EDFS: Election-based decentralized fog scheduling for IoMT

In recent times, researchers have been interested in developing fault-tolerant decentralized scheduling algorithms for fog networks, primarily focusing on orchestration for task offloading. In light of these challenges, incorporating a leader election scenario proves advantageous, as it eliminates the need for controller setup in an asynchronous environment. This work introduces Election-Based Decentralized Fog Scheduling for IoMT (EDFS), a leader election algorithm tailored for the Internet of Medical Things (IoMT) scenario. EDFS features an integrated fog computing setup and operates under realistic (weak) assumptions regarding communication synchrony and message delays. This work also involves fault tolerance of fairness mechanism. The proposed algorithm is conditionally validated and proves self-sufficient and easy to implement to extend support toward different IoMT setups. EDFS showcases its effectiveness in resource utilization and fault tolerance methods compared to existing models. The study reveals a consequential performance enhancement of 17.38% in score comparison and a significant 36.23% improvement in CPU utilization among fog nodes, surpassing other models examined in this research.