Identifying and scheduling loop chains using directives

Ian J. Bertolacci; Michelle Mills Strout; Stephen Guzik; Jordan Riley; Catherine Olschanowsky

doi:10.1109/WACCPD.2016.010

Identifying and scheduling loop chains using directives

Ian J. Bertolacci, Michelle Mills Strout, Stephen Guzik, Jordan Riley, Catherine Olschanowsky

Computer Science

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

7 Scopus citations

Abstract

Exposing opportunities for parallelization while explicitly managing data locality is the primary challenge to porting and optimizing existing computational science simulation codes to improve performance and accuracy. OpenMP provides many mechanisms for expressing parallelism, but it primarily remains the programmer's responsibility to group computations to improve data locality. The loopchain abstraction, where data access patterns are included with the specification of parallel loops, provides compilers with sufficient information to automate the parallelism versus data locality tradeoff. In this paper, we present a loop chain pragma and an extension to the omp for to enable the specification of loop chains and high-level specifications of schedules on loop chains. We show example usage of the extensions, describe their implementation, and show preliminary performance results for some simple examples.

Original language	English (US)
Title of host publication	Proceedings of WACCPD 2016
Subtitle of host publication	3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	57-67
Number of pages	11
ISBN (Electronic)	9781509061525
DOIs	https://doi.org/10.1109/WACCPD.2016.010
State	Published - Jan 30 2017
Event	3rd Workshop on Accelerator Programming using Directives, WACCPD 2016 - Salt Lake City, United States Duration: Nov 14 2016 → …

Publication series

Name	Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	3rd Workshop on Accelerator Programming using Directives, WACCPD 2016
Country/Territory	United States
City	Salt Lake City
Period	11/14/16 → …

ASJC Scopus subject areas

Hardware and Architecture
Software
Control and Optimization

Access to Document

10.1109/WACCPD.2016.010

Cite this

Bertolacci, I. J., Strout, M. M., Guzik, S., Riley, J., & Olschanowsky, C. (2017). Identifying and scheduling loop chains using directives. In Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 57-67). Article 7836581 (Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WACCPD.2016.010

Identifying and scheduling loop chains using directives. / Bertolacci, Ian J.; Strout, Michelle Mills; Guzik, Stephen et al.
Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. p. 57-67 7836581 (Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis).

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

Bertolacci, IJ, Strout, MM, Guzik, S, Riley, J & Olschanowsky, C 2017, Identifying and scheduling loop chains using directives. in Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis., 7836581, Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 57-67, 3rd Workshop on Accelerator Programming using Directives, WACCPD 2016, Salt Lake City, United States, 11/14/16. https://doi.org/10.1109/WACCPD.2016.010

Bertolacci IJ, Strout MM, Guzik S, Riley J, Olschanowsky C. Identifying and scheduling loop chains using directives. In Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2017. p. 57-67. 7836581. (Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/WACCPD.2016.010

Bertolacci, Ian J. ; Strout, Michelle Mills ; Guzik, Stephen et al. / Identifying and scheduling loop chains using directives. Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 57-67 (Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{5cacb9db25b648659e7cfebd39a3c09b,

title = "Identifying and scheduling loop chains using directives",

abstract = "Exposing opportunities for parallelization while explicitly managing data locality is the primary challenge to porting and optimizing existing computational science simulation codes to improve performance and accuracy. OpenMP provides many mechanisms for expressing parallelism, but it primarily remains the programmer's responsibility to group computations to improve data locality. The loopchain abstraction, where data access patterns are included with the specification of parallel loops, provides compilers with sufficient information to automate the parallelism versus data locality tradeoff. In this paper, we present a loop chain pragma and an extension to the omp for to enable the specification of loop chains and high-level specifications of schedules on loop chains. We show example usage of the extensions, describe their implementation, and show preliminary performance results for some simple examples.",

author = "Bertolacci, {Ian J.} and Strout, {Michelle Mills} and Stephen Guzik and Jordan Riley and Catherine Olschanowsky",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 3rd Workshop on Accelerator Programming using Directives, WACCPD 2016 ; Conference date: 14-11-2016",

year = "2017",

month = jan,

day = "30",

doi = "10.1109/WACCPD.2016.010",

language = "English (US)",

series = "Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "57--67",

booktitle = "Proceedings of WACCPD 2016",

}

TY - GEN

T1 - Identifying and scheduling loop chains using directives

AU - Bertolacci, Ian J.

AU - Strout, Michelle Mills

AU - Guzik, Stephen

AU - Riley, Jordan

AU - Olschanowsky, Catherine

PY - 2017/1/30

Y1 - 2017/1/30

N2 - Exposing opportunities for parallelization while explicitly managing data locality is the primary challenge to porting and optimizing existing computational science simulation codes to improve performance and accuracy. OpenMP provides many mechanisms for expressing parallelism, but it primarily remains the programmer's responsibility to group computations to improve data locality. The loopchain abstraction, where data access patterns are included with the specification of parallel loops, provides compilers with sufficient information to automate the parallelism versus data locality tradeoff. In this paper, we present a loop chain pragma and an extension to the omp for to enable the specification of loop chains and high-level specifications of schedules on loop chains. We show example usage of the extensions, describe their implementation, and show preliminary performance results for some simple examples.

AB - Exposing opportunities for parallelization while explicitly managing data locality is the primary challenge to porting and optimizing existing computational science simulation codes to improve performance and accuracy. OpenMP provides many mechanisms for expressing parallelism, but it primarily remains the programmer's responsibility to group computations to improve data locality. The loopchain abstraction, where data access patterns are included with the specification of parallel loops, provides compilers with sufficient information to automate the parallelism versus data locality tradeoff. In this paper, we present a loop chain pragma and an extension to the omp for to enable the specification of loop chains and high-level specifications of schedules on loop chains. We show example usage of the extensions, describe their implementation, and show preliminary performance results for some simple examples.

UR - http://www.scopus.com/inward/record.url?scp=85015177221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015177221&partnerID=8YFLogxK

U2 - 10.1109/WACCPD.2016.010

DO - 10.1109/WACCPD.2016.010

M3 - Conference contribution

AN - SCOPUS:85015177221

T3 - Proceedings of WACCPD 2016: 3rd Workshop on Accelerator Programming using Directives - Held in conjunction with SC 2016: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 57

EP - 67

BT - Proceedings of WACCPD 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd Workshop on Accelerator Programming using Directives, WACCPD 2016

Y2 - 14 November 2016

ER -

Identifying and scheduling loop chains using directives

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this