Dynamic Real-Time Scheduling on Distributed Hierarchical Fog Networks

We present a novel approach for efficient task scheduling on hierarchical fog nodes, catering to real-time (RT) and non-real-time (NRT) tasks with varying sizes and deadline constraints. Leveraging machine learning (ML) techniques, our proposed solution autonomously allocates tasks across fog nodes, dynamically adapting to changing workload patterns and system conditions. Our novel methodology integrates supervised learning algorithms to predict workload patterns and resource availability, enabling intelligent decision-making in task assignment. Our findings contribute to advancing the state-of-the-art in fog computing by offering a scalable and adaptive solution for dynamic task scheduling in hierarchical heterogeneous environments.