A Simplified Method of Measurement of Energy Consumption in Cloud and Virtualized Environment

Measuring energy consumption is an essential step in the development of policies for the management of energy in every IT system. There is a wide range of methods using both hardware and software for measuring energy consumed by the system accurately. However, most of these methods measure energy consumed by a machine or a cluster of machines. In environments such as Cloud that an application can be built from components with comparable characteristics, measuring energy consumed by a single component can be extremely beneficial. For example, if we can measure energy consumed by different HTTP servers, then we can establish which one consumes less energy performing a given task. As a result, the Cloud provider can provide incentives, so that, application developers use the HTTP server that consume less energy. Indeed, considering size of the Cloud, even a small amount of saving per Virtual Machine can add up to a substantial saving. In this paper, we propose a technique to measure energy consumed by an application via measuring energy consumed by the individual processes of the application. We shall deal with applications that run in a virtualized environment such as Cloud. We present two implementations of our idea to demonstrate the feasibility of the approach. Firstly, a method of measurement with the help of Kernel-Based Virtual Machine running on a typical laptop is presented. Secondly, in a commercial Cloud such as Elastic host, we describe a method of measuring energy consumed by processes such as HTTP servers. This will allow commercial providers to identify which product consumes less energy on their platform.

[1]  Paula Smith,et al.  VMmark: A Scalable Benchmark for Virtualized Systems , 2006 .

[2]  Xilong Qu,et al.  Virtual machine power measuring technique with bounded error in cloud environments , 2013, J. Netw. Comput. Appl..

[3]  Barry Press,et al.  Building the Power-Efficient PC: A Developer's Guide to ACPI Power Management with CD-ROM , 2001 .

[4]  C. Pipper,et al.  [''R"--project for statistical computing]. , 2008, Ugeskrift for laeger.

[5]  Zhen Liu,et al.  Traffic model and performance evaluation of Web servers , 2001, Perform. Evaluation.

[6]  Cees T. A. M. de Laat,et al.  Profiling Energy Consumption of VMs for Green Cloud Computing , 2011, 2011 IEEE Ninth International Conference on Dependable, Autonomic and Secure Computing.

[7]  Dr Michael J de Smith Statistical Analysis Handbook , 2014 .

[8]  Emily Halili,et al.  Apache JMeter , 2008 .

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

[10]  Ziming Zhang,et al.  Profiling and analysis of power consumption for virtualized systems and applications , 2010, International Performance Computing and Communications Conference.

[11]  Roy T. Fielding,et al.  The Apache HTTP Server Project , 1997, IEEE Internet Comput..

[12]  Ying Wang,et al.  An Online Power Metering Model for Cloud Environment , 2012, 2012 IEEE 11th International Symposium on Network Computing and Applications.

[13]  Vipin Chaudhary,et al.  VMeter: Power modelling for virtualized clouds , 2010, 2010 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW).

[14]  Liang Liu,et al.  GreenCloud: a new architecture for green data center , 2009, ICAC-INDST '09.