research-article

On-line analysis of hardware performance events for workload characterization and processor frequency scaling decisions

Authors:

Robert Schöne,

Daniel HackenbergAuthors Info & Claims

ICPE '11: Proceedings of the 2nd ACM/SPEC International Conference on Performance engineering

Pages 481 - 486

https://doi.org/10.1145/1958746.1958819

Published: 14 March 2011 Publication History

Get Access

Abstract

Energy efficiency optimizations of computational resources continue to be of growing importance for both classical datacenter workloads as well as high performance computing environments. New hardware generations introduce more and more energy efficiency features, resulting in a power consumption variation by at least a factor of four between idle and full load. Even the power consumption of different full-load workloads can vary substantially, clearly showing that there is energy saving potential apart from the traditional "race to idle". In this paper we present a configurable CPU frequency governor that adapts processor frequencies based on performance counter measurements instead of processor load. We use the SPEC OMP benchmark suite to determine the potential of our approach and present governor configurations for two up-to-date x86_64 microarchitectures. Moreover we show that substantial follow-up work is required to assess further efficiency optimization potential in this field.

References

[1]

V. Aslot, M. Domeika, R. Eigenmann, G. Gaertner, W. B. Jones, and B. Parady. Specomp: A new benchmark suite for measuring parallel computer performance. In In Workshop on OpenMP Applications and Tools, pages 1--10, 2001.

Digital Library

Google Scholar

[2]

H. Brunst, D. Hackenberg, G. Juckeland, and H. Rohling. Comprehensive Performance Tracking with Vampir 7. In M. S. Müller, M. M. Resch, A. Schulz, and W. E. Nagel, editors, Tools for High Performance Computing 2009, pages 17--29. Springer Berlin Heidelberg, 2010.

Crossref

Google Scholar

[3]

V. W. Freeh, F. Pan, D. K. Lowenthal, N. Kappiah, R. Springer, B. L. Rountree, and M. E. Femal. Analyzing the energy-time tradeoff in high-performance computing applications. IEEE Transactions on Parallel and Distributed Systems, 18(6):835--848, June 2007.

Digital Library

Google Scholar

[4]

D. Hackenberg, R. Schöne, D. Molka, M. S. Müller, and A. Knüpfer. Quantifying power consumption variations of hpc systems using spec mpi benchmarks. Computer Science - Research and Development, 25:155--163, 2010. 10.1007/s00450-010-0118-0.

Crossref

Google Scholar

[5]

V. Keller and R. Gruber. One joule per gflop for blas2 now! In ICNAAM 2010, pages 1321--1324, 2010.

Google Scholar

[6]

R. Kotla, S. Ghiasi, T. Keller, and F. Rawson. Scheduling processor voltage and frequency in server and cluster systems. In Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 11 - Volume 12, IPDPS '05, pages 234.2--, Washington, DC, USA, 2005. IEEE Computer Society.

Digital Library

Google Scholar

[7]

D. Li, D. S. Nikolopoulos, K. Cameron, B. R. de Supinski, and M. Schulz. Hybrid mpi/openmp power-aware computing. In IPDPS 2010, 2010.

Crossref

Google Scholar

[8]

V. Pallipadi and A. Starikovskiy. The ondemand governor: past, present and future. In Proceedings of Linux Symposium, vol. 2, pp. 223--238, 2006.

Google Scholar

[9]

R. Schoene, R. Tschueter, D. Hackenberg, and T. Ilsche. The VampirTrace Plugin Counter Interface: Introduction and Examples. In Proceedings of the EuroPar 2010 - Workshops (accepted), 2010.

Digital Library

Google Scholar

Cited By

View all

Fujita KNoguchi HFukuda HKaneko M(2023)Dynamic uncore frequency scaling based on performance events of Web applicationsProceedings of the 2023 8th International Conference on Cloud Computing and Internet of Things10.1145/3627345.3627350(31-37)Online publication date: 22-Sep-2023
https://dl.acm.org/doi/10.1145/3627345.3627350
Wang BKozhokanova ATerboven CMueller M(2023)RLP: Power Management Based on a Latency-Aware Roofline Model2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)10.1109/IPDPS54959.2023.00052(446-456)Online publication date: May-2023
https://doi.org/10.1109/IPDPS54959.2023.00052
Chinnakkonda DRajamani KSrinivas M(2022)Architecture slack exploitation for phase classification and performance estimation in server-class processorsJournal of Parallel and Distributed Computing10.1016/j.jpdc.2022.06.017169(157-170)Online publication date: Nov-2022
https://doi.org/10.1016/j.jpdc.2022.06.017
Show More Cited By

Index Terms

On-line analysis of hardware performance events for workload characterization and processor frequency scaling decisions

Recommendations

Workload-Aware CPU Performance Scaling for Transactional Database Systems
SIGMOD '18: Proceedings of the 2018 International Conference on Management of Data

Natural short term fluctuations in the load of transactional data systems present an opportunity for power savings. For example, a system handling 1000 requests per second on average can expect more than 1000 requests in some seconds, fewer in others. ...
Diverge-Merge Processor: Generalized and Energy-Efficient Dynamic Predication

The branch misprediction penalty is a major performance limiter and a major cause of wasted energy in high-performance processors. The diverge-merge processor reduces this penalty by dynamically predicating a wide range of hard-to-predict branches at ...
Memory power management via dynamic voltage/frequency scaling
ICAC '11: Proceedings of the 8th ACM international conference on Autonomic computing

Energy efficiency and energy-proportional computing have become a central focus in enterprise server architecture. As thermal and electrical constraints limit system power, and datacenter operators become more conscious of energy costs, energy ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

ICPE '11: Proceedings of the 2nd ACM/SPEC International Conference on Performance engineering

March 2011

470 pages

ISBN:9781450305198

DOI:10.1145/1958746

General Chairs:
Samuel Kounev
Karlsruhe Institute of Technology, Germany
,
Vittorio Cortellessa
Università dell'Aquila, Italy
,
Program Chairs:
Raffaela Mirandola
Politecnico di Milano, Italy
,
David Lilja
University of Minnesota, USA
,
Kai Sachs
SAP AG

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 14 March 2011

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Conference

ICPE'11

Sponsor:

ICPE'11: Second Joint WOSP/SIPEW International Conference on Performance Engineering

March 14 - 16, 2011

Karlsruhe, Germany

Acceptance Rates

Overall Acceptance Rate 252 of 851 submissions, 30%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

18
Total Citations
View Citations
339
Total Downloads

Downloads (Last 12 months)3
Downloads (Last 6 weeks)1

Reflects downloads up to 18 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Fujita KNoguchi HFukuda HKaneko M(2023)Dynamic uncore frequency scaling based on performance events of Web applicationsProceedings of the 2023 8th International Conference on Cloud Computing and Internet of Things10.1145/3627345.3627350(31-37)Online publication date: 22-Sep-2023
https://dl.acm.org/doi/10.1145/3627345.3627350
Wang BKozhokanova ATerboven CMueller M(2023)RLP: Power Management Based on a Latency-Aware Roofline Model2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)10.1109/IPDPS54959.2023.00052(446-456)Online publication date: May-2023
https://doi.org/10.1109/IPDPS54959.2023.00052
Chinnakkonda DRajamani KSrinivas M(2022)Architecture slack exploitation for phase classification and performance estimation in server-class processorsJournal of Parallel and Distributed Computing10.1016/j.jpdc.2022.06.017169(157-170)Online publication date: Nov-2022
https://doi.org/10.1016/j.jpdc.2022.06.017
Choi JPark GNam D(2019)Interference-aware co-scheduling method based on classification of application characteristics from hardware performance counter using data miningCluster Computing10.1007/s10586-019-02949-7Online publication date: 12-Jun-2019
https://doi.org/10.1007/s10586-019-02949-7
Wang BTerboven CMueller M(2018)Performance Prediction under Power Capping2018 International Conference on High Performance Computing & Simulation (HPCS)10.1109/HPCS.2018.00059(308-313)Online publication date: Jul-2018
https://doi.org/10.1109/HPCS.2018.00059
Wang BSchmidl DTerboven CMüller M(2017)Dynamic Application-aware Power CappingProceedings of the 5th International Workshop on Energy Efficient Supercomputing10.1145/3149412.3149413(1-8)Online publication date: 12-Nov-2017
https://dl.acm.org/doi/10.1145/3149412.3149413
Sakamoto RCao TKondo MInoue KUeda MPatki TEllsworth DRountree BSchulz M(2017)Production Hardware Overprovisioning: Real-World Performance Optimization Using an Extensible Power-Aware Resource Management Framework2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS)10.1109/IPDPS.2017.107(957-966)Online publication date: May-2017
https://doi.org/10.1109/IPDPS.2017.107
Georgiou YGlesser DTrystram D(2015)Adaptive Resource and Job Management for Limited Power ConsumptionProceedings of the 2015 IEEE International Parallel and Distributed Processing Symposium Workshop10.1109/IPDPSW.2015.118(863-870)Online publication date: 25-May-2015
https://dl.acm.org/doi/10.1109/IPDPSW.2015.118
Schöne RMolka DWerner M(2015)Wake-up latencies for processor idle states on current x86 processorsComputer Science - Research and Development10.1007/s00450-014-0270-z30:2(219-227)Online publication date: 1-May-2015
https://dl.acm.org/doi/10.1007/s00450-014-0270-z
Wang BSchmidl DMüller M(2015)Evaluating the Energy Consumption of OpenMP Applications on Haswell ProcessorsOpenMP: Heterogenous Execution and Data Movements10.1007/978-3-319-24595-9_17(233-246)Online publication date: 26-Nov-2015
https://doi.org/10.1007/978-3-319-24595-9_17
Show More Cited By

View Options

Login options

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

Cited By

Index Terms

Recommendations

Workload-Aware CPU Performance Scaling for Transactional Database Systems

Diverge-Merge Processor: Generalized and Energy-Efficient Dynamic Predication

Memory power management via dynamic voltage/frequency scaling