Distributed pattern matching: a key to flexible and efficient P2P search

Flexibility and efficiency are the prime requirements for any P2P search mechanism. Existing P2P systems do not provide satisfactory solution for achieving these two conflicting goals. Unstructured search protocols (as adopted in Gnutella and FastTrack) provide search flexibility but exhibit poor performance characteristics. Structured search techniques (mostly Distributed Hash Table (DHT)-based), on the other hand, can efficiently route queries but support exact-match semantic only. In this paper we have defined Distributed Pattern Matching (DPM) problem and have presented a novel P2P architecture, named Distributed Pattern Matching System (DPMS), as a solution. Possible application areas of DPM include P2P search, service discovery and P2P databases. In DPMS, advertised patterns are replicated and aggregated by the peers, organized in a lattice-like hierarchy. Replication Improves availability and resilience to peer failure, and aggregation reduces storage overhead. An advertised pattern can be discovered using any subset of its 1-bits. Search complexity in DPMS is logarithmic to the total number of peers in the system. Advertisement overhead and guarantee on search completeness is comparable to that of DHT-based systems. We have presented mathematical analysis and simulation results to demonstrate the effectiveness of DPMS.