Dynamic right-sizing for power-proportional data centers
Top Cited Papers
- 1 April 2011
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1098-1106
- https://doi.org/10.1109/infcom.2011.5934885
Abstract
Power consumption imposes a significant cost for data centers implementing cloud services, yet much of that power is used to maintain excess service capacity during periods of predictably low load. This paper investigates how much can be saved by dynamically `right-sizing' the data center by turning off servers during such periods, and how to achieve that saving via an online algorithm. We prove that the optimal offline algorithm for dynamic right-sizing has a simple structure when viewed in reverse time, and this structure is exploited to develop a new `lazy' online algorithm, which is proven to be 3-competitive. We validate the algorithm using traces from two real data center workloads and show that significant cost-savings are possible.Keywords
This publication has 23 references indexed in Scilit:
- Optimality, fairness, and robustness in speed scaling designsPublished by Association for Computing Machinery (ACM) ,2010
- Power-Aware Speed Scaling in Processor Sharing SystemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Power and performance management of virtualized computing environments via lookahead controlCluster Computing, 2008
- Workload Analysis and Demand Prediction of Enterprise Data Center ApplicationsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2007
- Managing energy and server resources in hosting centersPublished by Association for Computing Machinery (ACM) ,2001
- Dynamic TCP acknowledgement and other stories about e/(e-1)Published by Association for Computing Machinery (ACM) ,2001
- Better algorithms for unfair metrical task systems and applicationsPublished by Association for Computing Machinery (ACM) ,2000
- On-line Learning and the Metrical Task System ProblemMachine Learning, 2000
- An optimal on-line algorithm for metrical task systemJournal of the ACM, 1992
- Competitive snoopy cachingAlgorithmica, 1988