Energy-Aware Scheduling Using Nature-Inspired Algorithms in Cloud Computing

The ongoing growth in data processing requirements in today’s cloud computing environment has led to a notable increase in data center energy usage. This study undertakes a thorough investigation into the use of nature-inspired iterative algorithms in cloud computing energy-aware scheduling, such as ant colony optimization (ACO), particle swarm optimization (PSO), genetic algorithm (GA), grey wolf optimization (GWO), bat algorithm (BA) and cuckoo search (CS). Five literature databases were searched, including Web of Science, IEEE Xplore, SpringerLink, Scopus and Google Scholar. Applying the search criteria produced 1315 papers, of which 61 were selected: 20 papers on ACO, 9 on GA, 9 on GWO, 9 on CS, 8 on PSO and 6 on BA. The review scrutinized the optimization models, performance evaluation methods, termination criteria for algorithm iterations, and the characteristics of datasets used in experimentation. Most studies use multi-objective models (70.4% of those reporting the model) and are evaluated in simulation with synthetic workloads, mainly in CloudSim; statistical analysis is reported in only 16.4% of the studies, and about half of the studies (52.5%) do not report a stopping criterion; hybrid algorithms report consistent improvements over single-algorithm baselines (e.g., up to 18.56% lower energy consumption). Looking ahead, the review highlights the potential of hybrid optimization models as a promising research direction to address evolving challenges in energy-aware scheduling.