A Hybrid Approach to Multi-Objective Task Scheduling in Cloud Computing: Merging Estimation of Distribution and Genetic Algorithms

In the evolving landscape of cloud computing, efficient task scheduling plays a pivotal role in optimizing resource utilization and enhancing performance. This paper presents a novel hybrid algorithm that combines the strengths of Estimation of Distribution Algorithms (EDA) and Genetic Algorithms (GA) to address multi-objective task scheduling challenges. The proposed approach aims to balance conflicting objectives such as minimizing execution time, energy consumption, and cost while maximizing resource utilization. By leveraging the probabilistic modeling capabilities of EDA and the evolutionary search efficiency of GA, the hybrid algorithm achieves a more diverse solution set and faster convergence compared to traditional methods. Extensive simulations demonstrate the effectiveness of the proposed algorithm across various cloud environments and task complexities. The results highlight significant improvements in achieving Pareto-optimal solutions, offering a robust framework for cloud service providers to enhance scheduling efficiency. This study underscores the potential of hybrid metaheuristic techniques in addressing the dynamic and complex nature of cloud computing task scheduling, paving the way for more resilient and adaptive cloud infrastructures.