SPEC ACCEL: A standard application suite for measuring hardware accelerator performance

Guido Juckeland, William Brantley, Sunita Chandrasekaran, Barbara Chapman, Shuai Che, Mathew Colgrove, Huiyu Feng, Alexander Grund, Robert Henschel, Wen-Mei W Hwu, Huian Li, Matthias S. Müller, Wolfgang E. Nagel, Maxim Perminov, Pavel Shelepugin, Kevin Skadron, John Stratton, Alexey Titov, Ke Wang, Matthijs Van WaverenBrian Whitney, Sandra Wienke, Rengan Xu, Kalyan Kumaran

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

Abstract

Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail.

Original languageEnglish (US)
Title of host publicationHigh Performance Computing Systems
Subtitle of host publicationPerformance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers
EditorsSimon D. Hammond, Stephen A. Jarvis, Steven A. Wright
PublisherSpringer-Verlag
Pages46-67
Number of pages22
ISBN (Print)9783319172477
DOIs
StatePublished - Jan 1 2015
Event5th International Workshop on Performance Modeling, Benchmarking, and Simulation of High Performance Computing Systems, PMBS 2014 - New Orleans, United States
Duration: Nov 16 2014Nov 16 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8966
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other5th International Workshop on Performance Modeling, Benchmarking, and Simulation of High Performance Computing Systems, PMBS 2014
CountryUnited States
CityNew Orleans
Period11/16/1411/16/14

Fingerprint

Hardware Accelerator
Accelerator
Particle accelerators
Hardware
Benchmark
Performance Metrics
Viability
Power Consumption
High Performance
Electric power utilization
Metric
Standards
Evaluate
Vertex of a graph
Demonstrate

Keywords

  • Energy measurements
  • OpenACC
  • OpenCL
  • SPEC
  • SPEC ACCEL

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Juckeland, G., Brantley, W., Chandrasekaran, S., Chapman, B., Che, S., Colgrove, M., ... Kumaran, K. (2015). SPEC ACCEL: A standard application suite for measuring hardware accelerator performance. In S. D. Hammond, S. A. Jarvis, & S. A. Wright (Eds.), High Performance Computing Systems: Performance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers (pp. 46-67). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8966). Springer-Verlag. https://doi.org/10.1007/978-3-319-17248-4_3

SPEC ACCEL : A standard application suite for measuring hardware accelerator performance. / Juckeland, Guido; Brantley, William; Chandrasekaran, Sunita; Chapman, Barbara; Che, Shuai; Colgrove, Mathew; Feng, Huiyu; Grund, Alexander; Henschel, Robert; Hwu, Wen-Mei W; Li, Huian; Müller, Matthias S.; Nagel, Wolfgang E.; Perminov, Maxim; Shelepugin, Pavel; Skadron, Kevin; Stratton, John; Titov, Alexey; Wang, Ke; Van Waveren, Matthijs; Whitney, Brian; Wienke, Sandra; Xu, Rengan; Kumaran, Kalyan.

High Performance Computing Systems: Performance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers. ed. / Simon D. Hammond; Stephen A. Jarvis; Steven A. Wright. Springer-Verlag, 2015. p. 46-67 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8966).

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

Juckeland, G, Brantley, W, Chandrasekaran, S, Chapman, B, Che, S, Colgrove, M, Feng, H, Grund, A, Henschel, R, Hwu, W-MW, Li, H, Müller, MS, Nagel, WE, Perminov, M, Shelepugin, P, Skadron, K, Stratton, J, Titov, A, Wang, K, Van Waveren, M, Whitney, B, Wienke, S, Xu, R & Kumaran, K 2015, SPEC ACCEL: A standard application suite for measuring hardware accelerator performance. in SD Hammond, SA Jarvis & SA Wright (eds), High Performance Computing Systems: Performance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8966, Springer-Verlag, pp. 46-67, 5th International Workshop on Performance Modeling, Benchmarking, and Simulation of High Performance Computing Systems, PMBS 2014, New Orleans, United States, 11/16/14. https://doi.org/10.1007/978-3-319-17248-4_3
Juckeland G, Brantley W, Chandrasekaran S, Chapman B, Che S, Colgrove M et al. SPEC ACCEL: A standard application suite for measuring hardware accelerator performance. In Hammond SD, Jarvis SA, Wright SA, editors, High Performance Computing Systems: Performance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers. Springer-Verlag. 2015. p. 46-67. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-17248-4_3
Juckeland, Guido ; Brantley, William ; Chandrasekaran, Sunita ; Chapman, Barbara ; Che, Shuai ; Colgrove, Mathew ; Feng, Huiyu ; Grund, Alexander ; Henschel, Robert ; Hwu, Wen-Mei W ; Li, Huian ; Müller, Matthias S. ; Nagel, Wolfgang E. ; Perminov, Maxim ; Shelepugin, Pavel ; Skadron, Kevin ; Stratton, John ; Titov, Alexey ; Wang, Ke ; Van Waveren, Matthijs ; Whitney, Brian ; Wienke, Sandra ; Xu, Rengan ; Kumaran, Kalyan. / SPEC ACCEL : A standard application suite for measuring hardware accelerator performance. High Performance Computing Systems: Performance Modeling, Benchmarking, and Simulation - 5th International Workshop, PMBS 2014, Revised Selected Papers. editor / Simon D. Hammond ; Stephen A. Jarvis ; Steven A. Wright. Springer-Verlag, 2015. pp. 46-67 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{633ebb26ddc94a53b98c5f0a60eb643e,
title = "SPEC ACCEL: A standard application suite for measuring hardware accelerator performance",
abstract = "Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail.",
keywords = "Energy measurements, OpenACC, OpenCL, SPEC, SPEC ACCEL",
author = "Guido Juckeland and William Brantley and Sunita Chandrasekaran and Barbara Chapman and Shuai Che and Mathew Colgrove and Huiyu Feng and Alexander Grund and Robert Henschel and Hwu, {Wen-Mei W} and Huian Li and M{\"u}ller, {Matthias S.} and Nagel, {Wolfgang E.} and Maxim Perminov and Pavel Shelepugin and Kevin Skadron and John Stratton and Alexey Titov and Ke Wang and {Van Waveren}, Matthijs and Brian Whitney and Sandra Wienke and Rengan Xu and Kalyan Kumaran",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-319-17248-4_3",
language = "English (US)",
isbn = "9783319172477",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "46--67",
editor = "Hammond, {Simon D.} and Jarvis, {Stephen A.} and Wright, {Steven A.}",
booktitle = "High Performance Computing Systems",

}

TY - GEN

T1 - SPEC ACCEL

T2 - A standard application suite for measuring hardware accelerator performance

AU - Juckeland, Guido

AU - Brantley, William

AU - Chandrasekaran, Sunita

AU - Chapman, Barbara

AU - Che, Shuai

AU - Colgrove, Mathew

AU - Feng, Huiyu

AU - Grund, Alexander

AU - Henschel, Robert

AU - Hwu, Wen-Mei W

AU - Li, Huian

AU - Müller, Matthias S.

AU - Nagel, Wolfgang E.

AU - Perminov, Maxim

AU - Shelepugin, Pavel

AU - Skadron, Kevin

AU - Stratton, John

AU - Titov, Alexey

AU - Wang, Ke

AU - Van Waveren, Matthijs

AU - Whitney, Brian

AU - Wienke, Sandra

AU - Xu, Rengan

AU - Kumaran, Kalyan

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail.

AB - Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail.

KW - Energy measurements

KW - OpenACC

KW - OpenCL

KW - SPEC

KW - SPEC ACCEL

UR - http://www.scopus.com/inward/record.url?scp=84942519551&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942519551&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-17248-4_3

DO - 10.1007/978-3-319-17248-4_3

M3 - Conference contribution

AN - SCOPUS:84942519551

SN - 9783319172477

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

SP - 46

EP - 67

BT - High Performance Computing Systems

A2 - Hammond, Simon D.

A2 - Jarvis, Stephen A.

A2 - Wright, Steven A.

PB - Springer-Verlag

ER -