research-article

Utilizing Dark Silicon to Save Energy with Computational Sprinting

Authors:

Arun Raghavan,

Laurel Emurian,

Lei Shao,

Marios Papaefthymiou,

Kevin Pipe,

Thomas Wenisch,

Milo MartinAuthors Info & Claims

IEEE Micro, Volume 33, Issue 5

Pages 20 - 28

https://doi.org/10.1109/MM.2013.76

Published: 01 September 2013 Publication History

Abstract

Computational sprinting activates dark silicon to improve responsiveness by briefly but intensely exceeding a system's sustainable power limit. This article focuses on the energy implications of sprinting. The authors observe that sprinting can save energy even while improving responsiveness by enabling execution in chip configurations that, though thermally unsustainable, improve energy efficiency. Surprisingly, this energy savings can translate to throughput improvements even for long-running computations. Repeatedly alternating between sprint and idle modes while maintaining sustainable average power can outperform steady-state computation at the platform's thermal limit.

Cited By

View all

Azhar MZouzoula STrancoso PBartolini ARietveld KSchuman CMoreira J(2023)ARADAProceedings of the 20th ACM International Conference on Computing Frontiers10.1145/3587135.3592207(63-72)Online publication date: 9-May-2023
https://dl.acm.org/doi/10.1145/3587135.3592207
Azhar MPericàs MStenström P(2019)SaCProceedings of the 48th International Conference on Parallel Processing10.1145/3337821.3337865(1-12)Online publication date: 5-Aug-2019
https://dl.acm.org/doi/10.1145/3337821.3337865
Cai HCao QJiang H(2019)A Renewable Energy Driven Approach for Computational SprintingIEEE Transactions on Parallel and Distributed Systems10.1109/TPDS.2018.289023030:7(1449-1463)Online publication date: 13-Jun-2019
https://dl.acm.org/doi/10.1109/TPDS.2018.2890230
Show More Cited By

Recommendations

A Landscape of the New Dark Silicon Design Regime

Because of the breakdown of Dennard scaling, the percentage of a silicon chip that can switch at full frequency drops exponentially with each process generation. This utilization wall forces designers to ensure that, at any point in time, large ...
Limits of Parallelism and Boosting in Dim Silicon

Supply-voltage scaling has stagnated in recent technology nodes, leading to so-called dark silicon. To increase overall chip multiprocessor (CMP) performance, it is necessary to improve the energy efficiency of individual tasks so that more tasks can be ...
Scaling the Energy Proportionality Wall with KnightShift

Measuring energy proportionality accurately and understanding the reasons for disproportionality are critical first steps in designing future energy-efficient servers. This article presents two metrics—linear deviation and proportionality gap—that let ...

Comments

Information & Contributors

Information

Published In

IEEE Micro Volume 33, Issue 5

September 2013

56 pages

ISSN:0272-1732

Issue’s Table of Contents

Publisher

IEEE Computer Society Press

Washington, DC, United States

Publication History

Published: 01 September 2013

Author Tags

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

16
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 12 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Azhar MZouzoula STrancoso PBartolini ARietveld KSchuman CMoreira J(2023)ARADAProceedings of the 20th ACM International Conference on Computing Frontiers10.1145/3587135.3592207(63-72)Online publication date: 9-May-2023
https://dl.acm.org/doi/10.1145/3587135.3592207
Azhar MPericàs MStenström P(2019)SaCProceedings of the 48th International Conference on Parallel Processing10.1145/3337821.3337865(1-12)Online publication date: 5-Aug-2019
https://dl.acm.org/doi/10.1145/3337821.3337865
Cai HCao QJiang H(2019)A Renewable Energy Driven Approach for Computational SprintingIEEE Transactions on Parallel and Distributed Systems10.1109/TPDS.2018.289023030:7(1449-1463)Online publication date: 13-Jun-2019
https://dl.acm.org/doi/10.1109/TPDS.2018.2890230
Antoniadis AHuang COtt S(2019)A Fully Polynomial-Time Approximation Scheme for Speed Scaling with a Sleep StateAlgorithmica10.1007/s00453-019-00596-381:9(3725-3745)Online publication date: 1-Sep-2019
https://dl.acm.org/doi/10.1007/s00453-019-00596-3
Skach MArora MTullsen DTang LMars J(2018)Virtual melting temperatureProceedings of the 45th Annual International Symposium on Computer Architecture10.1109/ISCA.2018.00013(15-28)Online publication date: 2-Jun-2018
https://dl.acm.org/doi/10.1109/ISCA.2018.00013
Zahedi SFan SFaw MCole ELee B(2017)Computational SprintingACM Transactions on Computer Systems10.1145/301442834:4(1-26)Online publication date: 9-Jan-2017
https://dl.acm.org/doi/10.1145/3014428
Fan SZahedi SLee B(2016)The Computational Sprinting GameACM SIGARCH Computer Architecture News10.1145/2980024.287238344:2(561-575)Online publication date: 25-Mar-2016
https://dl.acm.org/doi/10.1145/2980024.2872383
Adileh AEyerman SJaleel AEeckhout L(2016)Maximizing Heterogeneous Processor Performance Under Power ConstraintsACM Transactions on Architecture and Code Optimization10.1145/297673913:3(1-23)Online publication date: 17-Sep-2016
https://dl.acm.org/doi/10.1145/2976739
Fan SZahedi SLee B(2016)The Computational Sprinting GameACM SIGPLAN Notices10.1145/2954679.287238351:4(561-575)Online publication date: 25-Mar-2016
https://dl.acm.org/doi/10.1145/2954679.2872383
Fan SZahedi SLee BConte TZhou Y(2016)The Computational Sprinting GameProceedings of the Twenty-First International Conference on Architectural Support for Programming Languages and Operating Systems10.1145/2872362.2872383(561-575)Online publication date: 25-Mar-2016
https://dl.acm.org/doi/10.1145/2872362.2872383
Show More Cited By

View Options

View options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Abstract

Cited By

Recommendations

A Landscape of the New Dark Silicon Design Regime

Limits of Parallelism and Boosting in Dim Silicon

Scaling the Energy Proportionality Wall with KnightShift

Comments

Information

Published In

Publisher

Publication History

Author Tags

Qualifiers

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

View options

Get Access

Login options

Full Access

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations