作者: Jingjiao Li , Yong Chen , Cholman Ho , Zhenlin Lu , None
DOI: 10.1109/ICOIN.2013.6496433
关键词:
摘要: In the network intrusion detection system (NIDS), there is a limitation on speed of software-based packet classification because processor performance, serial program execution and so on. It has become great challenge to develop scalable solutions for next-generation that support higher throughput, larger rule sets more header fields. For low-cost high performance embedded networking applications, best solution could be doing by special designed hardware, which can effectively release burden CPU. order improve classification, exhibit good memory quick update, high-speed based FPGA proposed in this paper. Taking advantage parallel processing, pipeline hardware circuit, throughput been improved greatly; defining size tree nodes binary tree, usage efficient. The structure generated through pre-processing computer, does not influence searching FPGA. During division, division field dynamic selected according rules. experimental results show time 50000 rules shorter than 0.051s, average rule-header Snort IDS 10 Gbps.