Small forwarding tables for fast routing lookups
SIGCOMM '97 Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication
Scalable high speed IP routing lookups
SIGCOMM '97 Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication
Fast address lookups using controlled prefix expansion
ACM Transactions on Computer Systems (TOCS)
IP lookups using multiway and multicolumn search
IEEE/ACM Transactions on Networking (TON)
IP-address lookup using LC-tries
IEEE Journal on Selected Areas in Communications
Survey and taxonomy of IP address lookup algorithms
IEEE Network: The Magazine of Global Internetworking
Hi-index | 0.01 |
IP address lookup is a challenging problem because of increasing forwarding table size, increasing Internet traffic, higher link speed, frequent prefix updates, migration to 128 bit IPv6 addresses and higher power consumption. IP address lookup need to do two dimensions match to find the longest match prefix. Traditional schemes implement IP address lookup using linear or binary search on prefix lengths or prefix values at the cost of slow lookup speed, complex pre-computation or high power consumption. A novel binary search algorithm based on prefix covered levels is proposed in this paper. At each level we use TCAMs to determine whether there is a match. TCAM entries need not be sorted because prefixes at each level are disjoint. Pre-computation is no longer needed and incremental updates are supported.IP address lookup can be done in O(log2max_level+1) TCAM clock cycle at the worst case where max_level is the max number of overlapping prefixes.The current max_level is 7 for IPv4 and 2 for IPv6. With single TCAM chip having several blocks and keeping one block working and the other blocks power off, or with serveral independent TCAM chips arranged in pipeline architecture, we can support 40Gbps linespeed forwording and reduce the power consumption about 50%. Complexity comparision and performance evaluation shows the proposed sheme has better performance over other schemes.