A DHT-based identifier-to-locator mapping approach for a scalable internet

It is commonly recognized that today's Internet routing and addressing system is facing serious scaling problems, which are mainly caused by the overloading of IP address semantics. That is, an IP address represents not only the location but also the identity of a host. To address this problem, several recent schemes propose to replace the IP namespace in today's Internet with a locator namespace and an identity namespace. The locator namespace consists of locators that are used to represent the locations of hosts. On the other hand, the identity namespace consists of identifiers that are used to represent the identities of hosts. For these schemes to work, there must be a mapping system that can supply an appropriate locator for any given end point identifier (EID). While prior related works mainly focus on aggregable EIDs, several recent works proposed the use of self-certifying EIDs for purpose of security and privacy. However, self-certifying EIDs are flat, unstructured and prior proposals cannot be used to deal with flat EIDs. In this paper, we propose a Distributed hash table (DHT)-based identifier-to-locator mapping scheme to resolve a locator for a flat identifier. We evaluate the performance of the proposed scheme. We show that, besides the capability to support flat EIDs, the scheme has good scalability and low resolution delay. We also show that the scheme is robust and can efficiently support mobility.