A content-based locality-aware collaborative P2P lookup algorithm

In structured systems, classical routing algorithms such as Chord, Pastry can always find resources within lower logic hops. However, they are independent of the physical network, so that they often find resources in a long delay due to undesirably long distances in some physical links. Resources are always stored in the simple form of <;key,value>; pair in these systems. Similarly, it will result in a long delay due to not considering the semantic properties of data objects and the data searched by clients. In order to solve the problems in peer-to-peer (P2P) networks, such as the mapping problem of mismatching between physical and logical network topology, and the problem of resource storage. Based on self-organizing content-aware collaborative P2P network, a new content-based locality-aware collaborative P2P lookup algorithm (CLP2P) is presented, which comprehensively considers the users' physical locations and interests. It maps nodes of physical proximity and same interest to proximal location in overlay network, and stores the resources on the interested nodes. The proposed algorithm has been assessed through a collection of numerical simulation experiments and the results show that the average hop count and the average routing latency among nodes are reduced, the performance of average resource search success rate is kept high, and the average resource search latency is reduced.