Date on Master's Thesis/Doctoral Dissertation


Document Type

Master's Thesis

Degree Name



Computer Engineering and Computer Science

Committee Chair

Ouyang, Ming

Author's Keywords

GPU parallel programming; Parallel NIDS; CUDA; NIDS GPU; Snort GPU; CUDA NIDS GPU


Computer networks--Security measures


As network speeds continue to increase and attacks get increasingly more complicated, there is need to improved detection algorithms and improved performance of Network Intrusion Detection Systems (NIDS). Recently, several attempts have been made to use the underutilized parallel processing capabilities of GPUs, to offload the costly NIDS pattern matching algorithms. This thesis presents an interface for NIDS Snort that allows porting of the pattern-matching algorithm to run on a GPU. The analysis show that this system can achieve up to four times speedup over the existing Snort implementation and that GPUs can be effectively utilized to perform intensive computational processes like pattern matching.