Uma comparação entre controladores de redes definidas por software em cenários de internet das coisas

Abstract

The Internet of Things (IoT) is a promising paradigm in making it possible to connect common objects to the Internet. However limitations of memory capacity as well as processing of these objects arise. Also, the management and optimization of network performance and IoT programmability have become complex. Because of this, a new approach to integrating IoT networks, such as Software Defined Networks (SDN), is needed as this architecture allows for greater flexibility in separating the data plane from the control plane. The overall objective of the paper is to compare SDN controllers: Ryu; POX and Floodlight; in IoT scenarios, and thus highlight which controller did the best with respect to the metrics: latency; memory; processor consumption; and execution time of the experiments. It is expected that the present work can be used by researchers or students in the field in the process of selecting the SDN controller to be used in studies or tests in IoT environments. One of the results obtained was that the Floodlight controller presented the lowest average latency with up to 104 IoT nodes.