Multi-objective task scheduling in cloud data centers: a differential evolution chaotic whale optimization approach

Cloud computing has transformed the accessibility of scalable computational resources, emphasizing the need for efficient task scheduling to optimize resource utilization. This study presents DECWOA, a novel Differential Evolution Chaotic Whale Optimization Algorithm designed to enhance scheduling in cloud data centers. DECWOA integrates concepts from Sine chaos theory, employing chaotic initialization processes based on sine functions to promote exploration diversity. Moreover, it incorporates adaptive inertia weights that dynamically adjust exploration and exploitation tendencies, along with differential variance to minimize solution space, thereby improving convergence. The algorithm achieves significant reductions in task and workflow execution durations, demonstrating a remarkable 64% decrease in execution time and an 11% reduction in data center costs. Through comprehensive comparisons with various scheduling algorithms and meta-heuristics using Cloudsim Plus, DECWOA consistently outperforms alternatives such as AIGA and GA. Furthermore, its adaptability to parameter variations ensures superior solutions across diverse configurations. This research underscores the effectiveness of DECWOA in multi-objective task scheduling, highlighting its pivotal role in accelerating convergence rates, cutting operational costs, and enhancing cloud service efficiency.