Power Consumption Modeling and Prediction in a Hybrid CPU-GPU-MIC Supercomputer

  • Alina SîrbuEmail author
  • Ozalp Babaoglu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9833)


Power consumption is a major obstacle for High Performance Computing (HPC) systems in their quest towards the holy grail of ExaFLOP performance. Significant advances in power efficiency have to be made before this goal can be attained and accurate modeling is an essential step towards power efficiency by optimizing system operating parameters to match dynamic energy needs. In this paper we present a study of power consumption by jobs in Eurora, a hybrid CPU-GPU-MIC system installed at the largest Italian data center. Using data from a dedicated monitoring framework, we build a data-driven model of power consumption for each user in the system and use it to predict the power requirements of future jobs. We are able to achieve good prediction results for over 80 % of the users in the system. For the remaining users, we identify possible reasons why prediction performance is not as good. Possible applications for our predictive modeling results include scheduling optimization, power-aware billing and system-scale power modeling. All the scripts used for the study have been made available on GitHub.


Job power modeling Job power prediction High performance computing Hybrid system Support vector regression 



BigQuery analysis was carried out through a generous Cloud Credits grant from Google. We are grateful to Prof. L. Benini and Dr. A. Bartolini for useful discussions regarding the data and to the HPC group at CINECA, in particular Dr. E. Rossi and Dr. C. Cavazzoni for providing access to the CINECA systems. We acknowledge the CINECA ISCRA PACNA and PM-HPC awards allowing access to HPC resources and support. This work was partially funded by the European project SoBigData Research Infrastructure — Big Data and Social Mining Ecosystem under the INFRAIA-H2020 program (grant agreement 654024).


  1. 1.
    Bartolini, A., et al.: Unveiling eurora-thermal and power characterization of the most energy-efficient supercomputer in the world. In: DATE 2014 (2014)Google Scholar
  2. 2.
    Borghesi, A., et al.: MS3: a Mediterranean-stile job scheduler for supercomputers-do less when it’s too hot! In: HPCS 2015, pp. 88–95 (2015)Google Scholar
  3. 3.
    Borghesi, A., Collina, F., Lombardi, M., Milano, M., Benini, L.: Power capping in high performance computing systems. In: Pesant, G. (ed.) CP 2015. LNCS, vol. 9255, pp. 524–540. Springer, Heidelberg (2015)Google Scholar
  4. 4.
    C. Storlie, C., et al.: Modeling and predicting power consumption of high performance computing jobs. arXiv preprint arXiv:14125247 (2014)
  5. 5.
    Cavazzoni, C.: Eurora: a european architecture toward exascale. In: Future HPC Systems: the Challenges of Power-Constrained Performance. ACM (2012)Google Scholar
  6. 6.
    Sîrbu, A., Babaoglu, O.: BigQuery and Python scripts. Github (2016).
  7. 7.
    Dargie, W.: A stochastic model for estimating the power consumption of a processor. IEEE Trans. Comput. 64(5), 1311–1322 (2015)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Fraternali, F., et al.: Quantifying the impact of variability on the energy efficiency for a next-generation ultra-green supercomputer. In: ISLPED 2014, pp. 295–298 (2014)Google Scholar
  9. 9.
    Pedregosa, F., et al.: Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)MathSciNetzbMATHGoogle Scholar
  10. 10.
    Gao, J.: Machine learning applications for data center optimisation. Google White Paper (2014)Google Scholar
  11. 11.
    Nagasaka, H., et al.: Statistical power modeling of GPU kernels using performance counters. In: IGCC 2010, pp. 115–122 (2010)Google Scholar
  12. 12.
    McCullough, J.C., et al.: Evaluating the effectiveness of model-based power characterization. In: USENIX ATC 2011, vol. 20 (2011)Google Scholar
  13. 13.
    Witkowski, M., et al.: Practical power consumption estimation for real life HPC applications. Future Gener. Comput. Syst. 29(1), 208–217 (2013)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Gschwandtner, P., et al.: Modeling CPU energy consumption of HPC applications on the IBM Power7. In: PDP 2014, pp. 536–543 (2014)Google Scholar
  15. 15.
    Shoukourian, H., Wilde, T.: Predicting the energy and power consumption of strong and weak scaling HPC applications. Supercomp Front Innov. 1(2), 20–41 (2014)Google Scholar
  16. 16.
    Smola, A., Vapnik, V.: Support vector regression machines. Adv. Neural Inf. Process. Syst. 9, 155–161 (1997)Google Scholar
  17. 17.
    Tigani, J., Naidu, S.: Google BigQuery Analytics. Wiley, Hoboken (2014)Google Scholar
  18. 18.
    Ma, X., et al.: Statistical power consumption analysis and modeling for GPU-based computing. In: ACM SOSP HotPower 2009 (2009)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversity of PisaPisaItaly
  2. 2.Department of Computer Science and EngineeringUniversity of BolognaBolognaItaly

Personalised recommendations