Detail publikace

Effective Hash-based IPv6 Longest Prefix Match

Originální název

Effective Hash-based IPv6 Longest Prefix Match

Anglický název

Effective Hash-based IPv6 Longest Prefix Match

Jazyk

en

Originální abstrakt

This paper introduces new longest prefix match algorithm Hash-Tree Bitmap suitable for IPv6 look-up. While most of the existing LPM algorithms focus on high density IPv4 prefix tables, the proposed algorithm is well suited for combination of both sparse and dense areas, which is typical for IPv6 rule sets. The Hash-Tree Bitmap algorithm use efficient and fast hash look-up for sparse areas and Tree Bitmap algorithm for high density trie areas. The proposed algorithm was mapped to the hardware architecture and implemented as a pipeline of processing elements. The resulting hardware architecture offers well balanced resource requirements and is able to achieve the wire-speed throughput for 100 Gbps networks.

Anglický abstrakt

This paper introduces new longest prefix match algorithm Hash-Tree Bitmap suitable for IPv6 look-up. While most of the existing LPM algorithms focus on high density IPv4 prefix tables, the proposed algorithm is well suited for combination of both sparse and dense areas, which is typical for IPv6 rule sets. The Hash-Tree Bitmap algorithm use efficient and fast hash look-up for sparse areas and Tree Bitmap algorithm for high density trie areas. The proposed algorithm was mapped to the hardware architecture and implemented as a pipeline of processing elements. The resulting hardware architecture offers well balanced resource requirements and is able to achieve the wire-speed throughput for 100 Gbps networks.

BibTex


@inproceedings{BUT76336,
  author="Jiří {Tobola} and Jan {Kořenek}",
  title="Effective Hash-based IPv6 Longest Prefix Match",
  annote="This paper introduces new longest prefix match algorithm Hash-Tree Bitmap
suitable for IPv6 look-up. While most of the existing LPM algorithms focus on
high density IPv4 prefix tables, the proposed algorithm is well suited for
combination of both sparse and dense areas, which is typical for IPv6 rule sets.
The Hash-Tree Bitmap algorithm use efficient and fast hash look-up for sparse
areas and Tree Bitmap algorithm for high density trie areas. The proposed
algorithm was mapped to the hardware architecture and implemented as a pipeline
of processing elements. The resulting hardware architecture offers well balanced
resource requirements and is able to achieve the wire-speed throughput for 100
Gbps networks.",
  address="IEEE Computer Society",
  booktitle="IEEE Design and Diagnostics of Electronic Circuits and Systems DDECS'2011",
  chapter="76336",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2011",
  month="april",
  pages="325--328",
  publisher="IEEE Computer Society",
  type="conference paper"
}