A Super-Peer Topology for Scalable P2P Systems

Abstract

Current peer-to-peer (P2P) file sharing applications are remarkably simple and robust, but their inefficiency can produce very high network loads. New P2P systems, called structured P2P systems, have been proposed and are much more efficient in terms of search time and bandwidth requirements, and are therefore more scalable. However, structured systems are considerably more complex, and likely to be less robust and flexible.

The use of hierarchy has been proposed to improve the performance of unstructured P2P systems. This hierarchy is implemented via an extra layer of super-peers who manage queries and index placement on behalf of a set of client peers. Unstructured P2P systems with super-peers have the potential to approach the performance and scalability of structured systems while retaining much of the generality and simplicity of (non-hierarchical, basic) unstructured P2P systems. There has, however, been little consensus on the best topology for connecting these super-peers, or how to construct the topology in a distributed, robust way.

In this thesis, we propose a Scalable Unstructured [P2P] System (SUPS). The unique aspect of SUPS is a protocol for the distributed construction of a super-peer topology that has highly desirable performance characteristics. The protocol is inspired by the theory of random graphs. We describe the protocol, and demonstrate experimentally that it produces a balanced and low-diameter super-peer topology at low cost. We show that the method is very robust to super-peer failures and inconsistent information, and compare it with other approaches.