北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2006, Vol. 29 ›› Issue (s2): 45-48.doi: 10.13190/jbupt.2006s2.45.306

• 论文 • 上一篇    下一篇

基于决策树的递归包分类算法

张艳军1,2,陈友1,2, 郭莉1, 程学旗1   

  1. 1. 中国科学院 计算技术研究所, 北京 100080; 2. 中国科学院 研究生院, 北京 100039
  • 收稿日期:2006-09-02 修回日期:1900-01-01 出版日期:2006-11-30 发布日期:2006-11-30
  • 通讯作者: 张艳军

A Recursive Packet Classification Algorithm Based on Decision Tree

ZHANG Yan-jun1,2, CHEN You1,2, GUO Li1, CHENG Xue-qi1   

  1. 1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China;
    2. Graduate University, Chinese Academy of Sciences, Beijing, China
  • Received:2006-09-02 Revised:1900-01-01 Online:2006-11-30 Published:2006-11-30
  • Contact: ZHANG Yan-jun1

摘要:

提出了一种新的包分类算法SRC(sensitive recursive classification).它建立在决策树基础之上,在以防火墙, 访问控制列表为种子的规则库中进行实验.实验结果表明:SRC内存使用比Hicuts (hierarchical intelligent cuttings)减少3~10倍,最坏查找速度比Hicuts提高5倍以上;SRC的内存使用比EGT-PC(extended grid-of-tries and path compression)减少2~8倍,最坏查找速度比EGT-PC提高4倍以上.

关键词: 包分类, 决策树, 映射

Abstract:

This paper introduces a classification algorithm called SRC(sensitive recursive classification). It is based on a decision tree structure. Doing many experiments, especially in FW(firewall) and ACL(access control list), we verified that SRC uses 3 to 10 times less memory than HiCuts(hierarchical intelligent cuttings), while the worst case search time is up to 5 times smaller. Compared with EGT-PC(extended grid-of-tries and path compression), SRC uses 2 to 8 times less memory while the worst case search time is up to 4 times smaller.

Key words: packet classification, decision tree, mapping

中图分类号: