Dynamic algorithms with worst-case performance for packet classification

Abstract

Packet classification involves — given a set of rules — finding the highest priority rule matching an incoming packet. When designing packet classification algorithms, three metrics need to be considered: query time, update time and storage requirements. The algorithms pro- posed to-date have been heuristics that exploit structure inherent in the classification rules, and/or trade off one or more metrics for others. In this paper, we describe two new simple dynamic classification algorithms, Heap-on-Trie or HoT and Binarysearchtree-on-Trie or BoT for general classifiers. The performance of these algorithms is considered in the worst-case, i.e., without assumptions about structure in the classification rules. They are also designed to perform well (though not necessarily the ''best") in each of the metrics simultaneously.