摘要
当前深度包检测算法通常需要将正则表达式转换为NFA或者DFA,但是随着网络带宽的不断增加,NFA和DFA状态占用的存储空间越来越大,极大地考验着系统的存储能力。为了应对这个问题,提出一种基于正则表达式相性的分组算法来对表达式进行分组,实验证明该算法能减少NFA和DFA状态的数量,提高匹配的效率。
The current Deep Packet Inspection (DPI) algorithm usually need to transform regular expres- sions into NFA or DFA. But with network bandwidth's increase, states of NFA and DFA take more and more memory space. It tests greatly system's memory capacity. In order to deal with this problem, puts forward a new concept: regular expression compatibility. And we divide regu- lar expressions into groups through a new grouping algorithm. The experiment shows it could re- duce the number of states of NFA and DFA and improve matching efficiency.
出处
《现代计算机》
2012年第12期3-5,9,共4页
Modern Computer