Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware

Denes Ban, R. Ferenc, Istvan Siket, Akos Kiss

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

2 Citations (Scopus)

Abstract

Heterogeneous environments are becoming commonplace so it is increasingly important to understand how and where we could execute a given algorithm the most efficiently. In this paper we propose a methodology that uses both static source code metrics and dynamic execution time, power and energy measurements to build configuration prediction models. These models are trained on special benchmarks that have both sequential and parallel implementations and can be executed on various computing elements, e.g., on CPUs or GPUs. After they are built, however, they can be applied to a new system using only the system's static metrics which are much more easily computable than any dynamic measurement. We found that we could predict the optimal execution configuration fairly accurately using static information alone.

Original languageEnglish
Title of host publicationProceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages178-183
Number of pages6
Volume3
ISBN (Print)9781467379519
DOIs
Publication statusPublished - Dec 2 2015
Event14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015 - Helsinki, Finland
Duration: Aug 20 2015Aug 22 2015

Other

Other14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015
CountryFinland
CityHelsinki
Period8/20/158/22/15

Fingerprint

Energy efficiency
Hardware
Electric power measurement
Program processors
Graphics processing unit

Keywords

  • configuration selection
  • Green computing
  • heterogeneous architecture
  • performance optimization
  • power-aware execution

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Ban, D., Ferenc, R., Siket, I., & Kiss, A. (2015). Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware. In Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015 (Vol. 3, pp. 178-183). [7345645] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/Trustcom.2015.629

Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware. / Ban, Denes; Ferenc, R.; Siket, Istvan; Kiss, Akos.

Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015. Vol. 3 Institute of Electrical and Electronics Engineers Inc., 2015. p. 178-183 7345645.

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

Ban, D, Ferenc, R, Siket, I & Kiss, A 2015, Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware. in Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015. vol. 3, 7345645, Institute of Electrical and Electronics Engineers Inc., pp. 178-183, 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015, Helsinki, Finland, 8/20/15. https://doi.org/10.1109/Trustcom.2015.629
Ban D, Ferenc R, Siket I, Kiss A. Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware. In Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015. Vol. 3. Institute of Electrical and Electronics Engineers Inc. 2015. p. 178-183. 7345645 https://doi.org/10.1109/Trustcom.2015.629
Ban, Denes ; Ferenc, R. ; Siket, Istvan ; Kiss, Akos. / Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware. Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015. Vol. 3 Institute of Electrical and Electronics Engineers Inc., 2015. pp. 178-183
@inproceedings{946ae919f3c043d59e0ac723d53e40bb,
title = "Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware",
abstract = "Heterogeneous environments are becoming commonplace so it is increasingly important to understand how and where we could execute a given algorithm the most efficiently. In this paper we propose a methodology that uses both static source code metrics and dynamic execution time, power and energy measurements to build configuration prediction models. These models are trained on special benchmarks that have both sequential and parallel implementations and can be executed on various computing elements, e.g., on CPUs or GPUs. After they are built, however, they can be applied to a new system using only the system's static metrics which are much more easily computable than any dynamic measurement. We found that we could predict the optimal execution configuration fairly accurately using static information alone.",
keywords = "configuration selection, Green computing, heterogeneous architecture, performance optimization, power-aware execution",
author = "Denes Ban and R. Ferenc and Istvan Siket and Akos Kiss",
year = "2015",
month = "12",
day = "2",
doi = "10.1109/Trustcom.2015.629",
language = "English",
isbn = "9781467379519",
volume = "3",
pages = "178--183",
booktitle = "Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Prediction Models for Performance, Power, and Energy Efficiency of Software Executed on Heterogeneous Hardware

AU - Ban, Denes

AU - Ferenc, R.

AU - Siket, Istvan

AU - Kiss, Akos

PY - 2015/12/2

Y1 - 2015/12/2

N2 - Heterogeneous environments are becoming commonplace so it is increasingly important to understand how and where we could execute a given algorithm the most efficiently. In this paper we propose a methodology that uses both static source code metrics and dynamic execution time, power and energy measurements to build configuration prediction models. These models are trained on special benchmarks that have both sequential and parallel implementations and can be executed on various computing elements, e.g., on CPUs or GPUs. After they are built, however, they can be applied to a new system using only the system's static metrics which are much more easily computable than any dynamic measurement. We found that we could predict the optimal execution configuration fairly accurately using static information alone.

AB - Heterogeneous environments are becoming commonplace so it is increasingly important to understand how and where we could execute a given algorithm the most efficiently. In this paper we propose a methodology that uses both static source code metrics and dynamic execution time, power and energy measurements to build configuration prediction models. These models are trained on special benchmarks that have both sequential and parallel implementations and can be executed on various computing elements, e.g., on CPUs or GPUs. After they are built, however, they can be applied to a new system using only the system's static metrics which are much more easily computable than any dynamic measurement. We found that we could predict the optimal execution configuration fairly accurately using static information alone.

KW - configuration selection

KW - Green computing

KW - heterogeneous architecture

KW - performance optimization

KW - power-aware execution

UR - http://www.scopus.com/inward/record.url?scp=84969270817&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84969270817&partnerID=8YFLogxK

U2 - 10.1109/Trustcom.2015.629

DO - 10.1109/Trustcom.2015.629

M3 - Conference contribution

SN - 9781467379519

VL - 3

SP - 178

EP - 183

BT - Proceedings - 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -