Through 22-09-2017",

}

TY - GEN

T1 - OpenMP® runtime instrumentation for optimization

AU - Doodi, Taru

AU - Peyton, Jonathan

AU - Cownie, Jim

AU - Garzaran, Maria

AU - Kalidas, Rubasri

AU - Kim, Jeongnim

AU - Mathuriya, Amrita

AU - Wilmarth, Terry

AU - Zheng, Gengbin

PY - 2017/1/1

Y1 - 2017/1/1

N2 - The OpenMP (The OpenMP name is a registered trademark of the OpenMP Architecture Review Board.) application programming interface provides a simple way for programmers to write parallel programs that are portable between machines and vendors. Programmers parallelize their programs to obtain higher performance, but, as the number of cores per processor increases, taking advantage of parallelism efficiently becomes more difficult. To facilitate efficient parallelization and avoid poor utilization of machine resources, programmers need to know where an application is spending time and what factors hinder scalability. In this paper, we present a Tool for Runtime Instrumentation of OpenMP programs (TRIO) that automatically collects statistics about an application’s use of the OpenMP runtime. TRIO provides statistics such as the total number of times an OpenMP construct is called, the time spent in each OpenMP construct, and the total time spent within the OpenMP runtime. TRIO helps to identify the runtime calls where a program spends most of the time and which constructs are called the most at runtime.

AB - The OpenMP (The OpenMP name is a registered trademark of the OpenMP Architecture Review Board.) application programming interface provides a simple way for programmers to write parallel programs that are portable between machines and vendors. Programmers parallelize their programs to obtain higher performance, but, as the number of cores per processor increases, taking advantage of parallelism efficiently becomes more difficult. To facilitate efficient parallelization and avoid poor utilization of machine resources, programmers need to know where an application is spending time and what factors hinder scalability. In this paper, we present a Tool for Runtime Instrumentation of OpenMP programs (TRIO) that automatically collects statistics about an application’s use of the OpenMP runtime. TRIO provides statistics such as the total number of times an OpenMP construct is called, the time spent in each OpenMP construct, and the total time spent within the OpenMP runtime. TRIO helps to identify the runtime calls where a program spends most of the time and which constructs are called the most at runtime.

KW - OpenMP constructs

KW - Runtime instrumentation

UR - http://www.scopus.com/inward/record.url?scp=85029391932&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029391932&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-65578-9_19

DO - 10.1007/978-3-319-65578-9_19

M3 - Conference contribution

AN - SCOPUS:85029391932

SN - 9783319655772

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

SP - 281

EP - 295

BT - Scaling OpenMP for Exascale Performance and Portability - 13th International Workshop on OpenMP, IWOMP 2017, Proceedings

A2 - de Supinski, Bronis R.

A2 - Chapman, Barbara M.

A2 - Terboven, Christian

A2 - Muller, Matthias S.

A2 - Olivier, Stephen L.

PB - Springer

T2 - 13th International Workshop on OpenMP, IWOMP 2017

Y2 - 20 September 2017 through 22 September 2017

ER -

OpenMP® runtime instrumentation for optimization

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other links

Fingerprint

Cite this