Consolidating VMs in Green Cloud Computing Using Harmony Search Algorithm

Throughout history, energy consumption was not a matter to humanity and mankind considered resources on earth limitless, as his knowledge grew, he found out that using so much energy and producing a lot of greenhouse gases, has endangered his life. Nowadays Cloud networks and data centers consume a lot of energy. In order to decrease energy consumption like heuristic and meta-heuristic algorithms are in wide range of use in this problem because a Virtual Machine(VM) consolidation is considered a NP-hard problem. Ant colony system, artificial bee colony, genetic algorithm and local regression are some of these heuristic methods. These mentioned algorithms tried to maintain service level agreement (SLA) while reducing energy consumption and live migrations. On the other hand, the harmony search algorithm has acceptable convergence rate compared to swarm methods, and has less computation time compared to genetic algorithm to find the optimum answer. In this paper, we are motivated to use newly presented meta-heuristic harmony search algorithm. This algorithm has proven its efficiency in power management systems. This algorithm benefits in way that other particle and genetic driven algorithms don't. Some of these benefits are faster result, no need for initial parameters, and no need for data derivation. Using this algorithm for virtual machine consolidation allows reduction in energy consumption, SLA violation, live migrations quantity and energy SLA violation multiplication (ESV), where the proposed method has proven its efficiency and offered improvement ranges from minimum of 18.56% in ESV to maximum of 1988.23% in migrations compared to previous methods.