XYZ: a scalable, partially centralized lookup service for large-scale peer-to-peer systems

Peer-to-peer (P2P) systems are characterized by direct access between peer computers, rather than through a centralized server. File sharing is the dominant P2P application on the Internet, allowing users to easily contribute, search and obtain content. P2P systems can be categorized by the degrees of centralization. For fully centralized systems, the lookup service will not be available when the central directory server is down. For purely decentralized and unstructured systems, since there is no information about which nodes are likely to have the relevant files, searching essentially amounts to random search. This makes the lookup service unscalable and unpredictable. The objective of this paper is to design a partially centralized, scalable and self-organizing lookup service (XYZ) for wide area P2P systems. A clustering method is used to create the system backbone by connecting the cluster heads together and a color clustering method is adopted to create color overlays and minimize the searching space. Simulations and analysis are also provided. Extensions are proposed to achieve better performance