GWO-Based Simulated Annealing Approach for Load Balancing in Cloud for Hosting Container as a Service

Abstract

Container-based virtualization has gained significant popularity in recent years because of its simplicity in deployment and adaptability in terms of cloud resource provisioning. Containerization technology is the recent development in cloud computing systems that is more efficient, reliable, and has better overall performance than a traditional virtual machine (VM) based technology. Containerized clouds produce better performance by maximizing host-level resource utilization and using a load-balancing technique. To this end, this article concentrates on distributing the workload among all available servers evenly. In this paper, we propose a Grey Wolf Optimization (GWO) based Simulated Annealing approach to counter the problem of load balancing in the containerized cloud that also considers the deadline miss rate. We have compared our results with the Genetic and Particle Swarm Optimization algorithm and evaluated the proposed algorithms by considering the parameter load variation and makespan. Our experimental result shows that, in most cases, more than 97% of the tasks were meeting their deadline and the Grey Wolf Optimization Algorithm with Simulated Annealing (GWO-SA) performs better than all other approaches in terms of load variation and makespan.