Enhancing QoS with Resource Optimization Technique Based on Harmony Search in Cloud Environment

Abstract

In cloud environment, enhancing QoS is a challenging task to solve the multi objective problems. Effective load balancing and scheduling is a solution to enhance QoS. Job scheduling is a NP complete problem and is difficult to get the best scheduling especially for large number of jobs to enhance QoS. Multiple activities may need to be scheduled efficiently on various resources (CPU, virtual machines) by reducing completion time and maximizing resource utilization. A meta-heuristic method has been proposed for optimizing multi-objective scheduling problems in cloud systems using harmony search algorithm. According to cloud user's expectations, throughput should be high such that most of the user tasks are completed before the deadline expires thereby resulting in a low task rejection rate. Major purpose of the HS-based, NEW HS scheduling algorithm is to arrange upcoming cloud applications at virtual machines in such a way that tasks are completed in least amount of time (execution and makespan time) with low execution cost and take deadline of cloud application as a QoS constraint.