GA-PSO-MIN: A HYBRID HEURISTIC ALGORITHM FOR MULTI-OBJECTIVE JOB SCHEDULING IN CLOUD COMPUTING

Due to the variable resource availability and diverse user needs, efficient task scheduling in cloud computing has become increasingly important. This study introduces GA-PSO-Min, a novel approach that synergistically combines genetic algorithms (GA), particle swarm optimization (PSO), and Min-Min strategy to improve scheduling efficiency in cloud environments. Unlike conventional approaches that prioritize single criteria, GA-PSO-Min emphasizes multi-objective optimization, minimizing the overall completion time while ensuring scalability and flexibility. The approach leverages the global search capabilities of GA and the fast convergence of PSO to initialize its population with a Min-Min solution, thereby outperforming standalone approaches. Compared to Min-Min, GA-PSO-Min reduces completion time by 2–7% in twelve distinct scenarios, including compute-intensive, I/O-intensive, and mixed workloads. The initial energy reduction is validated through a simple power model. It surpasses Min-Min with a temporal complexity of O(k⋅P⋅n⋅m), achieving a balance between enhanced performance and computational cost. The sensitivity analysis reveals the optimal resilience of the parameters (e.g., an inertia weight of 0.7), confirming GA-PSO-Min as an energy-efficient and scalable solution for modern cloud systems. Subsequent study will encompass improved QoS optimization and empirical validation.