Download presentation
Presentation is loading. Please wait.
Published byErin Wood Modified over 9 years ago
1
Lossy Compression of Packet Classifiers Author: Ori Rottenstreich, J’anos Tapolcai Publisher: 2015 IEEE International Conference on Communications Presenter: Yi-Hao Lai Date: 2015/12/09 Department of Computer Science and Information Engineering National Cheng Kung University, Taiwan R.O.C.
2
Introduction In recent years there has been a rapid growth in the size of classification and routing tables resulting in a scalability problem. Compression has gained attention recently as a way to deal with the expected increase of classifier size while keeping it semantically- equivalent to its original form. National Cheng Kung University CSIE Computer & Internet Architecture Lab 2
3
Introduction Measurement show that Internet traffic tends to follow the Zipf distribution and that a large portion of the traffic comes from a small number of flows. Accordingly traffic matches the classification rules in a biased distribution such that some of the classifier information is very seldom useful. National Cheng Kung University CSIE Computer & Internet Architecture Lab 3
4
Lossless compression Huffman coding and the Lempel-Ziv-Welch algorithm are well known lossless compression schemes. Lossless compression of packet classifiers has been deeply investigation in the last decades. The ORTC algorithm achieves an optimal representation with a minimal number of prefix rules. National Cheng Kung University CSIE Computer & Internet Architecture Lab 4
5
Lossless compression National Cheng Kung University CSIE Computer & Internet Architecture Lab 5
6
Lossy compression Lossy compression is a methodology for achieving higher compression ratios at the cost of losing some information about the represented object. National Cheng Kung University CSIE Computer & Internet Architecture Lab 6
7
Lossy compression In the main scheme of our approach, a unique action must be returned for all packets that cannot be classified due to the lossy representation of the classifier. For the unclassified packets we can then calculate the classification in an alternative slower module. National Cheng Kung University CSIE Computer & Internet Architecture Lab 7
8
Lossy compression Approximate Classification Cached Classification National Cheng Kung University CSIE Computer & Internet Architecture Lab 8
9
Approximate Classification National Cheng Kung University CSIE Computer & Internet Architecture Lab 9
10
Cached Classification National Cheng Kung University CSIE Computer & Internet Architecture Lab 10
11
Model and Notation National Cheng Kung University CSIE Computer & Internet Architecture Lab 11
12
Optimization problems National Cheng Kung University CSIE Computer & Internet Architecture Lab 12
13
Approximate Classification National Cheng Kung University CSIE Computer & Internet Architecture Lab 13
14
Cached Classification National Cheng Kung University CSIE Computer & Internet Architecture Lab 14
15
Greedy algorithm The prefix rule popularity National Cheng Kung University CSIE Computer & Internet Architecture Lab 15
16
Greedy algorithm National Cheng Kung University CSIE Computer & Internet Architecture Lab 16
17
Dynamic programming based algorithm National Cheng Kung University CSIE Computer & Internet Architecture Lab 17
18
Dynamic programming based algorithm start by setting the values of g(x,n,a) for a leaf (header) x Let y be a prefix that represents such a monochromatic subtree National Cheng Kung University CSIE Computer & Internet Architecture Lab 18
19
Dynamic programming based algorithm For a non-leaf node x and number of rules n ≥ 1, the function g(x,n,a) satisfies National Cheng Kung University CSIE Computer & Internet Architecture Lab 19
20
Dynamic programming based algorithm National Cheng Kung University CSIE Computer & Internet Architecture Lab 20
21
Dynamic programming based algorithm Cached classification The optimal approximation ratio satisfies National Cheng Kung University CSIE Computer & Internet Architecture Lab 21
22
More general classifiers National Cheng Kung University CSIE Computer & Internet Architecture Lab 22
23
Two-Dimensional Classifiers For a prefix x in the first field and a prefix y in the second, we calculate an optimal encoding of the headers in the rectangle (x,y). Such a rectangle represents the Cartesian product of the two subtrees that correspond to the prefixes x, y in the two fields. National Cheng Kung University CSIE Computer & Internet Architecture Lab 23
24
National Cheng Kung University CSIE Computer & Internet Architecture Lab 24
25
Experimental results National Cheng Kung University CSIE Computer & Internet Architecture Lab 25
26
Experimental results National Cheng Kung University CSIE Computer & Internet Architecture Lab 26
27
Experimental results National Cheng Kung University CSIE Computer & Internet Architecture Lab 27
28
Experimental results National Cheng Kung University CSIE Computer & Internet Architecture Lab 28
29
Experimental results National Cheng Kung University CSIE Computer & Internet Architecture Lab 29
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.