An Architecture for QoS-Aware Fog Service Provisioning

The proliferation of IoT technologies and the broad deployment of sensors and IoT devices are changing the way and the speed of delivery of many services. IoT data streams are typically transmitted to cloud services for processing. However, time-sensitive IoT applications cannot tolerate high latency they may experience when IoT data streams are sent to the cloud. Fog computing-based solutions for this kind of applications are becoming more and more attractive due to the low latency they can provide and guarantee. Given the growing deployments of fog nodes, we propose in this paper an architecture for quality of service (QoS) aware fog service provisioning, which allows to schedule the execution of IoT applications' tasks on a cluster of fog nodes. A fog broker component can implement various scheduling policies to help IoT applications fulfill their QoS requirements. The results of the simulations we performed show that using some simple strategies, it is possible to keep low the latency of applications and distribute the load among the fog nodes of the cluster.