Efficient query routing in peer-to-peer networks

摘要

In peer-to-peer (P2P) networks, efficient and scalable data retrieval represents a key problem. Unstructured P2P networks avoid the limitations of centralized systems and the drawbacks of structured approaches. However, their search algorithms are usually based on simple flooding schemes, showing severe inefficiencies. In this paper, in order to address this major limitation, we propose the adoption of a local adaptive routing protocol, that uses a smart neighbor selection process such that nodes with similar interest are likely to be grouped together as neighbors. Extensive simulations prove that the approach is able to dynamically adapt the topology to peer interests, organizing them into a small world network. The results also show that our algorithm has a very good impact on the successful rate, allowing to retrieve the resources searched even when they are sparse.