Binary searches on multiple small trees for IP address lookup

Abstract

Rapid growth of Internet traffic requires more Internet bandwidth and high-speed packet processing in Internet routers. IP address lookup in routers is an essential operation that should be performed in real-time for routers where hundreds of million packets arrive per second. In this paper, we propose a new software-based architecture for efficient IP address lookup. In the proposed scheme, a large routing table is divided into multiple balanced trees, and sequential binary searches are performed on those trees, and hence the number of memory accesses depends on the number of routing entries not on the length of routing prefixes. Performance evaluation results show that the proposed architecture requires a single 301.7 Kbyte SRAM to store about 41000 routing entries, and an address lookup is achieved by 11 memory accesses in average.