An Adaptive and Efficient Scheduling Framework for Decentralized Fog Environments: Design and Problem Formulation

The increasing demand for low-latency and energy-efficient processing in Internet of Things (IoT) applications has highlighted the limitations of centralized scheduling in fog computing environments. This paper proposes an adaptive and efficient scheduling framework for decentralized fog computing, where local schedulers at each fog node make real-time decisions based on dynamic resource conditions. A cooperative scheduling mechanism among neighboring nodes is introduced to improve resource utilization and reduce cloud dependency, with the cloud used only as a fallback layer. The scheduling problem is formulated as a multi-objective optimization model considering latency, energy consumption, and load balancing. An adaptive cost-based approach with learning-driven weight adjustment enables the system to dynamically respond to changing workloads. The proposed framework provides a scalable, decentralized, and efficient solution for fog computing environments.