Abstract:The classification of IPv6 network traffic components is helpful to better control the IPv6 network traffic behavior and application behavior. This research work is based on the original traffic of three campus networks connected to CERNET and CERNET2 Nanjing master node network. Apart from the preliminary identification of traffic with the open source software nDPI, a P2P traffic classifier is designed and implemented to identify the traffic that can not be recognized by nDPI. The test results show that the classifier can greatly reduce the proportion of unknown traffic in IPv6 traffic. On this basis, an existing encrypted video classifier is used to identify the video traffic in the HTTP/TLS unknown application layer protocol stream. The results show that the P2P component in the traffic is the main reason for the difference of traffic components between IPv4 and IPv6 networks. Finally, the differences of P2P applications in the two types of networks are also given.

IPv6 Network Traffic Component Analysis in Campus Network