OPTIMIZING CLOUD PERFORMANCE: A COMPARATIVE ANALYSIS OF BIRD SWARM AND ANT COLONY ALGORITHMS FOR LOAD BALANCING

Abstract Load balancing plays a vital role in the realm of cloud computing by efficiently dispersing workloads across multiple servers or resources, which serves to enhance overall performance, availability, and scalability. The primary goal of load balancing is to optimize resource utilization and prevent server overload, thereby optimizing the entire cloud infrastructure. In addressing the challenges associated with workload distribution and resource utilization optimization in the cloud, researchers have devised algorithms inspired by natural processes like evolution, swarm behavior, and genetics. This research assesses the performance of two such algorithms, namely ant colony optimization (ACO) and bird swarm optimization (BSO), with a focus on load balancing. A comparative analysis is carried out using various parameters, including fitness score, throughput, resource utilization, and makespan. The findings demonstrate that the BSO algorithm surpasses the ACO algorithm in terms of fitness score, throughput, resource utilization, and makespan. To conduct these experiments, the CloudSim simulator is utilized within the NetBeans development environment.