Cloud computation has revolutionized how resources are utilized and tasks are managed by offering scalable, efficient solutions. Scheduling task, a critical component of cloud systems, ensures optimal allocation of resources to minimize execution time and computational costs. Given its complexity as an NP-complete problem, heuristic approaches like Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) have proven effective. This study investigates the performance of PSO and GA in scheduling task, highlighting their unique strengths and limitations. This paper proposes a Hybrid PSO-GA model, which utilizes the global search capability of PSO with the exploration capability of GA to attain a balanced and efficient scheduling task. The experimental results have shown that, compared to PSO and GA on their own, the hybrid model is more effective in the optimization of execution time and communication costs, achieving efficient consumption of resources. The paper also reports task graph modelling, analyzes communication cost, and reports strategies for the resource mapping, to provide a more complete resource mapping approach for future research.