Energy conservation policies for web servers

作者: Ramakrishnan Rajamony , Mootaz Elnozahy , Michael Kistler

DOI:

关键词: Web serverComputer networkWorkloadOperating systemEnergy conservationServerEnergy managementPower managementComputer scienceEnergy consumptionDynamic voltage scaling

摘要: Energy management for servers is now necessary technical, financial, and environmental reasons. This paper describes three policies designed to reduce energy consumption in Web servers. The employ two power mechanisms: dynamic voltage scaling (DVS), an existing mechanism, request batching, a new mechanism introduced this paper. first policy uses DVS isolation, except that we extend recently task-based use server environments with many concurrent tasks. second batching conserve during periods of low workload intensity. third both mechanisms processor usage over wide range intensities. All the trade off system responsiveness save energy. However, feedback-driven control framework order while maintaining given quality service level, as defined by percentile-level response time. We evaluate using Salsa, web simulator has been extensively validated time against measurements from commodity server. Three daylong static workloads real systems are used quantify savings: Nagano Olympics98 server, financial services company site, disk intensive workload. Our results show when required maintain 90th-percentile 50ms, 8.7% 38% 3.1% 27% respectively CPU base system. polices provide these savings complementary combined effective all across broad intensities, saving 17% 42%

参考文章(12)
Arun K. Iyengar, Mark S. Squillante, Li Zhang, Analysis and characterization of large-scale Web server access patterns and performance World Wide Web. ,vol. 2, pp. 85- 100 ,(1999) , 10.1023/A:1019244621570
Taliver Heath, Enrique V. Carrera, Ricardo Bianchini, Eduardo Pinheiro, Load balancing and unbalancing for power and performance in cluster-based systems ,(2001) , 10.7282/T3-AGFW-YT73
Ramakrishnan Rajamony, Mootaz Elnozahy, Measuring client-perceived response times on the WWW usenix symposium on internet technologies and systems. pp. 16- 16 ,(2001)
E. N. Elnozahy, Michael Kistler, Ramakrishnan Rajamony, Energy-Efficient Server Clusters Power-Aware Computer Systems. pp. 179- 197 ,(2003) , 10.1007/3-540-36612-1_12
Kinshuk Govil, Edwin Chan, Hal Wasserman, Comparing algorithm for dynamic speed-setting of a low-power CPU Proceedings of the 1st annual international conference on Mobile computing and networking - MobiCom '95. pp. 13- 25 ,(1995) , 10.1145/215530.215546
Jeffrey S. Chase, Darrell C. Anderson, Prachi N. Thakar, Amin M. Vahdat, Ronald P. Doyle, Managing energy and server resources in hosting centers symposium on operating systems principles. ,vol. 35, pp. 103- 116 ,(2001) , 10.1145/502034.502045
David Mosberger, Tai Jin, httperf—a tool for measuring web server performance measurement and modeling of computer systems. ,vol. 26, pp. 31- 37 ,(1998) , 10.1145/306225.306235
J. Flinn, M. Satyanarayanan, PowerScope: a tool for profiling the energy usage of mobile applications workshop on mobile computing systems and applications. pp. 2- 10 ,(1999) , 10.1109/MCSA.1999.749272
David Brooks, Vivek Tiwari, Margaret Martonosi, Wattch: a framework for architectural-level power analysis and optimizations international symposium on computer architecture. ,vol. 28, pp. 83- 94 ,(2000) , 10.1145/339647.339657
Jacob R. Lorch, Alan Jay Smith, Improving dynamic voltage scaling algorithms with PACE Proceedings of the 2001 ACM SIGMETRICS international conference on Measurement and modeling of computer systems - SIGMETRICS '01. ,vol. 29, pp. 50- 61 ,(2001) , 10.1145/378420.378429