Content-oriented probabilistic routing with measured RTT

摘要

In CONs (Content-Oriented Networks), a router can have multiple entries for the same content name when multiple repositories produce the same content. This is the inherent feature of CONs that users do not care about the places of contents servers. When an Interest packet finds multiple entries at a router on a path, a routing strategy of the router plays a very important role for effective content retrieval. In this paper, we propose a new content routing method, probabilistic routing with measured RTT. In our method, selection probability of routing is calculated based on measured RTT so that an output interface giving smaller RTT has larger probability to be selected. In the performance evaluation, we reveal that our proposal can balance load of content servers and alleviate network congestion without control packets. As the result, our proposed method can reduce content retrieval delay of users.