Fast Name Lookup based on Component State Transfer in Content Centric Networking

Abstract

Content Centric Networking (CCN) is treated as one of the most representatives of the future Internet architecture structure, in which has attracted attention. Interest Packet forwarding relies on content name instead of IP address, the structure of packet name occur hierarchical and the length of name is unlimited, these pivotal features bring new challenges to meet high speed at large scale. In this paper, Fast Name Lookup based on Component State Transfer in Content Centric Networking(FNLCST) is proposed. Hash function and component coding are combined to construct virtual component Trie, then every component is compiled into unique code. In order to improve performance further, components in the same level in the virtual component Trie are stored to different hash tables according to their lengths, so the great number of children nodes can be dispersed by hash function. In addition, the number of memory access times can be reduced because each node in hash table can be allocated memory with specific size to avoid character pointer. Simulation results show FNLCST can achieve high lookup speed and exhibit good scalability to large-scale prefixes table.