Abstract
Abstract Network Intrusion Detection System (NIDS) has been widely used to protect computer systems from network attacks. Due to the ever-increasing number of attacks and network complexity, traditional software approaches on uni-processor become inadequate for the high-speed network. Graphics Processor Unit (GPU) has attracted a lot of attention due to its dramatic power of massive data parallel computing. In addition to graphic applications, GPU has achieved substantial progress than general-purpose CPUs for a range of non-graphical applications such as science and physics computations, electronic design automation, bioinformatics, and so on. In this paper, we propose a novel algorithm to speedup pattern matching on GPU. The experimental results show the new algorithm on GPU can achieve a significant speedup compared to the conventional Aho-Corasick algorithm on CPU for total Snort string patterns. In addition, the new algorithm also has improvement on memory requirements.