隨著科技進步，一些需求也隨之產生，如衛星成像、基因工程、全球氣候預報、核爆模擬等，數據量已經到達TB甚至PB，沒有相對的高速計算處理能力是無法解決的，而且日常生活中對於影像應用（遊戲、高畫質影片播放）的要求越來越高，所面臨的圖形和數據計算也越來越複雜，對於計算速度來說是個嚴峻的挑戰。
受到上述科技上所遇到的問題以及日常生活所需，GPU的性能提高速度很快，最近幾年來，GPU的性能每一年就可以成倍的成長，已經遠遠超過CPU遵照摩爾定律的發展速度，而目前主流GPU的單精度浮點運算能力已經到達了同期CPU的十倍左右，其中NVIDIA在2007年發表的CUDA是第一種不需借助圖形學API就可以使用類C語言進行通用計算的開發環境和軟體體系。
本研究就是利用GPU的優勢處理速度來協助處理日益龐大的網路封包流量，網路日漸普及，相對於此，各種不同的網路攻擊，或者因為網路而造成的安全問題（如個人隱私資料、公司機密資料保護）也就更受到大家的重視，如何在不影響傳輸速度下快速的處理各種的網路封包將有危害的封包排除在外就變成了相對重要的問題。

Several demands relied on high-performance computing come up with the advanced technologies, like Satellite Imaging, Genetic Engineering, Global Weather Forecast, Nuclear Explosion Emulation, and in the meantime, the amount of data usually approaches the rank of Tera-Bytes, even Peta-Bytes. Besides, we need practical image application in our daily life, such as Game, 3D Display, High-Definition Video, etc. These requirements of high-performance computing are rigorous challenge to current devices.
The performance of GPU (Graphic Processing Unit) is growing up rapidly in recent years. GPU doubles its computing power every year, which is far superior to CPU (Central Processing Unit) performance based on Moore’s Law. Nowadays, the computing power of GPU on the single-precision floating-point operations is ten times than that of CPU. Furthermore, CUDA (Compute Unified Device Architecture) is a parallel computing architecture proposed by NVIDIA at 2007, and it is the first C-like language software development environment without Graphics API.
In this research, we use GPU to assist network devices in filtering packets of the network flow, whose quantity is becoming more and more large. Due to the popularization of network, people pay attention to different types of network attacks or safety problems. Therefore, it is important to remove malicious packets from normal ones without degrading the network performance.

Abstract II
Outline III
Graph Catalogue V
1. Introduction 1
1.1. The motivation and objective 1
1.2. Thesis’s Structure 2
2. Introduction of GPU and CUDA 3
2.1. Summary and components of GPU 3
2.1.1. Host 3
2.1.1.1. CUDA (The Integrated software development environment) 3
2.1.2. Device 3
2.1.2.1. The Development of GPU 3
2.1.2.2. Introduction to components of device 5
2.2. Flow of CUDA processing 5
2.3. The speed of transmission between Host and Device 7
2.4. CUDA Thread Model 7
2.5. The Memory Model of GPU 8
2.6. Executing Model of GPU 10
3. IPS and Pattern-Matching Algorithm 11
3.1. Summary of IPS 11
3.2. Wu-Manber, A Fast Algorithm For Mutli-Pattern Searching 12
3.2.1. Implementation of Wu-Manber on C Language 13
3.2.2. Implementation of Wu-Manber on CUDA 14
3.2.3. The Packets 15
3.2.4. The Mechanism of parallel computing 16
3.2.5. Problem and Improvement 17
4. Implementation of pattern matching system and Performance Analysis 18
4.1. Design of the system 18
4.2. Flow of packets transmitting 18
4.3. Comparison of Performance between CPU/GPU 20
4.4. Number of patterns to the Performance 23
5. Conclusion 24
Reference 25

[1] Sun Wu, and Udi Manber “A FAST ALGORITHM FOR MULTI-PATTERN SEARCHING”, (May 1994)
[2] Aho, A. V., and M. J. Corasick, ‘‘Efficient string matching: an aid to bibliographic search,’’ (June 1975),
[3] Altschul S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman, ‘‘Basic local alignment search tool,’’ (1990)
[4] Baeza-Yates R. A., ‘‘Improved string searching,’’ Software — Practice and Experience (1989)
[5] Boyer R. S., and J. S. Moore, ‘‘A fast string searching algorithm,’’ (October 1977),
[6] Commentz-Walter, B, ‘‘A string matching algorithm fast on the average,’’ (1979)
[7] Tom R. Halfhill,” PARALLEL PROCESSINGWITH CUDA, Nvidia’s High-Performance Computing Platform Uses Massive Multithreading”(2008)
[8] Nen-Fu Huang+,*, Hsien-Wei Hung*, Sheng-Hung Lai+, Yen-Ming Chu*, Wen-Yen Tsai,” A GPU-based Multiple-pattern Matching Algorithm for Network Intrusion Detection Systems”
[9] J. Moscola, J. Lockwood, R. P. Loui, and M. Pachos, “Implementation of a content-scanning module for an internet firewall,” April 9–11, 2003
[10] S. Antonatos, K. Anagnostakis, and E. Markatos. Generating realistic workloads for network intrusion detection systems. January 2004.
[11] M. Attig and J. Lockwood. “A framework for rule processing in reconfigurable network systems”. 2005
[12] Z. K. Baker and V. K. Prasanna. “Time and area efficient pattern matching on FPGAs.” 2004
[13] H. Bos and K. Huang. “Towards software-based signature detection for intrusion prevention on the network card.” September 2005
[14] R. S. Boyer and J. S. Moore. “A fast string searching algorithm.” October 1977.
[15] J. B. D. Cabrera, J. Gosar,W. Lee, and R. K. Mehra. “On the statistical distribution of processing times in network intrusion detection.”, December 2004.
[16] Giorgos Vasiliadis, Spiros Antonatos, Michalis Polychronakis, Evangelos P., Markatos, and Sotiris Ioannidis,” Gnort: High Performance Network Intrusion Detection Using Graphics Processors”
[17] Nigel Jacob, Carla Brodley, “Offloading IDS Computation to the GPU,” Computer Security(2006)
[18] F. Yu , R. H. Katz , T. V. Lakshman, “Gigabit rate packet pattern-matching using TCAM,” Oct. 5–8, 2004
[19] GPGPU, http://www.gpgpu.org
[20] Snort, http://www.snort.org
[21] NVIDIA CUDA™ Programming Guide Version 2.2.1, http://www.nvidia.com
[22] 張舒,褚艷利,趙開勇,張鈺勃,GPU高性能運算之CUDA.
[23] NVIDIA CUDA Library Documentation, http://developer.download.nvidia.com/compute/cuda/3_0/toolkit/docs/online/