ABSTRACT
The services offered for computing from the aspects of both hardware and software can be referred to as the term ‘cloud’ in cloud computing. Cloud computing has made it possible to use data and content abundantly through a large number of wired or wireless devices. The environment provided by the cloud is required to provide services flexibly by contracting or expanding the different types of services to the application developers as per the demand. Hence, balancing load among several nodes and their instances with a suitable mechanism becomes important. Load balancing is mainly focused at assigning jobs among available nodes to attain effective resource utilization that also handles the jobs fairly. With no approach employed for load balancing, the system is usually seen to decrease the quality of service (QoS) of real-time applications. The aim of load balancing is to distribute load among the nodes to maximize the efficiency at each of the nodes, improving the performance as a whole. Load balancing can be achieved by migrating processes or virtual machines among the nodes in the system. The approach employed to identify the lightly loaded destination node to whom the processes/virtual machines migrate can be optimized by employing the genetic algorithm (GA). The GA optimizes the selection of the destination node on to which the load can be off-loaded. The fitness function employed by the GA ensures selecting the optimal node and thus facilitating load balancing.
