Modelling resource-aware fault tolerant task scheduling in dynamic fog environment: A tournament selection Approach

The contemporary computing landscape has witnessed the drift of services extended by the cloud to the network edge through the fog computing paradigm. The proximity to data sources and end users offers many advantages, including significantly reduced latency, enhanced real-time processing, and improved privacy. In this context, the orchestration of computational tasks through effective task scheduling becomes pivotal for maximizing resource utilization efficiency, latency minimization, and optimizing the overall performance of system. This work proposes a learning automata-based task scheduling technique for dynamic fog environments. The objective of the proposed work is to minimize latency and improve the responsive of the applications with stringent latency requirements. The proposed work uses the Tournament Selection algorithm for selecting the fog nodes. The model considers two critical factors as the foundation for fog node assignment for task execution, namely, residual energy and memory constraints. The learning automata enables the system to make informed and reliable decisions regarding allocating tasks to fog nodes, making the system fault tolerant. The experimental results demonstrates that the proposed approach has better performance than the existing scheduling strategies in reducing task response times, memory usage and energy consumption. The work proposed contributes significantly to the ongoing evolution of fog computing, highlighting the potential for substantial performance improvements and underlining the importance of resource-aware scheduling in the era of edge-enhanced computing.