Research And Implementation Of String Matching Algorithm In Peer-To-Peer Traffic Detection

The appearance of peer-to-peer network evolves the network architecture of traditional C/S mode. Every network node plays the same role in P2P network, they not only obtain service from other nodes as clients, but also provide service for other nodes as servers. P2P network makes it possible to excavate network resources deeply, and also presents a prospective future of computer network. So peer-to-peer network is a hot point in the research field of Internet currently.P2P is now widely used as an important application of Internet. Statistics show that peer-to-peer network traffic consumes the forty percent or seventy percent of the total network traffic of ISPs, and it becomes the most dangerous killer of network bandwidth. Now with the increasing prevalence of Internet and increasing complexity of network architecture, network security, management and the use of the traditional service has encountered a challenge, therefore research, detection, analysis and management about peer-to-peer traffic is very necessary, since ISPs would adjust the management strategy in time in order to accommodate the characteristic of peer-to-peer network, which is one of the pressing tasks of ISPs. Among those tasks, the most urgent task is the effective identification and management of peer-to-peer traffic. This paper is based on the project "research on traffic detection and management of P2P application"; it is the summarization about some project results. The paper presents research on features of peer-to-peer network, factors that affect the performance of string matching algorithm and performance emulation results and comparison of some string matching algorithms.The paper consists of five chapters. Chapter One provides a brief introduction about P2P technology and the research work of author completed in the project; Chapter two mainly covers the research status about peer-to-peer network and its development trend; Chapter three analyses features of some typical peer-to-peer network thoroughly, and presents some results of these P2P application on feature analysis; Chapter four describes related theory about string matching algorithm, analysis and experiment research of some classical and modified string matching algorithms. Emulation on performance comparison of some string matching algorithms and typical P2P application are given. Chapter five generalizes hardware-based peer-to-peer traffic identification technique, and finally summarizes the research work in the paper and a prospect of P2P research in the future.