Deadline constrained based dynamic load balancing algorithm with elasticity in cloud environment

The most challenging problem for a cloud service provider is maintaining the quality of service parameters like reliability, elasticity, keeping the deadline and minimizing the makespan time as also the task rejection ratio. Therefore, the cloud service provider needs a dynamic task scheduling algorithm that reduces the makespan time while increasing the utilization ratio of cloud resources and meeting the user defined QoS parameters. In this paper, we have developed a dynamic scheduling algorithm that balances the workload among all the virtual machines with elastic resource provisioning and deprovisioning based on the last optimal k-interval. Further, the algorithm has been tested on variable number of tasks (10 to 30) to achieve better scalability. The computational results (Figs. 5-10) show that the developed algorithm decreases the makespan time and increases the task to meet the deadline ratio compared with the min-min, the first come-first-serve and the shortest-job-first algorithms in all conditions.