A Hybrid AFSS-SHC Optimization Methodology for Balancing the Load Efficiently in Cloud Environment

Abstract

Cloud computing helps to share data and provide many resources to users. Users pay only for those resources as much they used. Rapid increase in load to these cloud framework cannot be predicted. Load balancing is one of the issues in cloud computing that distributes the workload to the nodes in such a way no node is overloaded or under - loaded. Load balancing is a main challenge in cloud environment. In this work, scheduling algorithm is applied for load balancing by considering the cost of task execution and make span. This scheduling algorithm efficiently maps task to available nodes in cloud and it is beneficial to user and service provider. Load balancing segregates assignments of tasks among all available virtual machines from datacenters. Assignment of tasks to virtual machines can be done with minimum delay. To enhance the make span, resource utilization, our proposed framework utilizes AFSS-SHC load balancing strategy. A metaheuristics swarm intelligence algorithm which is NP-hard have been suggested to balance load across devices. The algorithms taken into account are-HEFT,PSO and PSO-HC. The proposed methodology AFSS-SHC optimized the task scheduling. Random tasks have been taken for this purpose and simulated to show that the proposed methodology works efficiently to reduce the make span of tasks to reduce the cost.