Energy-Efficient Task Scheduling in IoT using a Hybrid Genetic Algorithm with Local Search
N. Sandhya, Seshendranath Balla Venkata, Saef Thallal, М. Лакшманан, Mukesh Soni
Abstract
In recent years, energy-aware scheduling has gained significant attention in the Internet of Things (IoT) and distributed computing systems with the aim of minimizing power consumption and maintaining performance and service quality. However, the existing bio-Inspired Energy Efficient Dynamic Task Scheduling (BEDTS) approach based on Adaptive Elephant Herding Optimization (AEHO) faced a lack of adaptability and optimization for IoT-fog-cloud systems to its convergence and cloud-centric design. Hence, this research proposes an Energy Aware Scheduling architecture for IoT environments based on a Genetic Algorithm enhanced by Local Search (GA-LS). In this research, an IoT system model is considered where heterogeneous IoT devices, fog nodes, and cloud resources are interconnected. Initially, tasks were dynamically generated by IoT devices and categorized according to their data size, deadline, and priority level. Both the processing and transmission energies are quantified using an energy model. In the scheduling process, GA-LS is employed to encode task-to-node mappings; evolve solutions through selection, crossover, and mutation; and apply a local search to further minimize energy consumption while maintaining deadlines and resource constraints. Therefore, the proposed GA-LS method achieved a 97.0 % of task guarantee ratio, 1.35 s an average response time, 7.04 an energy consumption and a resource utilization of 98.0 %.