Parameterizable Decision Tree Classifier on NetFPGA

  • Alireza Monemi
  • Roozbeh Zarei
  • Muhammad Nadzir Marsono
  • Mohamed Khalil-Hani
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 182)


Machine learning approaches based on decision trees (DTs) have been proposed for classifying networking traffic. Although this technique has been proven to have the ability to classify encrypted and unknown traffic, the software implementation of DT cannot cope with the current speed of packet traffic. In this paper, hardware architecture of decision tree is proposed on NetFPGA platform. The proposed architecture is fully parameterizable to cover wide range of applications. Several optimizations have been done on the DT structure to improve the tree search performance and to lower the hardware cost. The optimizations proposed are: a) node merging to reduce the computation latency, b) limit the number of nodes in the same level to control the memory usage, and c) support variable throughput to reduce the hardware cost of the tree.


Data Mining Machine Learning Search Tree 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bermak, A., Martinez, D.: A compact 3D VLSI classifier using bagging threshold network ensembles. IEEE Transactions on Neural Networks 14, 1097–1109 (2003)CrossRefGoogle Scholar
  2. 2.
    Breiman, L.: Random Forests. Machine Learning 45, 5–32 (2001)MATHCrossRefGoogle Scholar
  3. 3.
    Eklund, P., Kirkby, S.: Machine learning classifier performance as an indicator for data acquisition regimes in geographical field surveys. In: Proceedings of the Third Australian and New Zealand Conference on Intelligent Information Systems, pp. 264–269 (1995)Google Scholar
  4. 4.
    Erman, J., Mahanti, A., Arlitt, M., Cohen, I., Williamson, C.: Offline/realtime traffic classification using semi-supervised learning. Performance Evaluation 64(9-12), 1194–1213 (2007)CrossRefGoogle Scholar
  5. 5.
    Breiman, L., Friedman, J.H., Olshen, R.A., Stone, C.J.: Classification and regression trees. Wadsworth, Monterey (1984)MATHGoogle Scholar
  6. 6.
    Lockwood, J.W., McKeown, N., Watson, G., Gibb, G., Hartke, P., Naous, J., Raghuraman, R., Luo, J.: NetFPGA–An Open Platform for Gigabit-Rate Network Switching and Routing. In: Proceedings of 2007 IEEE International Conference on Microelectronic Systems Education (2007)Google Scholar
  7. 7.
    Lopez-Estrada, S., Cumplido, R.: Decision Tree Based FPGA-Architecture for Texture Sea State Classification. In: IEEE International Conference on Reconfigurable Computing and FPGA’s (2006)Google Scholar
  8. 8.
    Moore, A.W., Papagiannaki, K.: Toward the Accurate Identification of Network Applications. In: Dovrolis, C. (ed.) PAM 2005. LNCS, vol. 3431, pp. 41–54. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  9. 9.
    NetFPGA (2012),
  10. 10.
    Nguyen, T., Armitage, G.: Training on multiple sub-flows to optimise the use of machine learning classifiers in real-world IP networks. In: Proceedings of IEEE 31st Conference on Local Computer Networks, pp. 369–376 (2006)Google Scholar
  11. 11.
    Nguyen, T., Armitage, G.: A survey of techniques for internet traffic classification using machine learning. IEEE Communications Surveys Tutorials 10(4), 56–76 (2008)CrossRefGoogle Scholar
  12. 12.
    Qi, Y., Fong, J., Jiang, W., Xu, B., Li, J., Prasanna, V.: Multi-dimensional packet classification on FPGA: 100 Gbps and beyond. In: International Conference on Field-Programmable Technology, FPT, pp. 241–248 (2010)Google Scholar
  13. 13.
    Struharik, R., Novak, L.: Intellectual property core implementation of decision trees. IET, Computers Digital Techniques 3(3), 259–269 (2009)CrossRefGoogle Scholar
  14. 14.
    Wang, Y., Yu, S.Z.: Machine Learned Real-time Traffic Classifiers. In: Second International Symposium on Intelligent Information Technology Application, vol. 3, pp. 449–454 (2008)Google Scholar
  15. 15.
    Williams, N., Zander, S., Armitage, G.: A preliminary performance comparison of five machine learning algorithms for practical IP traffic flow classification. Special Interest Group on Data Communication (SIGCOMM) 36(5), 5–16 (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Alireza Monemi
    • 1
  • Roozbeh Zarei
    • 1
  • Muhammad Nadzir Marsono
    • 1
  • Mohamed Khalil-Hani
    • 1
  1. 1.Faculty of Electrical EngineeringUniversiti Teknologi MalaysiaJohor BahruMalaysia

Personalised recommendations