Scheduled System Maintenance:
On Monday, April 27th, IEEE Xplore will undergo scheduled maintenance from 1:00 PM - 3:00 PM ET (17:00 - 19:00 UTC). No interruption in service is anticipated.
By Topic

Divide-and-conquer: a scheme for IPv6 address longest prefix matching

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Zhenqiang Li ; Sch. of Comput. Sci. & Technol., Beijing Univ. of Posts & Telecommun. ; Xiaohong Deng ; Hongxiao Ma ; Yan Ma

Longest prefix matching (LPM) is a challenging subject because of the increasing routing table size, the increasing link speed, and the increasing Internet traffic with decreasing packet size. With the advent of IPv6, it requires reconsideration of the previous schemes particularly designed for IPv4. We introduce the first algorithm that we are aware of to employ divide-and-conquer method for IPv6 address 128-bit-long LPM. The algorithm divides an IPv6 address into 8 chunks of 16 bits each and the chunks are reduced recursively through several tunable phases according to the tradeoff between lookup performance and memory consumption. After exhaustive examination of IPv6 address allocation policies and 6 latest global backbone IPv6 BGP routing tables, we put forward some refinements to the basic scheme. The scheme is suitable for both software and hardware implementation. In measurements performed on a 1.1 GHz AMD Athlon machine using the real world IPv6 BGP routing tables, we can forward over several million IPv6 packets per second. When implemented in hardware adopting parallel chunk lookup and pipeline phase lookup architecture, the proposed scheme can achieve one IPv6 LPM lookup per memory access

Published in:

High Performance Switching and Routing, 2006 Workshop on

Date of Conference:

0-0 0