高速以太网中实时相似性匹配算法的研究

在高速以太网中, 针对用户行为的特殊性及数据流五元组的特点, 提出了一种自适应相似性匹配算法(ASMA-HN)。该算法充分考虑了数据流五元组的随机性分布特点, 通过一种改进的XOR-SHIFT算法将数据流分类索引, 并提出了同类数据库(SD)及异类数据库(HD)的概念, 基于距离测度分别设计了加权相似性比较函数, 使得特征值能更精准地匹配到符合要求的数据文件, 提高了整体的匹配效率。通过对校园网实际流量的现场测试, 证明该算法具有较高的精准性及可适用性, 且匹配时间较低, 能很好地满足高速网络中相似性匹配业务的需求。

Research on similarity mapping algorithm for high-speed Ethernet

This paper proposed an adaptive similarity mapping algorithm for high-speed Ethernet based on the special nature of user behavior and characteristics of the flow quintuple. It fully considered the random distribution of the quintuple in this algorithm and designed an improved XOR-SHIFT algorithm for flows classification. This paper raised concepts of similar database and heterogeneous database and designed similarity comparison functions for this two databases. With this method, similar eigenvalues can be mapped to data files precisely and matching efficiency is improved. It carried out some simulation tests in the campus network and proved that this mapping algorithm has higher precision and applicability than other methods. It is appropriate to search similarity flows in large data networks.