we address the problem of scalable distributed similarity searching. Our work is based on single-site metric space indexing algorithms. They provide efficient way to perform range and nearest neighbor queries on arbitrary data in general metric spaces. The metric spaces are excellent abstraction that allows comparison of very complex objects (such as audio files, DNA sequences, texts). We have exploited the SDDS (Scalable and Distributed Data Structures) paradigms and P2P (Peer to Peer) systems to form a metric space similarity searching structure in distributed environment. Our proposed method is fully scalable without any centralized part and it allows performing similarity queries on stored data.