A fast and scalable conflict detection algorithm for packet classifiers

Abstract

Packet filters are rules for classifying packets based on their header fields. A filter conflict occurs when two or more filters overlap, creating an ambiguity in packet classification. There has been prior works on conflict detection for multi-dimensional classifiers, but their efficiency and scalability are not good. A new algorithm is proposed, which uses hashing-based PATRICIA trie. The new algorithm can fast detect conflicts in classifiers and have high scalability. The technology of processing transport-level ports can bring more security than existed algorithms. © Springer-Verlag Berlin Heidelberg 2005.