Graption: A graph-based P2P traffic classification framework for the internet backbone

摘要: Monitoring network traffic and classifying applications are essential functions for administrators. Current classification methods can be grouped in three categories: (a) flow-based (e.g., packet sizing/timing features), (b) payload-based, (c) host-based. Methods from all categories have limitations, especially when it comes to detecting new applications, at the backbone. In this paper, we propose use of Traffic Dispersion Graphs (TDGs) remedy these limitations. Given a set flows, TDG is graph with an edge between any two IP addresses that communicate; thus TDGs capture network-wide interactions. Using TDGs, develop application framework dubbed Graption (Graph-based classification). Our provides systematic way classify by using information behavior flow-level characteristics Internet applications. As proof concept, instantiate our detect P2P traffic, show identify 90% flows 95% accuracy backbone traces, which particularly challenging other methods.