<![CDATA[Task scheduling is a critical component in cloud computing to ensure optimal resource allocation and execution time. This study proposes a hybrid algorithm that combines Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to minimize makespan in cloud task scheduling. The algorithm leverages the global exploration capability of GA and the local exploitation strength of PSO. The simulation was conducted using CloudSim 3.0.3 with three different workload scenarios: light, medium, and heavy. Each configuration varied in terms of cloudlet count, virtual machines, CPU capacity, task length, and memory allocation. Results indicate that the hybrid GA–PSO algorithm consistently outperformed standalone GA and PSO in minimizing makespan, particularly under heavy workloads. It achieved an 11.0% reduction compared to GA and 22.3% compared to PSO. Moreover, the hybrid approach demonstrated greater stability and adaptability in constrained resource environments. These findings support the practical use of the hybrid GA–PSO algorithm in dynamic cloud systems and highlight its potential for future development in multi-objective optimization and real-world deployments.]]>
Copyrights © 2026
