Abstract
Energy efficiency and power capping are critical concerns in server and cloud computing systems. They face growing challenges due to dynamic power variations from new client-directed web applications, as well as complex behaviors due to multicore resource sharing and hardware heterogeneity. This paper presents a new operating system facility called "power containers" that accounts for and controls the power and energy usage of individual fine-grained requests in multicore servers. This facility relies on three key techniques---1) online model that attributes multicore power (including shared maintenance power) to concurrently running tasks, 2) alignment of actual power measurements and model estimates to enable online model recalibration, and 3) on-the-fly application-transparent request tracking in multi-stage servers to isolate the power and energy contributions and customize per-request control. Our mechanisms enable new multicore server management capabilities including fair power capping that only penalizes power-hungry requests, and energy-aware request distribution between heterogeneous servers. Our evaluation uses three multicore processors (Intel Woodcrest, Westmere, and SandyBridge) and a variety of server and cloud computing (Google App Engine) workloads. Our results demonstrate the high accuracy of our request power accounting (no more than 11% errors) and the effectiveness of container-enabled power virus isolation and throttling. Our request distribution case study shows up to 25% energy saving compared to an alternative approach that recognizes machine heterogeneity but not fine-grained workload affinity.
- Intel Core2 Duo and Dual-Core thermal and mechanical design guidelines. http://www.intel.com/design/core2duo/documentation.htm.Google Scholar
- Apache Solr search server. http://lucene.apache.org/solr/.Google Scholar
- Stressful application test. http://code.google.com/p/stressapptest.Google Scholar
- Vosao content management system. http://www.vosao.org.Google Scholar
- Wikipedia data dumps. http://dumps.wikimedia.org/enwiki/.Google Scholar
- G. Banga, P. Druschel, and J. Mogul. Resource containers: A new facility for resource management in server systems. In Third USENIX Symp. on Operating Systems Design and Implementation (OSDI), pages 45--58, New Orleans, LA, Feb. 1999. Google Scholar
Digital Library
- P. Barham, A. Donnelly, R. Isaacs, and R. Mortier. Using Magpie for request extraction and workload modeling. In 6th USENIX Symp. on Operating Systems Design and Implementation (OSDI), pages 259--272, San Francisco, CA, Dec. 2004. Google Scholar
Digital Library
- le(2007)}barroso2007computerL. Barroso and U. Hölzle. The case for energy-proportional computing. IEEE Computer, 40 (12): 33--37, Dec. 2007. Google Scholar
Digital Library
- F. Bellosa. The benefits of event-driven energy accounting in power-sensitive systems. In ACM SIGOPS European Workshop, pages 37--42, Kolding, Denmark, Sept. 2000. Google Scholar
Digital Library
- R. Bertran, M. Gonzalez, X. Martorell, N. Navarro, and E. Ayguade. Decomposable and responsive power models for multicore processors using performance counters. In 24th ACM Int'l Conf. on Supercomputing (ICS), pages 147--158, Tsukuba, Japan, June 2010. Google Scholar
Digital Library
- A. Chanda, A. Cox, and W. Zwaenepoel. Whodunit: Transactional profiling for multi-tier applications. In Second EuroSys Conf., pages 17--30, Lisbon, Portugal, Mar. 2007. Google Scholar
Digital Library
- F. Chang, K. I. Farkas, and P. Ranganathan. Energy-driven statistical sampling: Detecting software hotspots. In Second Workshop on Power-Aware Computer Systems, pages 110--129, Cambridge, MA, Feb. 2002. Google Scholar
Digital Library
- J. S. Chase, D. C. Anderson, P. N. Thakar, A. M. Vahdat, and R. P. Doyle. Managing energy and server resources in hosting centers. In 18th ACM Symp. on Operating Systems Principles (SOSP), pages 103--116, Banff, Canada, Oct. 2001. Google Scholar
Digital Library
- B.-G. Chun, G. Iannaccone, G. Iannaccone, R. Katz, G. Lee, and L. Niccolini. An energy case for hybrid datacenters. In Workshop on Power Aware Computing and Systems, Big Sky, MT, Oct. 2009.Google Scholar
- X. Fan, W.-D. Weber, and L. Barroso. Power provisioning for a warehouse-sized computer. In 34th Int'l Symp. on Computer Architecture (ISCA), pages 13--23, San Diego, CA, June 2007. Google Scholar
Digital Library
- J. Flinn and M. Satyanarayanan. Energy-aware adaptation for mobile applications. In 17th ACM Symp. on Operating Systems Principles (SOSP), pages 48--63, Kiawah Island, SC, Dec. 2001. Google Scholar
Digital Library
- R. Fonseca, G. Porter, R. H. Katz, S. Shenker, and I. Stoica. X-Trace: A pervasive network tracing framework. In 4th USENIX Symp. on Networked Systems Design and Implementation (NSDI), Cambridge, MA, Apr. 2007. Google Scholar
Digital Library
- R. Fonseca, P. Dutta, P. Levis, and I. Stoica. Quanto: Tracking energy in networked embedded systems. In 8th USENIX Symp. on Operating Systems Design and Implementation (OSDI), pages 323--338, San Diego, CA, Dec. 2008. Google Scholar
Digital Library
- K. Ganesan, J. Jo, W. L. Bircher, D. Kaseridis, Z. Yu, and L. K. John. System-level max power (SYMPO): a systematic approach for escalating system-level power consumption using synthetic benchmarks. In 19th Int'l Conf. on Parallel Architecture and Compilation Techniques (PACT), pages 19--28, Vienna, Austria, Sept. 2010. Google Scholar
Digital Library
- S. Govindan, J. Choi, B. Urgaonkar, A. Sivasubramaniam, and A. Baldini. Statistical profiling-based techniques for effective power provisioning in data centers. In 4th EuroSys Conf., pages 317--330, Nuremberg, Germany, Apr. 2009. Google Scholar
Digital Library
- V. Gupta, P. Brett, D. Koufaty, D. Reddy, S. Hahn, K. Schwan, and G. Srinivasa. The forgotten "uncore": On the energy-efficiency of heterogeneous cores. In USENIX Annual Technical Conf., Boston, MA, June 2012. Google Scholar
Digital Library
- J. Hamilton. Where does the power go in high-scale data centers? Keynote speech at SIGMETRICS, June 2009.Google Scholar
- T. Heath, B. Diniz, E. V. Carrera, W. Meira Jr., and R. Bianchini. Energy conservation in heterogeneous server clusters. In 10th ACM Symp. on Principles and Practice of Parallel Programming (PPoPP), pages 186--195, Chicago, IL, June 2005. Google Scholar
Digital Library
- W. Huang, C. Lefurgy, W. Kuk, A. Buyuktosunoglu, M. Floyd, K. Rajamani, M. Allen-Ware, and B. Brock. Accurate fine-grained processor power proxies. In 45th Int'l Symp. on Microarchitecture (MICRO), pages 224--234, Vancouver, Canada, Dec. 2012. Google Scholar
Digital Library
- C. Isci and M. Martonosi. Phase characterization for power: Evaluating control-flow-based and event-counter-based techniques. In 12th Int'l Symp. on High-Performance Computer Architecture (HPCA), pages 121--132, Austin, TX, Feb. 2006.Google Scholar
- J. C. McCullough, Y. Agarwal, J. Chandrasheka, S. Kuppuswamy, A. C. Snoeren, and R. K. Gupta. Evaluating the effectiveness of model-based power characterization. In USENIX Annual Technical Conf., Portland, OR, June 2011. Google Scholar
Digital Library
- D. McIntire, K. Ho, B. Yip, A. Singh, W. Wu, and W. Kaiser. The low power energy aware processing (LEAP) embedded networked sensor system. In 5th Int'l Conf. on Information Processing in Sensor Networks, pages 449--457, Nashville, TN, Apr. 2006. Google Scholar
Digital Library
- D. Meisner, B. Gold, and T. Wenisch. PowerNap: Eliminating server idle power. In 14th Int'l Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pages 205--216, Washington, DC, Mar. 2009. Google Scholar
Digital Library
- E. Pinheiro, R. Bianchini, E. V. Carrera, and T. Heath. Dynamic cluster reconfiguration for power and performance. In Compilers and Operating Systems for Low Power, pages 75--93, 2003. Google Scholar
Digital Library
- E. Rotem, A. Naveh, D. Rajwan, A. Ananthakrishnan, and E. Weissmann. Power Management Architecture of the 2nd Generation Intel Core microarchitecture, formerly codenamed Sandy Bridge. In Hot Chips: A Symposium on High Performance Chips, Aug. 2011.Google Scholar
Cross Ref
- A. Roy, S. M. Rumble, R. Stutsman, P. Levis, D. Mazières, and N. Zeldovich. Energy management in mobile devices with the Cinder operating system. In 6th EuroSys Conf., pages 139--152, Salzburg, Austria, Apr. 2011. Google Scholar
Digital Library
- K. Shen. Request behavior variations. In 15th Int'l Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pages 103--116, Pittsburg, PA, Mar. 2010. Google Scholar
Digital Library
- K. Shen, M. Zhong, S. Dwarkadas, C. Li, C. Stewart, and X. Zhang. Hardware counter driven on-the-fly request signatures. In 13th Int'l Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pages 189--200, Seattle, WA, Mar. 2008. Google Scholar
Digital Library
- B. H. Sigelman, L. A. Barroso, M. Burrows, P. Stephenson, M. Plakal, D. Beaver, S. Jaspan, and C. Shanbhag. Dapper, a large-scale distributed systems tracing infrastructure. Technical report, Google, Apr. 2010.Google Scholar
- C. Stewart and K. Shen. Some Joules are more precious than others: Managing renewable energy in the datacenter. In Workshop on Power Aware Computing and Systems, Big Sky, MT, Oct. 2009.Google Scholar
- C. Stewart, M. Leventi, and K. Shen. Empirical examination of a collaborative web application. In IEEE Int'l Symp. on Workload Characterization, Seattle, WA, Sept. 2008.Google Scholar
- The Mathematical Association of America. WeBWorK: Online homework for math and science. http://webwork.maa.org/.Google Scholar
- M. Ware, K. Rajamani, M. Floyd, B. Brock, J. C. Rubio, F. Rawson, and J. B. Carter. Architecting for power management: The IBM POWER7 approach. In 16th Int'l Symp. on High-Performance Computer Architecture (HPCA), Bangalore, India, Jan. 2010.Google Scholar
- H. Zeng, C. S. Ellis, A. R. Lebeck, and A. Vahdat. ECOSystem: Managing energy as a first class operating system resource. In 10th Int'l Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pages 123--132, Boston, MA, Oct. 2002. Google Scholar
Digital Library
Index Terms
Power containers: an OS facility for fine-grained power and energy management on multicore servers
Recommendations
Power containers: an OS facility for fine-grained power and energy management on multicore servers
ASPLOS '13Energy efficiency and power capping are critical concerns in server and cloud computing systems. They face growing challenges due to dynamic power variations from new client-directed web applications, as well as complex behaviors due to multicore ...
Power containers: an OS facility for fine-grained power and energy management on multicore servers
ASPLOS '13: Proceedings of the eighteenth international conference on Architectural support for programming languages and operating systemsEnergy efficiency and power capping are critical concerns in server and cloud computing systems. They face growing challenges due to dynamic power variations from new client-directed web applications, as well as complex behaviors due to multicore ...
Towards better CPU power management on multicore smartphones
HotPower '13: Proceedings of the Workshop on Power-Aware Computing and SystemsAlthough multicore smartphones have become increasingly mainstream, it is unclear whether and how smartphone applications can utilize multicore CPUs to improve performance. In this paper we study the performance of mobile applications using multicore ...







Comments