M-KAD: A Scalable Peer-to-Peer System for Similarity Search in Metric Space

Similarity search plays an important part in data management field. In this paper. we present a novel data driven distributed storage structure for similarity search in Metric Space. We use Kademlia based peer-to-peer architecture for data storage. and design effect algorithms for data indexing and similarity query. At the end of this paper, we prove the system performance by simulation experiments.