14
Compact Trie Forest: Scalable architecture for IP Lookup on FPGAs Author: O˘guzhan Erdem, Aydin Carus and Hoang Le Publisher: ReConFig 2012 Presenter: Yu Hao, Tzeng Date: 2013/2/20

Compact Trie Forest: Scalable architecture for IP Lookup on FPGAs

  • Upload
    bozica

  • View
    41

  • Download
    0

Embed Size (px)

DESCRIPTION

Compact Trie Forest: Scalable architecture for IP Lookup on FPGAs. Author : O˘guzhan Erdem , Aydin Carus and Hoang Le Publisher : ReConFig 2012 Presenter : Yu Hao , Tzeng Date: 2013/2/20. Outline. Introduction Algorithm and Data Structure Architecture and Implementation on FPGA - PowerPoint PPT Presentation

Citation preview

Page 1: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Compact Trie Forest: Scalable architecture for IP Lookup on FPGAsAuthor: O˘guzhan Erdem, Aydin Carus and Hoang LePublisher: ReConFig 2012Presenter: Yu Hao, TzengDate: 2013/2/20

Page 2: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Outline

• Introduction• Algorithm and Data Structure• Architecture and Implementation on FPGA• Performance• Conclusion

Page 3: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Introduction

• Most hardware-based solutions for network routers• TCAM• DRAM / SRAM

• FPGA

• Compact Trie Forest• support up to 703K IPv4 and 418K IPv6 prefixes• sustain a throughput of 420 million lookups per second, or 135

Gbps for the minimum packet size of 40 Bytes

Page 4: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure

• Definitions and Notations• MSB ( Most Significant Bit )• LSB ( Least Significant Bit )• MSSB ( Most Significant Set Bit )• LSSB ( Least Significant Set Bit )• MSRB ( Most Significant Reset Bit )• LSRB ( Least Significant Reset Bit )• Example : 00110101*

• 0 and 1 values as MSB and LSB• 2, 7, 0 and 6 values for MSSB, LSSB, MSRB and LSRB positions

• Prefix Node• Non-Prefix Node

Page 5: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)• Definitions and Notations• Active Part

• MSSB and LSSB bits for (MSB, LSB) = (0, 0)• MSSB and LSRB bits for (MSB, LSB) = (0, 1)• MSRB and LSSB bits for (MSB, LSB) = (1, 0)• MSRB and LSRB bits for (MSB, LSB) = (1, 1)• Example A : 011011*• AP : 1

• Example B : 00101100*• AP : 01

• Conflicted Prefixes• For instance, the active part of 011001 and 001∗ 1010 are 10.∗

Page 6: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)• Prefix Table Conversion• prefix p = xyz can be represented as a triplet {|x|, y, |z|}• For example, 000010010100 can be represented as {5, 0010 , ∗ ∗

3},

Page 7: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)• Compact Trie Structure• In addition to the child pointers and the next hop information

fields, extra information (|x|, |z|, MSB and LSB values) are stored at each node to differentiate the conflicted prefixes.

Page 8: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)• Compact Trie Structure• set a limit for the number of conflicted prefixes per node Ptrie, and

move the excessive conflicted prefixes to a newly generated CT

Page 9: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)• IP Lookup Algorithm

Page 10: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Algorithm and Data Structure (Cont.)

Page 11: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Architecture and Implementation on FPGA

Page 12: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Architecture and Implementation on FPGA (Cont. )

Page 13: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Performance

Page 14: Compact  Trie  Forest: Scalable architecture for IP Lookup on FPGAs

Conclusion• Therefore, the algorithm can be used to improve the

performance (throughput and memory efficiency) of trie-based IPv4/v6 lookup schemes to satisfy fast internet link rates up to and beyond 100 Gbps at core routers, and compact memory footprint that can fit in the on-chip caches of multi-core and network processors.