Analysis of virtual machine live-migration as a method for power-capping

To reduce the construction cost of the power-supplying infrastructure in data centers and to increase the utilization of the existing one, many researchers have introduced software-based or hardware-based power-capping schemes. In servers with consolidated virtual machines, which can be easily found in cloud systems, exporting virtual machines to other light-loaded servers through live-migration is one of the key approaches to impose power-capping on servers. Up until now, most researchers who have tried to achieve power-capping through live-migration assumed that exporting a virtual machine instantly reduces the server power consumption. However, our analysis introduced in this paper reveals that the power consumption remains high or increases for a few seconds during a migration instance. This behavior contradicts the aim of power-capping, and may endanger the stability of servers. Based on this observation, we also propose and evaluate two power-suppressing live-migration schemes to resolve the power overshooting issue. Our evaluation shows that both approaches immediately limit the power consumption after live-migration is initiated.

[1]  Jie Liu,et al.  Power Budgeting for Virtualized Data Centers , 2011, USENIX Annual Technical Conference.

[2]  Bruce Nordman,et al.  Data Center Power Requirements: Measurements From Silicon Valley , 2003 .

[3]  Wolf-Dietrich Weber,et al.  Power provisioning for a warehouse-sized computer , 2007, ISCA '07.

[4]  Albert Y. Zomaya,et al.  Energy efficient utilization of resources in cloud computing systems , 2010, The Journal of Supercomputing.

[5]  Hai Jin,et al.  Live migration of virtual machine based on full system trace and replay , 2009, HPDC '09.

[6]  Sherief Reda,et al.  Pack & Cap: Adaptive DVFS and thread packing under power caps , 2011, 2011 44th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

[7]  Darryl Gove,et al.  CPU2006 working set size , 2007, CARN.

[8]  Bishop Brock,et al.  Architecting for power management: The IBM® POWER7™ approach , 2010, HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture.

[9]  Andrew Warfield,et al.  Live migration of virtual machines , 2005, NSDI.

[10]  Rajkumar Buyya,et al.  Cost of Virtual Machine Live Migration in Clouds: A Performance Evaluation , 2009, CloudCom.

[11]  Yellu Sreenivasulu,et al.  FAST TRANSPARENT MIGRATION FOR VIRTUAL MACHINES , 2014 .

[12]  Anand Sivasubramaniam,et al.  Power Consumption Prediction and Power-Aware Packing in Consolidated Environments , 2010, IEEE Transactions on Computers.

[13]  Anand Sivasubramaniam,et al.  Statistical profiling-based techniques for effective power provisioning in data centers , 2009, EuroSys '09.

[14]  Karsten Schwan,et al.  VPM tokens: virtual machine-aware power budgeting in datacenters , 2009, Cluster Computing.

[15]  Feng Zhao,et al.  Virtual machine power metering and provisioning , 2010, SoCC '10.

[16]  Mohsen Sharifi,et al.  Power-efficient distributed scheduling of virtual machines using workload-aware consolidation techniques , 2012, The Journal of Supercomputing.

[17]  Andrew Warfield,et al.  Xen and the art of virtualization , 2003, SOSP '03.

[18]  Mahadev Satyanarayanan,et al.  Internet suspend/resume , 2002, Proceedings Fourth IEEE Workshop on Mobile Computing Systems and Applications.

[19]  Randy H. Katz,et al.  Above the Clouds: A Berkeley View of Cloud Computing , 2009 .

[20]  Anand Sivasubramaniam,et al.  Profiling, Prediction, and Capping of Power Consumption in Consolidated Environments , 2008, 2008 IEEE International Symposium on Modeling, Analysis and Simulation of Computers and Telecommunication Systems.

[21]  Eyal de Lara,et al.  SnowFlock: rapid virtual machine cloning for cloud computing , 2009, EuroSys '09.

[22]  Xiaorui Wang,et al.  Server-Level Power Control , 2007, Fourth International Conference on Autonomic Computing (ICAC'07).

[23]  Laurent Lefèvre,et al.  Designing and evaluating an energy efficient Cloud , 2010, The Journal of Supercomputing.

[24]  Xiaorui Wang,et al.  Power capping: a prelude to power shifting , 2008, Cluster Computing.

[25]  Joonwon Lee,et al.  Energy Efficient Scheduling of Real-Time Tasks on Multicore Processors , 2008, IEEE Transactions on Parallel and Distributed Systems.

[26]  Hai Jin,et al.  Optimizing the live migration of virtual machine by CPU scheduling , 2011, J. Netw. Comput. Appl..

[27]  Gautam Kumar,et al.  CosMig: Modeling the Impact of Reconfiguration in a Cloud , 2011, 2011 IEEE 19th Annual International Symposium on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems.

[28]  Kartik Gopalan,et al.  Post-copy based live virtual machine migration using adaptive pre-paging and dynamic self-ballooning , 2009, VEE '09.

[29]  Vanish Talwar,et al.  No "power" struggles: coordinated multi-level power management for the data center , 2008, ASPLOS.

[30]  Qiang Huang,et al.  Power Consumption of Virtual Machine Live Migration in Clouds , 2011, 2011 Third International Conference on Communications and Mobile Computing.

[31]  Hai Jin,et al.  Performance and energy modeling for live migration of virtual machines , 2011, Cluster Computing.

[32]  Zhikui Wang,et al.  Feedback Control Algorithms for Power Management of Servers , 2008 .

[33]  M. Rosenblum,et al.  Optimizing the migration of virtual computers , 2002, OSDI '02.

[34]  Mor Harchol-Balter,et al.  Power Capping Via Forced Idleness , 2009 .

[35]  Satoshi Sekiguchi,et al.  Enabling Instantaneous Relocation of Virtual Machines with a Lightweight VMM Extension , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[36]  Kang G. Shin,et al.  LiteGreen: Saving Energy in Networked Desktops Using Virtualization , 2010, USENIX Annual Technical Conference.