Abstract
Power capping is a method to save power consumption of servers by limiting performance of the servers. Although users frequently run applications on different virtual machines (VMs) for keeping their performance and having them isolated from the other applications, power capping may degrade performance of all the applications running on the server. We present fine-grained power capping by limiting performance of each application individually. For keeping performance defined in Quality of Service (QoS) requirements, it is important to estimate applications’ performance and power consumption after the fine-grained power capping is applied. We propose the estimation method of physical CPU usage when limiting virtual CPU usage of applications on VMs. On servers where multiple VMs run, VM’s usage of physical CPU is interrupted by the other VMs, and a hypervisor uses physical CPU to control VMs. These VMs’ and hypervisor’s behaviors make it difficult to estimate performance and power consumption by straightforward methods, such as linear regression and polynomial regression. The proposed method uses Piecewise Linear Regression to estimate physical CPU usage by assuming that VM’s access to physical CPU is not interrupted by the other VMs. Then we estimate how much physical CPU usage is reduced by the interruption. Because physical CPU usage is not stable soon after limiting CPU usage, the proposed method estimates a convergence value of CPU usage after many interruptions are repeated.
- Emad Aboelela. 2003. Network Simulation Experiments Manual. Elsevier, Burlington, MA. Google ScholarDigital Library
- Thomas Begin and Alexandre Brandwajn. 2016. Predicting the system performance by combining calibrated performance models of its components: A preliminary study. In Proceedings of the 7th ACM/SPEC on International Conference on Performance Engineering (ICPE’16). ACM, New York, NY, 95--100. Google ScholarDigital Library
- Arka A. Bhattacharya, David Culler, Aman Kansal, Sriram Govindan, and Sriram Sankar. 2012a. The need for speed and stability in data center power capping. In Proceedings of the 2012 International Green Computing Conference (IGCC’12). 1--10. Google ScholarDigital Library
- Arka A. Bhattacharya, David Culler, Aman Kansal, Sriram Govindan, and Sriram Sankar. 2012b. The need for speed and stability in data center power capping. In Proceedings of the 2012 International Green Computing Conference (IGCC’12). 1--10. Google ScholarDigital Library
- Van Bui, Boyana Norris, Kevin Huck, Lois Curfman McInnes, Li Li, Oscar Hernandez, and Barbara Chapman. 2008. A component infrastructure for performance and power modeling of parallel scientific applications. In Proceedings of the 2008 compFrame/HPC-GECO Workshop on Component Based High Performance (CBHPC’08). 6:1--6:11. Google ScholarDigital Library
- Dayle Chettiar, Arindam Das, and Olivia Das. 2014. Performance modeling of cloud-based web systems to estimate response time distribution. In Proceeedings of Workshop on Software Architectures for Adaptive Autonomous Systems. 41--46.Google Scholar
- Ryan Cochran, Can Hankendi, Ayse K. Coskun, and Sherief Reda. 2011. Pack 8 cap: Adaptive DVFS and thread packing under power caps. In Proceedings of the 44th Annual IEEE/ACM Int. Symposium on Microarchitecture. 175--185. Google ScholarDigital Library
- Maxime Colmant, Mascha Kurpicz, Pascal Felber, Loïc Huertas, Romain Rouvoy, and Anita Sobe. 2015. Process-level power estimation in VM-based systems. In Proceedings of the 10th European Conference on Computer Systems (EuroSys’15). 14:1--14:14. Google ScholarDigital Library
- Brian F. Cooper, Adam Silberstein, Erwin Tam, Raghu Ramakrishnan, and Russell Sears. 2010. Benchmarking cloud serving systems with YCSB. In Proceedings of the 1st ACM Symposium on Cloud Computing (SoCC’10). 143--154. Google ScholarDigital Library
- M. Dayarathna, Y. Wen, and R. Fan. 2016. Data center energy consumption modeling: A survey. IEEE Commun. Surv. Tutor. 18, 1 (2016), 732--794.Google ScholarDigital Library
- Tamás Élteto, Cécile Germain-Renaud, Pascal Bondon, and Michèle Sebag. 2010. Discovering piecewise linear models of grid workload. In Proceedings of the 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing. 474--484. Google ScholarDigital Library
- L. Fan, B. Gao, Z. Wang, W. An, and Y. Wang. 2016. A semi-automatic approach of transforming applications to be multi-tenancy enabled. IEEE Trans. Serv. Comput. 9, 2 (2016), 227--240. Google ScholarDigital Library
- Can Hankendi, Sherief Reda, and Ayse K. Coskun. 2013. vCap: Adaptive power capping for virtualized servers. In Proceedings of the 2013 International Symposium on Low Power Electronics and Design. 415--420. Google ScholarDigital Library
- T. Hastie, R. Tibshirani, and J. Friedman. 2009. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, Berlin.Google Scholar
- Ajay Kattepur and Manoj Nambiar. 2016. Performance modeling of multi-tiered web applications with varying service demands. Int. J. Netw. Comput. 6, 1 (2016), 64--86. 2185-2847 http://www.ijnc.org/index.php/ijnc/article/view/118Google ScholarCross Ref
- Charles Lefurgy, Xiaorui Wang, and Malcolm Ware. 2008. Power capping: A prelude to power shifting. Renew. Sust. Energy Rev. 11, 2 (2008), 461--478. Google ScholarDigital Library
- Y. Li, Y. Wang, B. Yin, and L. Guan. 2012. An online power metering model for cloud environment. Proceedings of IEEE 11th International Symposium on Network Computing and Applications (NCA’12). 175--180. Google ScholarDigital Library
- Angelo Marletta. 2015. cpulimit. Retrieved January 29, 2015 from https://github.com/opsengine/cpulimit.Google Scholar
- Paul Menage. 2004. cgroups. Retrieved December 27, 2016 from https://www.kernel.org/doc/Documentation/cgroup-v1/cgroups.txt. (2004).Google Scholar
- Ha Tuan Minh, Masaki Samejima, and Norihisa Komoda. 2014. A restraint usage estimation method of server resource for peak shaving in data center by piecewise linear regression. In Proceedings of the 3rd Asian Conference on Information Systems (ACIS’14). 41--47.Google Scholar
- C. Möbius, W. Dargie, and A. Schill. 2014. Power consumption estimation models for processors, virtual machines, and servers. IEEE Trans. Parallel Distrib. Syst. 25, 6 (June 2014), 1600--1614. Google ScholarDigital Library
- R. Nathuji, P. England, P. Sharma, and A. Singh. 2009. Feedback driven QoS-aware power budgeting for virtualized servers. In Proceedings of 4th International Workshop on Feedback Control Implementation and Design in Computing Systems and Networks (FeBID’09).Google Scholar
- Mohammed S. Obaidat and Noureddine A. Boudriga. 2010. Fundamentals of Performance Evaluation of Computer and Telecommunications Systems. Wiley-Interscience, New York, NY. Google ScholarDigital Library
- K. Salah and R. Boutaba. 2012. Estimating service response time for elastic cloud applications. In Proceedings of the 2012 IEEE 1st International Conference on Cloud Networking (CLOUDNET’12). 12--16.Google Scholar
- Masaki Samejima, Ha Tuan Minh, and Norihisa Komoda. 2014. Flexible peak shaving in data center by suppression of application resource usage. In Proceedings of 16th International Conference on Enterprise Information Systems (ICEIS’14). 355--360. Google ScholarDigital Library
- Pierluigi Siano. 2014. Demand response and smart grids—A survey. Renew. Sust. Energy Rev. 30 (2014), 461--478.Google ScholarCross Ref
- Rahul Singh, Prashant Shenoy, Maitreya Natu, Vaishali Sadaphal, and Harrick Vin. 2011. Predico: A system for what-if analysis in complex data center applications. In Proceedings of the 12th ACM/IFIP/USENIX International Conference on Middleware. 123--142. Google ScholarDigital Library
- Patrick Thibodeau. 2014. Data centers are the new polluters. In Computerworld. Retrieved December 27, 2016 from http://www.computerworld.com/article/2598562/data-center/data-centers-are-the-new-polluters.html.Google Scholar
- Yuan Tian, Chuang Lin, and Keqin Li. 2014. Managing performance and power consumption tradeoff for multiple heterogeneous servers in cloud computing. Cluster Comput. 17, 3 (2014), 943--955. Google ScholarDigital Library
- Daniel Versick, Ingolf Wassmann, and Djamshid Tavangarian. 2013. Power consumption estimation of CPU and peripheral components in virtual machines. SIGAPP Appl. Comput. Rev. 13, 3 (2013), 17--25. Google ScholarDigital Library
- X. Xu, K. Teramoto, A. Morales, and H. H. Huang. 2013. DUAL: Reliability-aware power management in data centers. Proceedings of 13th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing (CCGrid’13), 530--537.Google Scholar
- Yan Zhai, Xiao Zhang, Stephane Eranian, Lingjia Tang, and Jason Mars. 2014. HaPPy: Hyperthread-aware power profiling dynamically. In Proceedings of the 2014 USENIX Annual Technical Conference. 211--218. Google ScholarDigital Library
- Albert Y. Zomaya and Young Choon Lee. 2012. Energy Efficient Distributed Computing Systems. Wiley-IEEE Computer Society Press, New York, NY Google ScholarDigital Library
Index Terms
- Power and Performance Estimation for Fine-Grained Server Power Capping via Controlling Heterogeneous Applications
Recommendations
Power Optimization with Performance Assurance for Multi-tier Applications in Virtualized Data Centers
ICPPW '10: Proceedings of the 2010 39th International Conference on Parallel Processing WorkshopsModern data centers must provide performance assurance for complex system software such as multi-tier web applications. In addition, the power consumption of data centers needs to be minimized to reduce operating costs and avoid system overheating. ...
Power consumption evaluation of an MHD simulation with CPU power capping
CCGRID '14: Proceedings of the 14th IEEE/ACM International Symposium on Cluster, Cloud, and Grid ComputingRecently to achieve the Exa-flops next generation computer system, the power consumption becomes the important issue. On the other hand, the power consumption character of application program is not so considered now. In this study we examine the power ...
Power Capping of CPU-GPU Heterogeneous Systems using Power and Performance Models
SMARTGREENS 2015: Proceedings of the 4th International Conference on Smart Cities and Green ICT SystemsRecent high performance computing (HPC) systems and supercomputers are built under strict power budgets
and the limitation will be even severer. Thus power control is becoming more important, especially on the
systems with accelerators such as GPUs, ...
Comments