Optimization And Implementation Of Netflow Generation System Based On Online Clustering

With the rapid growth of Internet traffic, effective management of the Internet has become an important issue. Network traffic statistics provide network management and maintenance important information, which are very important for network architecture planning, network performance improvement and network security. So the Internet traffic monitoring and analysis of the Internet has become a prerequisite for effective management of the Internet.This subject comes from the project the data set generation technology research based on playback and annotation of NetFlow, which discusses the implement of extensible NetFlow generation system based on IPFIX protocol. It resolves problems such as serious packets loss of current flow-based network traffic monitoring, online clustering for high-speed network, single for NetFlow output format, less information containedand so on.The paper first introduces the basic concepts of NetFlow, extensible NetFlow protocol IPFix, high-speed networks capture technology PF_ring, clustering and other related technologies. Then it describes the architecture of extensible NetFlow generation system, and describes the network packet capture module, the network packet streaming module, the NetFlow clustering module, the NetFlow export module and the NetFlow collection module. Then it gives a detailed description of the clustering module, introduces the basic concepts involved in the algorithm, and the framework of the algorithm, and the function, significance of use and implement of each sub-function. The paper also describes the test for the system, and the test results show that the system can achieve the desired objective. Finally, the paper summarizes the research work and the future prospect.