A Superlative Approach to Improve the QoS by Load balancing in Cloud Computing

  • Vrajesh Sharmaa et. al


Abstract: Scheduling of tasks and resources is a big problem in cloud computing, as there are many factors such as priority, cost, quality of service and deadline which need to be taken care of before devising any scheduling strategy. Efficient job scheduling algorithm enables the optimal utilization of resources in cloud computing platform. Sometimes while scheduling, some virtual machines (VMs) get over-loaded and some remain under-loaded which produces adverse effect on the throughput of the system. The quest to balance the load during scheduling of cloudlets paves the path for the research in the load balancing mechanisms. Prevalent priority based job scheduling strategies are silent in deciding scheduling scheme for tasks with the same priority and strive hard inĀ  appropriately allocating jobs to virtual machines. A Credits based task scheduling algorithm was rendered using modified K-means for clustering of jobs and VMs but it was observed that for providing optimized performance, this arrangement further needed some load balancing strategy to balance the load. Therefore, Honey Bee Foraging behaviour inspired Load balancing technique was roped in for load balancing. Work pertaining to the use of Honey Bee Foraging Load Balancing Algorithm coupled with credits based scheduling and modified K-means clustering technique is not available. Results indicate that the proposed scheduling algorithm has excelled existing priority-based scheduling strategy and it has been empirically proven with experimental/simulated results in this paper.