Advertisement

TUE, a New Energy-Efficiency Metric Applied at ORNL’s Jaguar

  • Michael K. Patterson
  • Stephen W. Poole
  • Chung-Hsing Hsu
  • Don Maxwell
  • William Tschudi
  • Henry Coles
  • David J. Martinez
  • Natalie Bates
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7905)

Abstract

The metric, Power Usage Effectiveness (PUE), has been successful in improving energy efficiency of data centers, but it is not perfect. One challenge is that PUE does not account for the power distribution and cooling losses inside IT equipment. This is particularly problematic in the HPC (high performance computing) space where system suppliers are moving cooling and power subsystems into or out of the cluster. This paper proposes two new metrics: ITUE (IT-power usage effectiveness), similar to PUE but “inside” the IT and TUE (total-power usage effectiveness), which combines the two for a total efficiency picture. We conclude with a demonstration of the method, and a case study of measurements at ORNL’s Jaguar system. TUE provides a ratio of total energy, (internal and external support energy uses) and the specific energy used in the HPC. TUE can also be a means for comparing HPC site to HPC site.

Keywords

HPC energy-efficiency metrics data center 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Malone, C., Belady, C.: Metrics to Characterize Data Center & IT Equipment Energy Use. In: Proceedings of 2006 Digital Power Forum, Richardson, TX (2006)Google Scholar
  2. 2.
    Uptime Institute, Uptime Institute 2007 Data Center Industry Survey (2007), http://www.uptimeinstitute.com
  3. 3.
    The Green Grid, White Paper #6 “Green Grid Data Center Power Effieicny Metrics: PUE and DCiE (December 2008), http://www.thegreengrid.org/~/media/WhitePapers/White_Paper_6_-_PUE_and_DCiE_Eff_Metrics_30_December_2008.pdf?lang=en
  4. 4.
    The Green Grid, White Paper #49-PUE: A Comprehensive Examination of the Metric (October 2, 2012), http://www.thegreengrid.org/en/Global/Content/white-papers/WP49-PUEAComprehensiveExaminationoftheMetric
  5. 5.
    EPA, DOE, TGG, ASHRAE, et.al. Recommendations for Measuring and Reporting Over-all Data Center Efficiency, (May 17, 2011), EPA Website: http://www.energystar.gov/ia/partners/prod_development/downloads/Data_Center_Metrics_Task_Force_Recommendations_V2.pdf?7438-21e8
  6. 6.
    Top500 [Internet]. Top 500 Supercomputing Sites (c2000-2012), http://top500.org/project/linpack (cited October 31, 2012)
  7. 7.
    Intel, Power Joint Engineering Team System Power Calculator (January 2013) (unpublished)Google Scholar
  8. 8.
    Bland, A.S., Joubert, W., Kendall, R.A., Kothe, D.B., Rogers, J.H., Shipman, G.M.: Jaguar: The world’s most powerful computer system – an update. Cray Users Group (May 2010)Google Scholar
  9. 9.
    Hennecke, M., Frings, W., Homberg, W., Zitz, A., Knobloch, M., Böttiger, H.: Measuring power consumption on IBM Blue Gene/P. In: International Conference on Energy-Aware High Performance Computing (September 2011)Google Scholar
  10. 10.
    Maeda, H., Kubo, H., Shimamori, H., Tamura, A., Wei, J.: System packaging technologies for the K computer. Fujitsu Scientific and Technical Journal 48(3), 286–294 (2012)Google Scholar
  11. 11.
    Rogers, J., Hoehn, B., Kelley, D.: Deploying large scale XT systems at ORNL. Cray Users Group (May 2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Michael K. Patterson
    • 1
    • 5
  • Stephen W. Poole
    • 2
    • 5
  • Chung-Hsing Hsu
    • 2
    • 5
  • Don Maxwell
    • 2
  • William Tschudi
    • 3
    • 5
  • Henry Coles
    • 3
    • 5
  • David J. Martinez
    • 4
    • 5
  • Natalie Bates
    • 5
  1. 1.Intel Architecture GroupIntel CorporationDupontUSA
  2. 2.Oak Ridge National LaboratoryOak RidgeUSA
  3. 3.Lawrence Berkeley National LaboratoryBerkeleyUSA
  4. 4.Sandia National LaboratoriesAlbuquerqueUSA
  5. 5.Energy Efficient HPC Working GroupAnderson IslandUSA

Personalised recommendations