Implementation of Fog Nodes in the Tree-Based Fog Computing (TBFC) Model of the IoT

Abstract

The IoT (Internet of Things) is so scalable that not only computers like servers but also sensors and actuators installed in various things are interconnected in networks. In the cloud computing model, application processes to process sensor data are performed on servers, this means networks are congested and servers are overloaded to handle a huge volume of sensor data. The fog computing model is proposed to efficiently realize the IoT. Here, subprocesses of an application process are performed on not only servers but also fog nodes. Servers finally receive data processed by fog nodes. Thus, traffic to process sensor data in severs and to transmit sensor data in networks can be reduced in the fog computing model. In this paper, we take the tree-based fog computing (TBFC) model where fog nodes are hierarchically structured in a height-balanced tree. We implement types of subprocesses of fog nodes in Raspbery PI. In experiment of the implemented TBFC model, we show the total execution time of nodes in the TBFC model is shorter than the cloud computing model.