Metaheuristic-based task scheduling for latency-sensitive IoT applications in edge computing

The increasing amount of data produced by Internet of Things (IoT) devices imposes significant limitations on the resources available in conventional cloud data centers, undermining their capacity to accommodate time-sensitive IoT applications. Cloud-fog computing has emerged as a promising paradigm that extends cloud services to the network edge. However, the distribution of tasks in a cloud-fog environment presents new challenges. Our research paper introduces a semi-dynamic real-time task scheduling system designed explicitly for the cloud-fog environment. This algorithm effectively assigns jobs while minimizing energy consumption, cost, and makespan. An adapted version of the grey wolf optimizer is introduced to optimize task scheduling by considering various criteria such as task duration, resource requirements, and execution time. Our approach outperforms existing methods, such as genetic algorithm, particle swarm optimization, and artificial bee colony algorithm, in terms of makespan, total execution time, cost, and energy consumption.