| Summary: | 碩士 === 國立清華大學 === 資訊工程學系 === 95 === By the development of network applications, network security issues are getting more and more important. The technologies of deep packet inspection are also in progress. There are multiple fragment shaders doing graphics processing jobs simultaneously in the graphics processing unit (GPU), this characteristic is used to speed up the pattern matching part of deep packet inspection. In this thesis, a Wu-Manber like pattern matching algorithm is proposed to process the content string of a single packet in parallel. The data structures used in the proposed are transformed into the data structures executed in the GPUs, and the pattern matching task is performed by the GPU programming related languages. The performance of the proposed approach is analyzed through evaluations such as using various texture formats and different implementations. In the experimental result, the performance of the proposed approach is twice the performance of the modified Wu-Manber algorithm used in Snort. The proposed approach can make a commodity graphics card with the market price less than US$200 a feasible pattern matching co-processor.
|
|---|
