Scaling IDS construction based on Non-negative Matrix factorization using GPU computing

Attacks on the computer infrastructures are becoming an increasingly serious problem. Whether it is banking, e-commerce businesses, health care, law enforcement, air transportation, or education, we are all becoming increasingly reliant upon the networked computers. The possibilities and opportunities are limitless; unfortunately, so too are the risks and chances of malicious intrusions. Intrusion detection is required as an additional wall for protecting systems despite of prevention techniques and is useful not only in detecting successful intrusions, but also in monitoring attempts to security, which provides important information for timely countermeasures. This paper presents some improvements to some of our previous approaches using a Non-negative Matrix factorization approach. To improve the performance (detection accuracy) and computational speed (scaling) a GPU implementation is detailed. Empirical results indicate that the speedup was up to 500x for the training phase and up to 190x for the testing phase.

[1]  Ajith Abraham,et al.  Intrusion Detection Using Ensemble of Soft Computing Paradigms , 2003 .

[2]  Gerhard Wellein,et al.  Data access optimizations for highly threaded multi-core CPUs with multiple memory controllers , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[3]  Lars Elden,et al.  Matrix methods in data mining and pattern recognition , 2007, Fundamentals of algorithms.

[4]  Sugata Sanyal,et al.  Adaptive neuro-fuzzy intrusion detection systems , 2004, International Conference on Information Technology: Coding and Computing, 2004. Proceedings. ITCC 2004..

[5]  Václav Snásel,et al.  Matrix Factorization Approach for Feature Deduction and Design of Intrusion Detection Systems , 2008, 2008 The Fourth International Conference on Information Assurance and Security.

[6]  Tobias Preis,et al.  Accelerated fluctuation analysis by graphic cards and complex pattern formation in financial markets , 2009 .

[7]  P. J. Narayanan,et al.  Singular value decomposition on GPU using CUDA , 2009, 2009 IEEE International Symposium on Parallel & Distributed Processing.

[8]  Ajith Abraham,et al.  Hybrid Feature Selection for Modeling Intrusion Detection Systems , 2004, ICONIP.

[9]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[10]  Jan Platos,et al.  Designing Light Weight Intrusion Detection Systems: Non-Negative Matrix Factorization Approach , 2012 .

[11]  András A. Benczúr,et al.  Large-scale principal component analysis on LiveJournal friends network , 2008 .

[12]  I K Fodor,et al.  A Survey of Dimension Reduction Techniques , 2002 .

[13]  Alexander Thomasian,et al.  CSVD: Clustering and Singular Value Decomposition for Approximate Similarity Search in High-Dimensional Spaces , 2003, IEEE Trans. Knowl. Data Eng..

[14]  Naren Ramakrishnan,et al.  Accelerator-Oriented Algorithm Transformation for Temporal Data Mining , 2009, 2009 Sixth IFIP International Conference on Network and Parallel Computing.

[15]  Richard P. Lippmann,et al.  1999 DARPA Intrusion Detection Evaluation: Design and Procedures , 2001 .

[16]  David Skillicorn,et al.  Using Matrix Decompositions for Data Mining (Chapman & Hall/Crc Data Mining and Knowledge Discovery Series) , 2007 .

[17]  Václav Snásel,et al.  On the Implementation of Boolean Matrix Factorization , 2008, 2008 19th International Workshop on Database and Expert Systems Applications.

[18]  Dionysios Politis,et al.  Socioeconomic and Legal Implications of Electronic Intrusion , 2009 .

[19]  Andrew H. Sung,et al.  Modeling intrusion detection systems using linear genetic programming approach , 2004 .

[20]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[21]  Khalid Saeed,et al.  Normalization Impact on SVD-Based Iris Recognition , 2009, 2009 International Conference on Biometrics and Kansei Engineering.

[22]  David B. Skillicorn,et al.  Understanding Complex Datasets: Data Mining with Matrix Decompositions , 2007 .

[23]  Andrew H. Sung,et al.  Intrusion Detection Systems Using Adaptive Regression Splines , 2004, ICEIS.

[24]  Václav Snásel,et al.  Evolutionary Approaches to Linear Ordering Problem , 2008, 2008 19th International Workshop on Database and Expert Systems Applications.

[25]  Mircea Andrecut,et al.  Parallel GPU Implementation of Iterative PCA Algorithms , 2008, J. Comput. Biol..

[26]  Andrew H. Sung,et al.  Intrusion Detection Systems Using Adaptive Regression Splines , 2004, ICEIS.

[27]  Václav Snásel,et al.  Detecting Insider Attacks Using Non-negative Matrix Factorization , 2009, 2009 Fifth International Conference on Information Assurance and Security.

[28]  M.W. Berry,et al.  Computational Methods for Intelligent Information Access , 1995, Proceedings of the IEEE/ACM SC95 Conference.