Enhanced Manta Ray Foraging Algorithm for Scheduling Scientific Workflows in Cloud Computing Environments Using Levy Flight and Heuristic Operator

Abstract

In modern computing, efficient task scheduling in cloud environments, especially for large-scale scientific workflows, presents a significant challenge as it is classified as a NP-hard problem. This study introduces an improved version of the Manta Ray Foraging Optimization Algorithm, named Lévy-Heuristic Manta Ray Foraging Optimization Algorithm (LH-MRFOA), which is enhanced with Lévy flight and heuristic search techniques to address these challenges. The Lévy flight mechanism is integrated to enhance the algorithm’s exploration capabilities, allowing it to avoid local optima effectively and achieve global convergence. Meanwhile, the heuristic search method is employed to improve the exploitation capability of the algorithm while ensuring more efficient resource utilization and reduced processing time. The proposed LH-MRFOA, which mimics the natural foraging behavior of manta rays, combines these enhancements to deliver superior performance in task scheduling by minimizing makespan, processing cost, storage cost, and bandwidth utilization across varying workflow sizes. Experimental evaluations on a heterogeneous cloud infrastructure reveal that the LH-MRFOA outperforms bio-inspired algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), particularly in scenarios that require high scalability and balanced resource allocation. This research substantially advances cloud task scheduling optimization, offering a robust solution for enhancing resource management and cost efficiency in real-world cloud applications.