27
1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National Cheng Kung University, Taiwan R.O.C. Authors: Pi-Chung Wang, Chun-Liang Lee, Chia-Tai Chan, and Hung-Yi Chang Publisher: IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 15, NO. 4, AUGUST 2007 Present: Yu-Tso Chen Date: February, 12, 2008

1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

  • View
    218

  • Download
    2

Embed Size (px)

Citation preview

Page 1: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

1

Performance Improvement of Two-Dimensional Packet

Classification by Filter Rephrasing

Department of Computer Science and Information Engineering National Cheng Kung University, Taiwan R.O.C.

Authors: Pi-Chung Wang, Chun-Liang Lee, Chia-Tai Chan, and Hung-Yi Chang

Publisher: IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 15, NO. 4, AUGUST 2007

Present: Yu-Tso Chen

Date: February, 12, 2008

Page 2: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

2

Outline

1. Introduction 2. Definitions and problem description 3. Tuple Reorganization by Filter Rephr

asing 4. Performance Evaluation

Page 3: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

3

Introduction

Our idea is inspired from the observation that the performance of the tuple-based schemes ties to the length distribution of the prefixes.

Hence, we exploit the property of prefix nesting by encoding the prefixes as well as the filters to revise the length distributions.

Page 4: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

4

Outline

1. Introduction 2. Definitions and problem descriptio

n 3. Tuple Reorganization by Filter Rephr

asing 4. Performance Evaluation

Page 5: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

5

Algorithms for Tuple Space Search

Page 6: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

6

Rectangle search algorithm

Page 7: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

7

Relationship Between filters and markers

The size explosion is caused by an unbalanced distribution of filter lengths

Page 8: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

8

Length Distribution of Original Filters

Page 9: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

9

Prefix nesting distribution of the routing table

The number of nested prefixes for each routing prefix in the existing routing tables from some major network access points

Page 10: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

10

Outline

1. Introduction 2. Definitions and problem description 3. Tuple Reorganization by Filter Rep

hrasing 4. Performance Evaluation

Page 11: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

11

Basic Prefix Encoding maximal 7 (=2+3+2) bits are required to

represent the original prefixes, but the number of distinct lengths is reduced from 7 to 3.

Page 12: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

12

Exclusive Prefix Encoding

“Level II” IDs could be reduced to 2 bits and the maximal length is reduced to 6.

separate the bitstreams based on their concatenated prefixes at each level

Page 13: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

13

Compare the lengths of the encoded prefixes

Page 14: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

14

Length Distribution of Encoded Filters

The encoded filters only occupy seven tuples The markers are inserted into only four tuples

Page 15: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

15

Search Procedure and Implement

Heuristic methods Select the field with more distinct prefix l

engths as the horizontal axis in the tuple space and perform prefix encoding only on this field.• Only one field is encoded, only one BMP look

up is required.

Choose the field of destination prefix as the horizontal axis

Page 16: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

16

Handling Dynamic Filters

Propose a hybrid approach by maintain two tuple spaces• 1) For encoded filters

• 2) For original filters

New filters is only inserted into the data structure of the tuple pruning search.• Two location lists, one for all filters and one fo

r the new filters.

• New filter inserted into both location lists.

Page 17: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

17

Insertion of Updates

Rectangle search and tuple pruning search are performed simultaneously.

Once the lookup performance of tuple pruning search degrades to threshold, the new filters are merged into a rect-angle search and trigger reconstruction.

During the period, the tuple pruning search looks up all filters by referring to the location list of all filters.

Page 18: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

18

Filter Updates

Page 19: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

19

Deletion of Updates

The pre-computation of the least-cost matching filter is modified to “pre-compute and record all matching filters.”• Listed according to their cost in ascending order.

Once B is deleted, B and best matching filter A are inserted into the tuple pruning search.

And B is recognized as deleted.

Page 20: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

20

Example of Filter Deletion

Page 21: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

21

Outline

1. Introduction 2. Definitions and problem description 3. Tuple Reorganization by Filter Rephr

asing 4. Performance Evaluation

Page 22: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

22

Comparison with Rectangle Search

Hash table and hash entry present the required storage for 1-D BMP lookups.

NLANR routing table

Page 23: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

23

Performance Comparisons

Page 24: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

24

Filter Length Distribution of synthetic Filter Databases

Original filters

Encoded filters, marker, and 1-D hash entries

Filters and markers

Page 25: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

25

Comparison with Rectangle Search(cont.)

Synthetic filter databases

Page 26: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

26

Performance Comparisons

Page 27: 1 Performance Improvement of Two-Dimensional Packet Classification by Filter Rephrasing Department of Computer Science and Information Engineering National

27

Performance Comparison with Existing Schemes