An Enhanced Approach of Genetic and Ant colony based Load Balancing in Cloud Environment

Abstract

Cloud Computing is an economic, flexible delivery platform providing business or consumer IT services over the Internet. It allows users to take benefit from all technologies, without the need of deep knowledge or expertise in it. Load balancing is one of the key processes in cloud computing which avoids the situation where nodes become overloaded. Load balancing stabilizes the Quality of Service (QOS) which includes response time, cost, throughput, performance and resource utilization. At peak time it is difficult for the servers to handle the incoming requests with the available number of virtual machines, so some extra virtual machines were in need to continue the execution without any fault and delay. In this proposed system, the additional virtual machines were included using genetic approach so that the best virtual machines could be allocated to handle the requests. The allotment of best virtual machines could handle the requests in a very effective and fast manner. During the execution, if some virtual machines were overloaded with requests, the load could be balanced using ant colony optimization technique. The above technique would share the extra load to other lightly loaded and idle virtual machines. On the other hand the overall energy consumption is optimized by switching off the virtual machines after their work completion or when they were idle.