A TLV-structured data naming scheme for content-oriented networking

Content-oriented networking (CON) addresses the inefficiency of IP networks in supporting content distribution, by decoupling location from identity at the network level, and retrieving a content object through its name or identifier, instead of its storage location or host IP address. A naming scheme is critical in CON designs because a name is used to identify the object and acts as the key for location resolution and routing. This paper proposes a new TLV-structured naming scheme with Bloom filter summarization. The proposed scheme satisfies the requirements of name uniqueness and persistence, and enables more secure content-oriented trust model and effective routing state aggregation for scalability. It also addresses the “suffixhole” problem encountered in conventional prefix-based routing aggregation. We analyzed the performance of the proposed TLV-structured naming and Bloom filter aggregation scheme and our early evaluation results shows that the proposed scheme greatly outperforms the prefix-based aggregation in terms of routing resolution error probability.