Deadline-Aware Task Scheduling for Cloud-Fog Systems

The integration of fog computing with cloud service reduces transmission times to ensure that Internet of Things (IoT) devices satisfy deadline-aware tasks. Existing researches focus on the effective reduction of task transmission delays in hierarchical fog-cloud networks. However, an improper task execution order increases the waiting time and the risk of missing deadlines. To address this problem, the concept of Deadline-Aware Task Offloading and Scheduling (DTOS) is introduced. To balance the load of all fog nodes, this research proposes an offloading strategy using a Genetic Algorithm that encodes multiple chromosomes with different offloading decisions for each task. Each evolution round refines the offloading positions, generating a strategy that approaches the optimal solution. Subsequently, the scheduling algorithm finds an effective execution sequence for each fog node based on the generated offloading decision. The simulation results demonstrate that the proposed strategy is more efficient than the baseline algorithms compared in this research.