ABSTRACT
We present and evaluate a surrogate model, based on hardware performance counter measurements, to estimate computer system power consumption. Power and energy are especially important in the design and operation of large data centers and of clusters used for scientific computing. Tradeoffs are made between performance and power consumption, this needs to be dynamic because activity varies over time. While it is possible to instrument systems for fine-grain power monitoring, such instrumentation is costly and not commonly available. Furthermore, the latency and sampling periods of hardware power monitors can be large compared to time scales at which workloads can change and dynamic power controls can operate. Given these limitations, we argue that surrogate models of the kind we present here can provide low-cost and accurate estimates of power consumption to drive on-line dynamic control mechanisms and for use in off-line tuning.
In this brief paper, we discuss a general approach to building system power estimation models based on hardware performance counters. Using this technique, we then present a model for an Intel Core i7 system that has an absolute estimation error of 5.32 percent (median) and acceptable data collection overheads on varying workloads, CPU power states (frequency and voltage), and number of active cores. Since this method is based on event sampling of hardware counters, one can make a tradeoff between estimation accuracy and data-collection overhead.
- }}D. Bailey, J. Barton, T. Lasinski, and H. Simon. The NAS parallel benchmarks. RNR-91-002, NASA Ames Research Center, August 1991.Google Scholar
- }}D. Bedard, M. Lim, R. Fowler, and A. Porterfield. Powermon: Fine-grained and integrated power monitoring for commodity computer systems. In IEEE SoutheastCon conference. IEEE, 2010.Google ScholarCross Ref
- }}K. Cameron, R. Ge, and X. Feng. High-performance, power-aware distributed computing for scientific applications. Computer, 38(11):40--47, 2005. Google ScholarDigital Library
- }}K. Cameron, R. Ge, and X. Feng. Designing Computational Clusters for Performance and Power. Advances in Computers, 69(2), May 2007.Google Scholar
- }}M. Curtis-Maury, A. Shah, F. Blagojevic, D. Nikolopoulos, B. de Supinski, and M. Schulz. Prediction models for multi-dimensional power-performance optimization on many cores. In PACT '08: Proceedings of the 17th international conference on Parallel architectures and compilation techniques, Toronto, Ontario, Canada, 2008. Google ScholarDigital Library
- }}N. R. Draper and H. Smith. Applied Regression Analysis. Wiley-Interscience, 1998.Google ScholarCross Ref
- }}X. Fan, W. Weber, and L. A. Barroso. Power provisioning for a warehouse-sized computer. In ISCA '07: Proceedings of the 34th annual international symposium on Computer architecture, San Diego, California, USA, 2007. Google ScholarDigital Library
- }}W. Feng, X. Feng, and R. Ge. Green Supercomputing Comes of Age. IT Professional, 10(1), January 2008. Google ScholarDigital Library
- }}Peter J Huber. Robust statistics, 1981.Google Scholar
- }}Intel. Intel turbo boost technology in intel core microarchitecture (Nehalem) based processors. 2008.Google Scholar
- }}B. Lee and D. Brooks. Accurate and efficient regression modeling for microarchitectural performance and power prediction. SIGOPS Oper. Syst. Rev., 40(5):185--194, 2006. Google ScholarDigital Library
- }}M. Y. Lim. Improving Power and Performance Efficiency in Parallel and Distributed Computing Systems. PhD thesis, North Carolina State University, 2009.Google Scholar
- }}M. Y. Lim, V. Freeh, and D. Lowenthal. Adaptive, transparent frequency and voltage scaling of communication phases in MPI programs. In SC '06: Proceedings of the 2006 ACM/IEEE conference on Supercomputing, Tampa, FL, USA, 2006. Google ScholarDigital Library
- }}The R Project for Statistical Computing. http://www.r-project.org/.Google Scholar
- }}Perfmon2: the hardware-based performance monitoring interface. http://perfmon2.sourceforge.net/.Google Scholar
- }}B. Rountree, D. Lownenthal, B. de Supinski, M. Schulz, V. Freeh, and T. Bletsch. Adagio: making DVS practical for complex HPC applications. In ICS '09: Proceedings of the 23rd international conference on Supercomputing, Yorktown Heights, NY, USA, 2009. Google ScholarDigital Library
- }}K. Singh, M. Bhadauria, and S. A. McKee. Real time power estimation and thread scheduling via performance counters. SIGARCH Comput. Archit. News, 37(2):46--55, 2009. Google ScholarDigital Library
- }}C. Spearman. The proof and measurement of association between two things. Amer. J. Psychol, 15, 1974.Google Scholar
- }}Z. Wang, C. McCarthy, X. Zhu, P. Ranganathan, and V. Talwar. Feedback control algorithms for power management of servers. Technical report, Hewlett-Packard Laboratories, 2008.Google Scholar
Index Terms
- SoftPower: fine-grain power estimations using performance counters
Recommendations
Decomposable and responsive power models for multicore processors using performance counters
ICS '10: Proceedings of the 24th ACM International Conference on SupercomputingPower modeling based on performance monitoring counters (PMCs) attracted the interest of researchers since it became a quick approach to understand and analyse power behavior on real systems. As a result, several power-aware policies use power models to ...
Managing power constraints in a single-core scenario through power tokens
Current microprocessors face constant thermal and power-related problems during their everyday use, usually solved by applying a power budget to the processor/core. Dynamic voltage and frequency scaling (DVFS) has been an effective technique that ...
GPUWattch: enabling energy optimizations in GPGPUs
ICSA '13General-purpose GPUs (GPGPUs) are becoming prevalent in mainstream computing, and performance per watt has emerged as a more crucial evaluation metric than peak performance. As such, GPU architects require robust tools that will enable them to quickly ...
Comments