An energy-saving strategy based on multi-server vacation queuing theory in cloud data center

Energy consumption is a growing concern in cloud data centers because underutilization of servers results in significant wasted power. Thus, improving server utilization for optimal energy use is now an urgent issue. We propose an energy-saving strategy based on multi-server vacation queuing theory that switches servers between on and sleep in groups. The strategy incorporates both synchronous and asynchronous strategies. When the number of idle servers reaches to a given threshold, idle servers enter sleep mode synchronously as a group. Varying workloads cause groups of servers to sleep asynchronously. We model the data center with our strategy as an M/M/H vacation queuing system and construct a two-dimensional continuous-time Markov chain to formulate the queuing system. Using a powerful matrix-geometric method, we obtain the stationary probability distribution for the system states. We use results from theoretical and simulated experiments to estimate the performance of our approach. The results are valuable for studying the power-performance trade-off in cloud data centers.

[1]  Paul J. Kühn,et al.  Automatic energy efficiency management of data center resources by load-dependent server activation and sleep modes , 2015, Ad Hoc Networks.

[2]  Junzhou Luo,et al.  Dynamic Pricing Based Energy Cost Optimization in Data Center Environments: Dynamic Pricing Based Energy Cost Optimization in Data Center Environments , 2014 .

[3]  Harry G. Perros,et al.  Service Performance and Analysis in Cloud Computing , 2009, 2009 Congress on Services - I.

[4]  Tan Yiming,et al.  Policy of Energy Optimal Management for Cloud Computing Platform with Stochastic Tasks: Policy of Energy Optimal Management for Cloud Computing Platform with Stochastic Tasks , 2012 .

[5]  Ideguchi Tetsuo,et al.  Queuing Theoretic Approach to Server Allocation Problem in Time-delay Cloud Computing Systems , 2011 .

[6]  Zhe George Zhang,et al.  Analysis of multi-server queue with a single vacation (e, d)-policy , 2006, Perform. Evaluation.

[7]  Tian Nai-shuo,et al.  The M/M/c Queue with -policy and asynchronous multiple vacation of partial servers , 2006 .

[8]  Sanjay Ranka,et al.  Dynamic slack allocation algorithms for energy minimization on parallel machines , 2010, J. Parallel Distributed Comput..

[9]  Yuan-Shun Dai,et al.  Performance evaluation of cloud service considering fault recovery , 2009, The Journal of Supercomputing.

[10]  Tuan Phung-Duc,et al.  Impacts of Retrials on Power-Saving Policy in Data Centers , 2016, QTNA.

[11]  Ying Wang,et al.  An Energy-Saving Task Scheduling Strategy Based on Vacation Queuing Theory in Cloud Computing , 2015 .

[12]  Gregor von Laszewski,et al.  Towards Energy Aware Scheduling for Precedence Constrained Parallel Tasks in a Cluster with DVFS , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[13]  Yu Gong,et al.  Energy and performance management in large data centers: A queuing theory perspective , 2015, 2015 International Conference on Computing, Networking and Communications (ICNC).

[14]  Sherali Zeadally,et al.  A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems , 2016, Computing.

[15]  Hermann de Meer,et al.  Performance tradeoffs of energy-aware virtual machine consolidation , 2013, Cluster Computing.

[16]  Shobha Vasudevan,et al.  Verifying dynamic power management schemes using statistical model checking , 2012, 17th Asia and South Pacific Design Automation Conference.

[17]  Shoji Kasahara,et al.  Multi-server Queue with Job Service Time Depending on a Background Process , 2015, QTNA.

[18]  Binh Minh Nguyen,et al.  Enhancing service capability with multiple finite capacity server queues in cloud data centers , 2016, Cluster Computing.

[19]  Luo Jun,et al.  Dynamic Pricing Based Energy Cost Optimization in Data Center Environments , 2013 .

[20]  Tan Yi,et al.  Policy of Energy Optimal Management for Cloud Computing Platform with Stochastic Tasks , 2012 .

[21]  Yi Zhong,et al.  State-of-the-art research study for green cloud computing , 2011, The Journal of Supercomputing.