Using Queueing Theory for Controlling the Number of Computing Servers

Jukka Kommeri, Mark Sevalnev, Samuli Aalto, Tapio Niemi

Abstract

We have tested how queueing theory can be applied to improve energy efficiency of scientific computing clusters. Our method calculates the number of required servers based on the arrival rate of computing jobs and turns on and off computing nodes based on this estimate. Our tests indicated that this method decreases energy consumption. However simultaneously the average lead time tends to increase because of higher waiting times in cases when the arrival intensity goes up.

References

  1. Barroso, L. A. and Hölzle, U. (2007). The case for energyproportional computing. Computer, 40:33-37.
  2. beam (2006). Lhc beam-beam studies. http://lhc-beambeam.web.cern.ch/lhc-beam-beam.
  3. Choi, K., Soma, R., and Pedram, M. (2004). Dynamic Voltage and Frequency Scaling based on Workload Decomposition. In Int. Symp on Low Power Electronics and Design.
  4. Crovella, M. and Bestavros, A. (1995). Explaining world wide web traffic self-similarity.
  5. Elnozahy, E. M., Kistler, M., and Rajamony, R. (2002). Energy-efficient server clusters. In In Proceedings of the 2nd Workshop on Power-Aware Computing Systems, pages 179-196.
  6. Fan, X., Weber, W.-D., and Barroso, L. A. (2007). Power provisioning for a warehouse-sized computer. SIGARCH Comput. Archit. News, 35:13-23.
  7. Gandhi, A., Harchol-Balter, M., Das, R., and Lefurgy, C. (2009). Optimal power allocation in server farms. pages 157-168. ACM.
  8. Govindan, S., Sivasubramaniam, A., and Urgaonkar, B. (2011). Benefits and limitations of tapping into stored energy for datacenters. In ISCA, pages 341-352.
  9. Herr, W. and Zorzano, M. P. (2001). Coherent dipole modes for multiple interaction regions. Technical report, LHC Project Report 461.
  10. Horvath, T. and Skadron, K. (2008). Multi-mode energy management for multi-tier server clusters. In Proceedings of the 17th international conference on Parallel architectures and compilation techniques, PACT 7808, pages 270-279, New York, NY, USA. ACM.
  11. Kaxiras, S. and Martonosi, M. (2008). Computer Architecture Techniques for Power-Efficiency. Morgan and Claypool Publishers, 1st edition.
  12. Kleinrock, L. (1976). Queueing Systems, volume II: Computer Applications. Wiley Interscience. (Published in Russian, 1979. Published in Japanese, 1979.).
  13. Lin, M., Wierman, A., Andrew, L. L. H., and Thereska, E. (2011). Dynamic right-sizing for power-proportional data centers. In INFOCOM, pages 1098-1106. IEEE.
  14. Liu, Z., Lin, M., Wierman, A., Low, S. H., and Andrew, L. L. H. (2011). Greening geographical load balancing. In SIGMETRICS, pages 233-244.
  15. Meisner, D., Gold, B. T., and Wenisch, T. F. (2009). Powernap: eliminating server idle power. SIGPLAN Not., 44:205-216.
  16. Miyoshi, A., Lefurgy, C., Hensbergen, E. V., Rajamony, R., and Rajkumar, R. (2002). Critical power slope: Understanding the runtime effects of frequency scaling. In In Proceedings of the 16th Annual ACM International Conference on Supercomputing, pages 35-44.
  17. Nedevschi, S., Popa, L., Iannaccone, G., Ratnasamy, S., and Wetherall, D. (2008). Reducing network energy consumption via sleeping and rate-adaptation. In Proceedings of the 5th USENIX Symposium on Networked Systems Design and Implementation, NSDI'08, pages 323-336, Berkeley, CA, USA. USENIX Association.
  18. Pakbaznia, E. and Pedram, M. (2009). Minimizing data center cooling and server power costs. In Proceedings of the 14th ACM/IEEE international symposium on Low power electronics and design, ISLPED 7809, pages 145-150, New York, NY, USA. ACM.
  19. Paxson, V. and Floyd, S. (1995). Wide-area traffic: The failure of poisson modeling. IEEE/ACM Transactions on Networking, pages 226-244.
  20. Rao, L., Liu, X., Xie, L., and Liu, W. (2010). Minimizing electricity cost: Optimization of distributed internet data centers in a multi-electricity-market environment. In INFOCOM, pages 1145-1153.
  21. Wang, Z., Zhu, X., McCarthy, C., Ranganathan, P., and Talwar, V. (2008). Feedback control algorithms for power management of servers. In Third International Workshop on Feedback Control Implementation and Design in Computing Systems and Networks (FeBid), Annapolis.
  22. Wendell, P., Jiang, J. W., Freedman, M. J., and Rexford, J. (2010). Donar: decentralized server selection for cloud services. In SIGCOMM, pages 231-242.
  23. Wu, Q., Juang, P., Martonosi, M., Peh, L.-S., and Clark, D. W. (2005). Formal control techniques for powerperformance management. IEEE Micro, 25(5):52-62.
Download


Paper Citation


in Harvard Style

Niemi T., Aalto S., Sevalnev M. and Kommeri J. (2012). Using Queueing Theory for Controlling the Number of Computing Servers . In Proceedings of the Sixth International Symposium on e-Health Services and Technologies and the Third International Conference on Green IT Solutions - Volume 1: ICGREEN, ISBN 978-989-8565-27-3, pages 83-88. DOI: 10.5220/0004474900830088


in Bibtex Style

@conference{icgreen12,
author={Tapio Niemi and Samuli Aalto and Mark Sevalnev and Jukka Kommeri},
title={Using Queueing Theory for Controlling the Number of Computing Servers},
booktitle={Proceedings of the Sixth International Symposium on e-Health Services and Technologies and the Third International Conference on Green IT Solutions - Volume 1: ICGREEN,},
year={2012},
pages={83-88},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004474900830088},
isbn={978-989-8565-27-3},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Sixth International Symposium on e-Health Services and Technologies and the Third International Conference on Green IT Solutions - Volume 1: ICGREEN,
TI - Using Queueing Theory for Controlling the Number of Computing Servers
SN - 978-989-8565-27-3
AU - Niemi T.
AU - Aalto S.
AU - Sevalnev M.
AU - Kommeri J.
PY - 2012
SP - 83
EP - 88
DO - 10.5220/0004474900830088