Alsadie Deafallah, Alsulami Musleh
Department of Computer Science and Artificial Intelligence, College of Computing, Umm Al-Qura University, 21961, Makkah, Saudi Arabia.
Department of Software Engineering, College of Computing, Umm Al-Qura University, 21961, Makkah, Saudi Arabia.
Sci Rep. 2025 Apr 27;15(1):14730. doi: 10.1038/s41598-025-99837-5.
Fog-cloud computing has emerged as a transformative paradigm for managing the growing demands of Internet of Things (IoT) applications, where efficient task scheduling is crucial for optimizing system performance. However, existing task scheduling methods often struggle to balance makespan minimization and energy efficiency in dynamic and resource-constrained fog-cloud environments. Addressing this gap, this paper introduces a novel Task Scheduling algorithm based on a modified Grey Wolf Optimization approach (TS-GWO), tailored specifically for IoT requests in fog-cloud systems. The proposed TS-GWO incorporates innovative operators to enhance exploration and exploitation capabilities, enabling the identification of optimal scheduling solutions. Extensive evaluations using both synthetic and real-world datasets, such as NASA Ames iPSC and HPC2N workloads, demonstrate the superior performance of TS-GWO over established metaheuristic methods. Notably, TS-GWO achieves improvements in makespan by up to 46.15% and reductions in energy consumption by up to 28.57%. These results highlight the potential of TS-GWO to effectively address task scheduling challenges in fog-cloud environments, paving the way for its application in broader optimization tasks.
雾云计算已成为一种变革性范式，用于管理物联网（IoT）应用不断增长的需求，其中高效的任务调度对于优化系统性能至关重要。然而，在动态和资源受限的雾云环境中，现有的任务调度方法往往难以平衡完工时间最小化和能源效率。为了弥补这一差距，本文介绍了一种基于改进灰狼优化方法（TS-GWO）的新型任务调度算法，该算法专门针对雾云系统中的物联网请求量身定制。所提出的TS-GWO纳入了创新算子，以增强探索和利用能力，从而能够识别最优调度解决方案。使用合成数据集和真实世界数据集（如美国国家航空航天局艾姆斯实验室的iPSC和HPC2N工作负载）进行的广泛评估表明，TS-GWO的性能优于既定的元启发式方法。值得注意的是，TS-GWO可将完工时间最多缩短46.15%，并将能耗最多降低28.57%。这些结果凸显了TS-GWO有效应对雾云环境中任务调度挑战的潜力，为其在更广泛的优化任务中的应用铺平了道路。
