Optimizing Virtual Machines in Cloud Resources using Gaussian Median based Tunicate Swarm Optimization

The growing demand for cloud services, coupled with the need for immediate provisioning of virtual machines (VMs) with varying resource sizes and types, creates an imbalance in cloud data centres. Although current cloud-based frameworks offer VM services, they also face challenges such as high energy consumption, increased computational costs, and other inefficiencies. This research aims to propose the Gaussian Medium based Tunicate Swarm Optimization (GM-TSA) approach for effective task scheduling in cloud environments. A combination of GM improves the convergence rate, permitting the approach to identify optimal solutions more rapidly. This research considers the three main multi-objective functions makespan, execution time and Resource Utilization (RU) to efficiently schedule the loss. The experimental results show that the proposed GM-TSA approach achieves the minimum makespan, minimum execution time and maximum RU of 58.8s, 720s and 92.0% on the 200 tasks, respectively, as opposed to the existing method, Particle Swarm Optimization with Grey Wolf Optimization (PSGWO) approach.