Effect of Memory Allocation on Power Consumption in Virtual Machine: Case Study Microsoft Hyper-V
Main Article Content
Abstract
Virtualization technology consolidates a number of servers to work together in a single hardware platform. Since memory is a shared resource in the virtual environment, it is necessary to allocate memory to all virtual servers. Initially, static memory allocation was the only way to allocate memory to virtual servers. At present, it is possible to have dynamic memory allocation, which has been introduced by Microsoft Hyper-V. However, this allocation requires some resources to be performed that cause overhead costs regarding energy consumption. Therefore, this research was performed in order to find differences in the power consumption in the virtual server environment when using static and dynamic memory allocation. The experiment was set up to measure power consumption on I/O and memory-intensive applications in the same virtual server environment; one experiment used static memory allocation and the other used dynamic memory allocation. We found that the power consumption of the dynamic memory allocation was 5–10% more than that of the static memory allocation, and this number can be reduced to 0.16–1.02% by setting the memory buffer value to 20%. In application requirements focusing on energy savings, it is recommended that the static memory allocation should be applied to both I/O and memory-intensive applications . If the dynamic memory allocation is a must in virtual machines, we recommend setting up the memory buffer value to 20% for the best results in terms of energy consumption with virtual servers.
Article Details
The articles published are the opinion of the author only. The author is responsible for any legal consequences. That may arise from that article.
References
[2] H. Fayyad-Kazan, L. Perneel, and M. Timmerman, “Benchmarking the performance of Microsoft Hyper-V server, VMware ESXi and Xen Hypervisors,” Journal of Emerging Trends in Computing and Information Sciences, vol. 4, no. 12, pp. 922–933, December 2013.
[3] O. Thomas, Training Guide: Administering Windows Server 2012 R2, Redmond, WA: Microsoft Press, 2014.
[4] A. Finn, P. Lownds, M. Luescher, and D. Flynn, Windows Server 2012 Hyper-V Installation and Configuration Guide, Indianapolis, IN: John Wiley & Sons Inc., 2013.
[5] J. W. Smith and I. Sommerville, “Workload classification & software energy measurement for efficient scheduling on private cloud platforms,” presented at the ACM SOCC, Cascais, Portugal, October 26–28, 2011.
[6] VMware, “Understanding full virtualization paravirtualization, and hardware assist,” VMware Inc., Palo Alto, CA, Tech. Rep. WP-028-PRD-01-01, March 2008.
[7] J. O. Iglesias, P. Perry, L. Murphy, T. S. Buda, and J. Thorburn, “An experimental methodology to evaluate energy efficiency and performance in an enterprise virtualized environment,” in Proceedings ICPE, 2014, pp. 51–62.
[8] B. Posri and B. Papasratorn, “Cloud computing resources provisioning using virtual server size expansion method,” The Journal of King Mongkut’s University of Technology North Bangkok, vol. 23, no. 1, pp. 180–187, 2013 (in Thai).
[9] A.Desai, R. Oza, P. Sharma, and B. Patel, “Hypervisor: A survey on concepts and taxonomy,” International Journal of Innovative Technology and Exploring Engineering, vol. 2, no. 3, pp. 222–225, 2013.
[10] V. Malik and C. R. Barde, “Survey on architecture of leading hypervisors and their live migration techniques,” International Journal of Computer Science and Mobile Computing, vol. 3, no. 11, pp. 65–72, 2014.
[11] Microsoft. (2010 July 12). Hyper-V Dynamic Memory Configuration Guide. [Online]. Available: http://technet.microsoft.com/en-us/library
[12] M. Tulloch, Training Guide: Installing and Configuring Windows Server 2012, Redmond, WA: Microsoft Press, 2012.
[13] Microsoft. (2012). Windows Server 2012: Server Virtualization. [Online]. Available: http://download. microsoft.com/documents/italy/SBP/server2012/
[14] A. Kansal, F. Zhao, J. Liu, N. Kothari, and A. A. Bhattacharya, “Virtual machine power metering and provisioning,” presented at the ACM SOCC, Indianapolis, IN, USA, June 10, 2010.
[15] X. Zhang, J. Lu, and X. Qin, “Energy consumption characteriation of heterogeneous servers,” in ChinaGrid Annual Conference, 2013, pp. 14–21.
[16] SPEC. (2006). SPEC CPU2006 Run and Reporting Rules. [Online]. Available: http://www.spec.org/cpu2006/Docs/runrules.html