Research And Realization On Network Traffic Classification Based On Octeon Multi-core Network Processor

With the rapid development of Internet technology, the need for network security is more and more urgent. Network traffic classification technology is playing an increasingly important role in network management, service quality assurance and network security as the basic technology and important means to realize network controllability.In this paper, the existing network traffic classification techniques are summarized, and their advantages and disadvantages based on port number matching, feature field analysis, transport layer behavior model and flow statistics based machine learning methods are summarized respectively. In view of the advantages of high accuracy and scalability of network traffic classification based on machine learning algorithm, random forest was selected as the classification method of network traffic, and improved strategy was put forward for traditional random forest. Finally the random forest strategy based on improved to OCTEON platform, the network traffic classification system based on OCTEON multi-core processor is designed and implemented. Finally, a system test environment is set up,and the system is validated.The main work has the following aspects:1. Based on the basis of fully understanding the traditional stochastic forest algorithm, according to the deficiency of single node model training and the difference of strength and weakness of the basic classifier, this paper proposes an improved stochastic forest from the aspects of parallelism and weighted adjustment.2. The OCTEON operating mode, inter-core synchronous communication mechanism and memory allocation mechanism are combined with the random forest strategy of the modified machine to realize the improved random forest model on OCTEON platform.3. The architecture of the network traffic classification system based on OCTEON multi-core processor is realized. The architecture design, the detailed design and realization of each function module are introduced in detail. The system test environment is built up. Traffic classification of the excellent performance.