FoGMatch

Abstract

Cloud computing has long been the main backbone that IoT devices rely on to accommodate their storage and analytical needs. However, the fact that cloud systems are often deployed in locations that are quite far from the IoT devices and the emergence of delay-critical IoT applications (e.g., health monitoring, real-time machine learning, etc.) urged the need for extending the cloud architecture to support delay-critical services. In this context, the notion of fog computing has been projected to furnish data analytics and decision-making closer to the IoT devices. Given that fog nodes are characterized by small resource capabilities compared to the cloud systems, the problem of matching the IoT services to the appropriate fog nodes while guaranteeing minimal delay for the IoT services and efficient resource utilization on the fog nodes becomes quite challenging. Several approaches have been proposed in the literature in an attempt to address this challenge. The main limitation of these approaches is that they address the scheduling problem from one side point of view, i.e., either fog nodes or IoT devices. This results in an unfair situation wherein the needs of one of the parties are ignored in the scheduling process. To address this problem, we propose in this paper a multi-criteria intelligent IoT scheduling approach in fog computing environments using matching game theory. Our solution consists of (1) two optimization problems, one for the IoT devices and one for the fog nodes, (2) preference functions for both the IoT and fog layers to help them rank each other on the basis of several criteria such latency and resource utilization, and (3) centralized and distributed intelligent scheduling algorithms that consider the preferences of both the fog and IoT layers to improve the performance of the overall IoT ecosystem. Simulation results reveal that our solution outperforms the two common scheduling algorithms (i.e., Min-Min and Max-Min) in terms of IoT services execution makespan, fog nodes resource utilization efficiency and execution time.