Modified Particle Swarm Optimization

Cloud computing has emerged as a pivotal technology for managing and processing data, with a primary objective to offer efficient resource access while minimizing expenses. The allocation of resources is a critical aspect that can significantly reduce costs. This process necessitates the continuous assessment of the current status of each resource to design algorithms that optimize allocation and enhance overall system performance. Numerous algorithms have been developed to address the challenge of resource allocation, yet many fail to satisfy requirements of time efficiency and load balancing in cloud computing environments. This paper introduces a novel approach that classifies tasks according to their resource demands, employs a modified particle swarm optimization (PSO) algorithm, and incorporates load balancing strategies. The proposed method initially clusters tasks based on their resource requirements, subsequently utilizes the PSO algorithm to determine the best task-to-resource assignments, and finally implements a load balancing algorithm to reduce costs through balanced load distribution. The validity of the proposed method is tested and simulated using the Cloudsim tool. The simulation results indicate that the proposed method achieves lower average response time, waiting times, and energy consumption than existing baseline methods.