Research On Building P2P Search System Based On Tapestry

Along with the augmentation of storing space and the drop of storing price, even if a small group of P2P users can share a great deal of resources as well. Abundant shared resources of P2P systems attract large amount of users, but it is difficult to find desirable resources from the huge amount of resources.Based on distributed hash table (DHT) idea, a Metadata-based Peer-to-Peer Information System (MPIS) is designed and implemented in this paper. Because of the diversity and complexity of network resources, it is necessary to characterize each resource with several properties from various points of view so as to express users' requirements more accurately and to make the search result meet the users' requirements better, and metadata is formed as a result. Using the ability of DHT that it is able to map a single key to a peer in the network, and MPIS publish; the inverted indices of each property to certain peers respectively while listing resources, and routes the search request to these peers possessing the inverted indices according to given searching conditions of each property while searching resources. It's required for MPIS to collect the inverted indices of each property and compute the intersection of them as the result of search. Based on DHT component of Tapestry, we implemented the publication and search of a class of typical resources that possess metadata: MP3 audio file, in the environment of Linux using Java programming language. As to those properties that always are used together while searching for resources, we adopt combined properties method to publish and search for resources. We store groups of synonyms in our system so as to expand the scope of searching. Using virtual nodes, we simulate many peers to check our system, and this way also has the function of load-balancement.In this paper, we introduce the conception of P2P, survey the searching methods of various P2P models and analyze the advantages and drawbacks of each method firstly. Then we study how to develop applications on the basis of Tapestry. After this, we describe the design scheme and implementation of MPIS in detail. At last we publish and search some resources using MPIS as experiment, so as to analyze the efficiency and some other problem and show the usability of MPIS.