Job Scheduling on Fog Computing Infrastructure based on QoS Parameters

Abstract

The increasing demand for cloud computing in the Internet of Things (IoT) device ecosystem has highlighted the need for a more suitable infrastructure. Edge and Fog environments have emerged as viable options, as they allow for computation to be performed closer to the relevant resources, thereby reducing latency. The dynamic nature of IoT traffic and the demand for low latency services further underscore the importance of efficient service placement, which is facilitated by Fog computing through the implementation of load balancing. In this study, the authors focus on two critical parameters for IoT devices: load balancing and Quality of Service (QoS) improvement. To address these concerns, the authors propose a decentralized, agent-based solution that uses edgeto- cloud nodes to cooperatively manage the network's input load. The solution utilizes agents that generate potential request paths for resource allocation and choose the path that maximizes edge usage while minimizing service execution cost. The proposed solution is expected to effectively balance the input load throughout the network and reduce the costof- service execution in the IoT device ecosystem.