Energy Aware Next Fit Allocation Approach for Placement of VMs in Cloud Computing Environment

Cloud computing enables the IT giants to outsource their infrastructure, by providing a sharable pool of computing sources. These sources consume a huge amount of energy that not only increase the running expenses but also produce CO2 emission in the environment. Therefore, the main issue is to manage and optimize the available resources for saving the energy. It can best be done by dividing the physical machines into virtual machines and maintaining the number of active machines according to the dynamic workload. This process of server consolidation includes finding the overloaded hosts, selection of VMs from the hosts with excess or under load and, finally, placing them all over the available physical hosts dynamically. In this context, a novel approach for placing virtual machines has been proposed that aims to reduce energy consumption and SLA violation. Inspired from the bin packing problem, Next fit allocation policy is tested for placing a VM over the available hosts. Suitability of hosts is defined primarily on the basis of minimum energy consumption by a VM on a host while placement. However, searching for the hosts is optimized using next-fit policy. Experiments are performed in the cloudsim simulator tool and results are compared with the existing policy of best-fit. Proposed approach has identified better results for various performance matrices considered during the experiments.