Software-Defined Networking (SDN) Traffic Analysis Using Big Data Analytic Approach

Abstract

Network Traffic Analysis is an essential component of network management, especially for ensuring the proper operation of large-scale networks like the Internet. Monitoring and analyzing network traffic has become increasingly difficult due to the growing complexity of Internet services and the volume of network traffic. The development of the network concept with a virtualization approach such as Software-Defined Networking provides new problems in monitoring and analyzing traffic on the network. The SDN concept that distinguishes data flow and control makes network traffic data different from conventional networks. On the other hand, applications such as traffic classification and policy making in computer networks require a real-time approach and scalability. Anomaly detection and security techniques must rapidly recognize and respond to unanticipated events while processing millions of heterogeneous events extracted from computer network traffic data. Finally, the system must gather, store, and process enormous historical data sets in preparation for analysis. Volume, Velocity, Variety, and Veracity are challenges that must be faced in managing traffic analysis. This study explains how traffic analysis on SDN networks and Big Data analytics are combined to take full advantage of the potential of network data. This study aims to discuss the extent to which traffic analysis can take advantage of the potential of Big Data technology and describe the challenges and opportunities of using Big Data and machine learning technology for traffic analysis. A prototype approach is used to build Big Data Analytics architecture and apply machine learning methods for data analysis. Based on the result big data approach can be used to classify attack traffic on SDN networks. The results of the train scores and test scores on the classification using the decision tree are as follows: the training score is: 0.998265782638848, and the Test score is: 0.9982486670135102.