摘要
文章以经典的多模式匹配算法—AC算法为例,通过对CUDA特性的分析,提出了基于CUDA的并行模型,设计了适合CUDA并行技术的AC匹配算法。实验结果表明,基于CUDA的AC匹配算法较CPU上获得了22倍的加速比,有效提高了入侵检测系统的性能。
By analyzing the characteristic of CUDA, a parallel module of AC matching algorithm based on CUDA is proposed. Experiment shows that the AC multi-pattern matching algorithm based on GPU gets 22 times speedup ratio than it based on CPU, and it improves the performance of intrusion detection system effectively.
出处
《信息网络安全》
2011年第9期126-128,共3页
Netinfo Security
关键词
CUDA
并行技术
多模式匹配
AC算法
CUDA
parallel technology
multi-pattern matching
AC algorithm
