International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 2 Issue: 9 2887 – 2890 _______________________________________________________________________________________________ 2887 IJRITCC | September 2014, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Energy Efficient Policies in Cloud Computing Sandeep Kaur 1 , 1 Department of Computer Engineering, Punjabi University Yadavindra College of Engineering, Talwandi Sabo ,Punjab, India s.simmidhillon052@gmail.com Meenakshi Bansal 2 2 Assistant Professor, Department of Computer Engineering, Yadavindra College of Engineering, Talwandi Sabo ,Punjab, Punjabi University, India ermeenu10@gmail.com Abstract: - Energy efficient processing is known as Green Cloud Computing. It is getting popular with the demand of time and become important to reduce the energy consumption of the Cloud datacenters in order to create a more nature friendly Cloud environment. In this paper, two policies are proposed i.e. minimum CPU utilization policy and low utilization host selection policy. These two policies are deciding which VM will be picked for migration and which host will be selected to reallocate that VM. The Min-Min scheduling algorithm is used to schedule the tasks at Cloud datacenter. The load balancing concept is proposed that further helps in reducing the energy consumption. The results of proposed energy efficient Cloud scenario is compared with power aware and non-power aware Clouds. Keywords: - Cloud Computing, Energy Efficient, Power Aware Cloud. __________________________________________________*****_________________________________________________ I. Introduction Cloud Computing is a technology in which resources are provided to user in terms of services. There are three types of services provided by Cloud service provider i.e. software as a service, platform as a service and infrastructure as a service. The main motive of Cloud service provider is to offer quality of service to its customer. The SLA (service level agreement) is a document in which QoS constraints are defined and signed between Cloud service provider and Cloud user. The Cloud service provider offer services from Cloud datacenter and try to optimize the energy consumption of datacenter. These Cloud data centers need electric power for its working so, to reduce the operation cost. The Cloud service provider makes the data centre energy efficient. Energy efficient cloud computing also known as Green Cloud Computing is the demand of the time. As cloud computing is getting popular, it is becoming important to reduce the energy consumption of the cloud data centers in order to create a more nature friendly cloud environment which is also called Green Cloud. Moreover creating a cloud that is low in energy consumption will increase ROI (return on investment). The rest of the paper is structured as follow: The related work is presented in section 2. The proposed approach is presented in section 3. Experimental results and comparison are presented in section 4 and Section 5 concludes the work carried out. II. Related Work The scheduling is performed [1] using genetic algorithm to allocate reliable virtual machines to workflow tasks. Genetic algorithm results into highly power utilization at Cloud datacenter. In [2] there is needs to make a policy that reduce the energy consumption at datacenters. In [3] propose power aware policy which results into saving in energy consumption at datacenter. The power aware policy switch off the idle nodes and transfer there load to some underutilized virtual machines. The results are evaluated on the basis of energy consumption, SLA violation and number of virtual machines migrated. In [4] proposed a policy in which live virtual machines are rearranged to reduce the overall energy consumption of datacenter. Cloud global optimization algorithm is proposed that results into high energy saving and provides infrastructure as a service. In [5] present a review on various energy efficient methods. From the literature review we feel that there is a need to make the datacenter energy efficient when scheduling the task on it. III. Proposed Approach In the proposed work, there are two policies on the basis of which the energy consumption is reduced at Cloud datacenter. The proposed work finds a better selection policy for the virtual machine to be migrated and find a policy to decide which host will be selected for reallocation of the virtual machine. The minimum CPU utilization policy and low utilization host selection policy decide which virtual machine will be picked for migration and which host will be selected to reallocate that virtual machine. A. Energy Efficient Policies 1. Selection Policy In our research we are proposing a minimum CPU utilization selection policy. According to our approach Vm whose CPU utilization is less than 10%, should be picked for migration so that underutilized host should be shut down. We will migrates virtual machine of the underutilized host to the other hosts, we are proposing that Vm should be