基于OpenFlow的报文分类算法研究与实现  

Study and implementation of packet classification algorithms for OpenFlow

作　　者：吕昭[1] 李韬[1] 

机构地区：[1]国防科学技术大学计算机学院,湖南长沙410073

出　　处：《计算机工程与科学》2014年第5期860-865,共6页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(61202483)

摘　　要：随着软件定义网络、OpenFlow等技术的兴起,传统的基于5元组的报文分类技术已不能满足OpenFlow基于多元组的细粒度流量控制需求。因此,以分析已有的报文分类算法为基础,采用分而治之的思想,针对OpenFlow报文分类的精确匹配需求,设计实现了一种基于Hash的计数型链表Bloom Filter算法——OF_CBF算法。针对OpenFlow报文分类的通配匹配需求,借鉴正则表达式匹配算法思想,设计实现了基于有限自动机的报文匹配算法——OF_FSMP算法。对两种算法进行分析验证,并初步对两种算法进行了性能分析。With the emergence of software-defined network and OpenFlow in recent years, fine- grained flow control based on the multi-tuple packet classification makes a huge challenge to the tradi- tional 5-tuple packet classification. The basic concept of packet classification, some typical packet classi- fication algorithms and the packet classification for OpenFlow are introduced. For the demand of exact match in the OpenFlow packet classification, a counting and linked list Bloom Filter base on Hash, named OF CBF, is designed and implemented. And it is tested and analyzed. For the demand of wild- card match in the OpenFlow packet classification, an algorithm based on finite state machine, named OF FSMP is designed, implemented, tested and analyzed.

关 键 词：网络处理器 报文分类 OpenFlow BLOOMFILTER 有限自动机 

分 类 号：TP393.02[自动化与计算机技术—计算机应用技术]