Energy efficient resource provisioning with dynamic VM placement using energy aware load balancer in cloud

Cloud Computing is merely defined as delivering IT services over the Internet. It is ubiquitous, on-demand provisioning of resources. The fundamental principle of reusability of IT capabilities is the basis of cloud computing. To meet the growing demand for computations and large volumes of data, the model of cloud computing environment provides high performance servers and persistent storage devices like hard disk drives. These resources are the major source of the enormous power consumption in cloud data center along with air conditioning and cooling equipment. Energy consumption and diffusion of carbon that contaminate the atmosphere go hand in hand. The more the energy consumed, higher is the carbon footprint, which in turn maximizes the operating cost. Energy aware load balancer like Weighted First-cum-First-Served, helps in avoiding overheating and overloading of servers, by evenly balancing the workload across all the servers in the data center, hence reducing the amount of energy consumed. Dynamic placement of virtual machines in the data center enables to ensure availability of computing resources, tries to minimize overloaded servers and save the power consumption by shutting down some under-utilized, idle servers and thereby in maximizing the usage of resources. The main aim of the dynamic VM placement ensures the mapping of virtual machines to physical machines at run time, in such a way that the physical machine can be utilized to their maximum efficiency, and is non-preemptive in nature, where the already running VMs are neither shut down nor interrupted but in turn provides a list of live VM migrations to be executed to reach the optimal solution, thereby saves energy consumption to a greater extent. The central objective of this work is dedicated to develop an energy efficient resource provisioning framework with dynamic virtual machine placement using energy aware load balancer in cloud.