一种高速多维报文分类算法

报文分类作为一种核心技术,在消息处理,内容过滤,Web信息检索和支持QoS的路由器等领域有重要应用。RFC算法是目前具有代表性的一种报文分类算法,具有匹配速度快,占用内存少、支持范围匹配等优点,但是它不能处理具有变长字符串域。结合多模式匹配算法的思想,对RFC算法进行了有益扩充,使新算法能够根据变长字符串域进行分类。试验结果表明,改进的算法在有效扩展KFC算法的适用范围的基础上,仍具有良好的性能。

Fast packet classification algorithm applying to multiple fields

Packet classification is a key technology which is widely used in many areas such as message handling,content filtering,web information indexing and router supported with Qos.The recursive flow classification(RFC) algorithm proposed by P.Gupta and N.McKeown is a typical flow classification algorithm.It performs classification very quickly,consumes less memory,and supports scope matching,but it can't perform classification according to variable length string fields.Combined with the idea of Aho-Corasick multi-pattern matching algorithm,the RFC algorithm is improved;And the improved algorithm can effectively deal with the classification of variable length string fields.The experimental result illustrates that it not only extends the application scope of RFC algorithm but also has good performance.