基于矢量化加速的网络安全应用多模式匹配  

Multi-pattern Matching of Network Security Applications Based on Vectorization Acceleration

作　　者：刘文涛 LIU Wen-tao(School of Computer and Information,Anhui Normal University,Wuhu 241000,China)

机构地区：[1]安徽师范大学计算机与信息学院,安徽芜湖241000

出　　处：《西安文理学院学报（自然科学版）》2022年第4期11-16,共6页Journal of Xi’an University(Natural Science Edition)

基　　金：2021年安徽省自然科学基金面上项目(2108085MF206):“基于深度字典学习的低质指纹识别关键技术研究”。

摘　　要：为提高网络安全应用多模式匹配效率,设计了一个基于矢量化加速的网络安全应用多模式匹配方法.首先对网络安全应用多模式匹配进行了预处理,然后为避免匹配失败时不必要的回溯,引入下一个数组,根据模式串的特征计算下一个序列,最后提出除发生匹配不成功情况外基本的匹配过程,实现基于矢量化加速的网络安全应用多模式匹配.实验结果表明,所研究的基于矢量化加速的网络安全应用多模式匹配方法有效提高了匹配效率,并提高了匹配的成功率.In order to improve the efficiency of multi-pattern matching in network security applications,a multi-pattern matching method based on vectorization acceleration is designed.Firstly,the multi-pattern matching of network security applications is preprocessed.Then,in order to avoid unnecessary backtracking when matching fails,the next array is introduced,and the next sequence is calculated according to the characteristics of the pattern string.Finally,the basic matching process except for unsuccessful matching is proposed to realize multi-pattern matching for network security applications based on vectorization acceleration.The experimental results show that the network security application multi-pattern matching method based on vectorization acceleration effectively improves the matching efficiency and the success rate of matching.

关 键 词：矢量化加速 网络安全应用 多模式 匹配 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]