Developing a deep learning based approach for traffic engineering in software defined networks

Traffic engineering is essential for network management, particularly in today's large networks carrying massive amounts of data. Traffic engineering aims to increase the network's efficiency and reliability through intelligent allocation of resources. In this thesis, we propose a deep learning-based traffic engineering system in software-defined networks (SDN) to improve bandwidth allocation among various applications. The proposed system implements traffic classification based on deep neural network (DNN) and one dimensional convolution neural network (1-D CNN) models, and implements a traffic engineering module on the SDN controller to direct traffic flows in the network according to traffic classes by considering the Quality of Service. The system aims to improve the Quality of Service (QoS) by identifying flows from various applications and distributing the identified flow to multiple queues where each queue has a different priority. Next, it applies traffic shaping in order to manage network bandwidth and the volume of incoming traffic. To increase the network's performance and avoid traffic congestion, we implement a technique that considers the port capacity to accomplish general load balancing. We solved the issue of imbalanced dataset by implementing an oversampling technique called Synthetic Minority Over-Sampling Technique (SMOTE). The performance of DNN and 1-D CNN have been compared and evaluated with some of machine learning models, such as KNN, SVM, DT, and RF. The results showed 1-D CNN and DNN are able to achieve high accuracy of traffic captured at 5s and 10s timeout, while KNN and RF are able to achieve high accuracy of traffic captured at 15s, 30s, 60s and 120s timeout, while CNN and DNN were able to achieve better accuracy than DT and SVM at 15 s, 30 s, 60 s and 120 s timeout. Different scenarios have been used to evaluate the QoS-based traffic engineering system. The results showed that applying traffic shaping to identified flows increases network performance by regulating of bandwidth usage helps in guaranteeing resources for each type of application and bandwidth availability by controlling the rate and volume of incoming flows.