A Memory Accessing Method for the Parallel Aho-Corasick Algorithm on GPU

Abstract

In this paper, we propose a memory accessing method of Parallel Failureless Aho-Corasick (PFAC) algorithm considering Graphic Processing Unit (GPU) memory architecture for throughput improvement. Compared with Aho-Corasick (AC) Algorithm using Central Processing Unit (CPU) and Data-Parallel Aho-Corasick (DPAC) using Open Multi-Processing (OpenMP), PFAC using GPU achieves high performance advancement in the string matching. However, due to the increasing network speed, there is need to increase the throughput of the string matching engine. In order to solve the problem of frequent memory accesses of the string matching engine based the AC algorithm, we propose the method of memory accesses for the PFAC algorithm. In the experiments using attack pattern from Snort and input stream from DEFCON on NVIDIA GTX 660. Our memory accessing method achieves about 10% improvement of throughput, compared with the original PFAC algorithm.