Metrics for energy-aware software optimisation

Stephen I. Roberts; Steven A. Wright; Suhaib A. Fahmy; Stephen A. Jarvis

doi:10.1007/978-3-319-58667-0_22

Metrics for energy-aware software optimisation

Stephen I. Roberts^*, Steven A. Wright, Suhaib A. Fahmy, Stephen A. Jarvis

^*Corresponding author for this work

Engineering and Physical Sciences

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

3 Citations (Scopus)

Abstract

Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Etⁿ) family, are unsuitable for energyaware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Etⁿ metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.

Original language	English
Title of host publication	High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings
Editors	Julian M. Kunkel, Pavan Balaji, David Keyes, Rio Yokota
Publisher	Springer Verlag
Pages	413-430
Number of pages	18
ISBN (Print)	9783319586663
DOIs	https://doi.org/10.1007/978-3-319-58667-0_22
Publication status	Published - 2017
Event	32nd International Conference, ISC High Performance, 2017 - Frankfurt, Germany Duration: 18 Jun 2017 → 22 Jun 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10266 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	32nd International Conference, ISC High Performance, 2017
Country/Territory	Germany
City	Frankfurt
Period	18/06/17 → 22/06/17

Bibliographical note

Funding Information:
The authors would like to thank Thomas Ilsche and the Center of Information Services and High Performance Computing (ZIH) at TU Dresden. This research is funded in part by research grants from AWE, ATOS and Allinea. Professor Stephen Jarvis is an AWE William Penney Fellow.

Publisher Copyright:
© Springer International Publishing AG 2017.

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-319-58667-0_22

Cite this

Roberts, S. I., Wright, S. A., Fahmy, S. A., & Jarvis, S. A. (2017). Metrics for energy-aware software optimisation. In J. M. Kunkel, P. Balaji, D. Keyes, & R. Yokota (Eds.), High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings (pp. 413-430). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10266 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-58667-0_22

Roberts, Stephen I. ; Wright, Steven A. ; Fahmy, Suhaib A. et al. / Metrics for energy-aware software optimisation. High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings. editor / Julian M. Kunkel ; Pavan Balaji ; David Keyes ; Rio Yokota. Springer Verlag, 2017. pp. 413-430 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{868997451feb4080a432a28d0da720f8,

title = "Metrics for energy-aware software optimisation",

abstract = "Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Etn) family, are unsuitable for energyaware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Etn metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.",

author = "Roberts, {Stephen I.} and Wright, {Steven A.} and Fahmy, {Suhaib A.} and Jarvis, {Stephen A.}",

note = "Funding Information: The authors would like to thank Thomas Ilsche and the Center of Information Services and High Performance Computing (ZIH) at TU Dresden. This research is funded in part by research grants from AWE, ATOS and Allinea. Professor Stephen Jarvis is an AWE William Penney Fellow. Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 32nd International Conference, ISC High Performance, 2017 ; Conference date: 18-06-2017 Through 22-06-2017",

year = "2017",

doi = "10.1007/978-3-319-58667-0_22",

language = "English",

isbn = "9783319586663",

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

publisher = "Springer Verlag",

pages = "413--430",

editor = "Kunkel, {Julian M.} and Pavan Balaji and David Keyes and Rio Yokota",

booktitle = "High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings",

}

Roberts, SI, Wright, SA, Fahmy, SA & Jarvis, SA 2017, Metrics for energy-aware software optimisation. in JM Kunkel, P Balaji, D Keyes & R Yokota (eds), High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10266 LNCS, Springer Verlag, pp. 413-430, 32nd International Conference, ISC High Performance, 2017, Frankfurt, Germany, 18/06/17. https://doi.org/10.1007/978-3-319-58667-0_22

Metrics for energy-aware software optimisation. / Roberts, Stephen I.; Wright, Steven A.; Fahmy, Suhaib A. et al.
High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings. ed. / Julian M. Kunkel; Pavan Balaji; David Keyes; Rio Yokota. Springer Verlag, 2017. p. 413-430 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10266 LNCS).

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

TY - GEN

T1 - Metrics for energy-aware software optimisation

AU - Roberts, Stephen I.

AU - Wright, Steven A.

AU - Fahmy, Suhaib A.

AU - Jarvis, Stephen A.

N1 - Funding Information: The authors would like to thank Thomas Ilsche and the Center of Information Services and High Performance Computing (ZIH) at TU Dresden. This research is funded in part by research grants from AWE, ATOS and Allinea. Professor Stephen Jarvis is an AWE William Penney Fellow. Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Etn) family, are unsuitable for energyaware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Etn metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.

AB - Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Etn) family, are unsuitable for energyaware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Etn metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.

UR - http://www.scopus.com/inward/record.url?scp=85026266811&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-58667-0_22

DO - 10.1007/978-3-319-58667-0_22

M3 - Conference contribution

AN - SCOPUS:85026266811

SN - 9783319586663

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

SP - 413

EP - 430

BT - High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings

A2 - Kunkel, Julian M.

A2 - Balaji, Pavan

A2 - Keyes, David

A2 - Yokota, Rio

PB - Springer Verlag

T2 - 32nd International Conference, ISC High Performance, 2017

Y2 - 18 June 2017 through 22 June 2017

ER -

Roberts SI, Wright SA, Fahmy SA, Jarvis SA. Metrics for energy-aware software optimisation. In Kunkel JM, Balaji P, Keyes D, Yokota R, editors, High Performance Computing - 32nd International Conference, ISC High Performance 2017, Proceedings. Springer Verlag. 2017. p. 413-430. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-58667-0_22

Metrics for energy-aware software optimisation

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

UN SDGs

Access to Document

Fingerprint

Cite this