1 / 13

Fast Packet Classification Using Multi-Dimensional Encoding

Fast Packet Classification Using Multi-Dimensional Encoding. Author: Manfred Georg, Christoph Jechlitschek, Sergey Gorinsky Publisher: High Performance Switching and Routing, 2007 Presenter: Chun-Yi Li Date: 2008/11/12. Outline. Related Work Bitmap intersection

liana
Download Presentation

Fast Packet Classification Using Multi-Dimensional Encoding

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Fast Packet Classification Using Multi-Dimensional Encoding Author: Manfred Georg, Christoph Jechlitschek, Sergey Gorinsky Publisher:High Performance Switching and Routing, 2007 Presenter: Chun-Yi Li Date: 2008/11/12

  2. Outline • Related Work • Bitmap intersection • Aggregated Bit Vector (ABV) • Bit Compression Algorithm (BCA) • Multi-Dimensional Encoding • Performance • Conclusion

  3. Bitmap intersection P1: X2&Y4 ->0011 & 1111 =0011 P2: X4&Y4 ->1101 & 1111 =1101

  4. Aggregated Bit Vector • Aggregation of bit vector - Aggregation tries to decrease the memory access time by adding ABV • Generates false matching. - Rule rearrangement. • Faster than bitmap intersection, but use more space.

  5. Bit Compression Algorithm • Memory storage - θ(d.N.㏒N) • Require additional time for decompression

  6. Bit Compression Algorithm

  7. Introduction • New Coding Vector – Layer Coding Vectors(LCVs) • Memory storage - θ(L.N.㏒N) L : Number of collision-free rule set N : Number of rules

  8. Layer Coding Vector • Find minimum collision-free sets • To minimize memory storage, the wildcards of X dimension are only stored in the LCV of Y dimension

  9. P 4

  10. Number Of Layer Analysis β : The probability that prefix PA is a prefix of prefix PB

  11. Performance

  12. Conclusion • Storage complexity Bitmap intersection - O(d N2) Multi-Dimensional Encoding - θ(L.N.㏒N) • Less memory than bitmap intersection and bit compression • Requires additional processing time for decoding, it still outperforms bitmap intersection, bit-compression and even ABV in terms of classification speed.

More Related