Efficient IP address lookup with a counting Bloom filter in Trie-based algorithms

Previous research has shown that the search performance of Trie-based algorithms can be improved by adding on-chip Bloom filters. The Bloom filter can identify the membership of a node in an off-chip Trie, and the number of off-chip memory accesses is reduced. But the Trie-based structure has the shortcoming that it is difficult to delete or insert nodes once the Trie has been constructed. In this paper, we propose a new method of utilizing a counting Bloom filter for the IP address lookup of longest prefix match which can make some operations with the Trie nodes. Simulation results show that the best matching prefix can be found with limited off-chip access and much better performance of search can be achieved in our proposed method.