Parallelizing FIB Lookup in Content Centric Networking

Abstract

As a promising future Internet architecture, CCN(Content Centric Networking) is a pragmatic shift from host-to-host to named content model. Providing each piece of content with a specified name, CCN can either directly route a user's Interest for a specific piece of content to the content server, or simply pick the right piece of content from the in network cache and send it back to the user. The combination of content name and in-network cache makes CCN suitable for content distribution. Name based FIB (Forwarding Information Base) table lookup thus plays an important role in CCN routing, as FIB records the forwarding face set for different name prefixes. This paper leverages multi-core platform to accelerate the FIB lookup in CCN router. Based on TILEPro64 platform, which has 64 identical tiles, two parallel zed lookup algorithms, one is based on hash table and the other is based on Bloom filter, are implemented and evaluated. To better evaluate the performance, the size of FIB is estimated and the statistics of URLs are measured to simulate the generation of practical Interests packet traffic. Through theoretical evaluation and experiment, it shows that the parallelization can improve the FIB lookup performance efficiently.