Multi-Objective Load Balancing in Cloud Computing: A Meta-Heuristic Approach

Workload balancing in cloud computing continues to be a significant topic, notably within the Infrastructure as a Service (IaaS) paradigm. Site or web service being overcrowded or underloaded impairs the waiting times or culminates in a system malfunction, and it is indeed an issue that need not happen with cloud storage. To overcome such issues, a suitable access schedule should be devised, allowing the organization to spread work and overall resources available, a process known as load balancing. As a response, a unique multi-objective load balancing framework is presented in this research for optimum load balancing in the cloud. The suggested multi-objective load balancing paradigm considers power usage, bandwidth consumption, migration costs, memory usage, and load balancing parameters (Response time, Turnaround time, Server load). A novel hybrid optimization technique-Mouse Customized Golden Eagle Optimization (MCGEO) model is developed for optimal load balancing, which is conceptual combination of traditional Golden Eagle Optimizer (GEO) and Cat and Mouse-Based Optimizer (CMBO). Therefore, the newly developed hybrid optimization model improves convergence and addresses the optimization problems in load balancing. The proposed MCGEO achieved a throughput value of ∼281.6255, at 150 tasks for the cloud environment-2 this proves the superiority of the proposed approach.