Load Balancing in Cloud Computing Using Meta Heuristic Algorithm

Cloud computing has transformed the delivery of computing resources, offering scalable and flexible solutions. Cloud load balancing optimizes resource utilization, minimizes response time, and ensures high availability by distributing workloads across multiple computing resources. This complex problem involves challenges like overloading, where servers exceed capacity, leading to performance degradation, and underloading, causing resource inefficiency. Achieving efficient load balancing is critical for improving system performance. This study develops two meta-heuristic approaches for multi-objective cloud load balancing, focusing on reducing makespan (total task completion time) and improving CPU utilization. Inspired by natural phenomena, particle swarm optimization and grey wolf optimization are explored as solutions to optimize cloud load balancing. The algorithms are evaluated using the CloudSim simulator to assess their performance. Results indicate that grey wolf optimization outperforms particle swarm optimization by converging faster and achieving lower total execution times, demonstrating its superiority in balancing workloads and optimizing resource use in cloud environments.