Streaming via SDN: Resource forecasting for video streaming in a Software-Defined Network

Abstract

With the advancement in network devices and the proliferation of new technologies such as Software-Defined Networking (SDN), managing a network becomes more difficult. In an SDN network, a single physical device acts as a firewall and load balancer at the same time. The management of those devices and the prevention of the resources being exhausted is a challenging task for the network administrator. In this direction, this paper presents an approach to predict resources on a switch in an SDN-based network. For this purpose, a video streaming scenario is deployed in an SDN network and performance metrics are captured. The resources are predicted using four machine learning algorithms. Specifically, the paper proposes a testbed implementation of a video streaming scenario to evaluate the performance of the proposed approach. The proposed approach can help network operators optimize network performance, ensure efficient use of resources, and enhance user experience.