ROIN: reputation-oriented inverted indexing for the P2P network

Abstract

Traditional DHT-based P2P networks cannot provide fuzzy search for resources. On the contrary, inverted index-based DHT network can provide semantic-based fuzzy search for resources similar to search engines over the Internet. However, inverted indexing in P2P networks may bring about extra costs for the maintenance of the index network, due to the dynamic nature of the P2P network. In this paper, we present ROIN, an efficient index network framework for the P2P network. In the ROIN framework, an inverted index network for the fully distributed P2P network is established by assigning nodes with high reputations as index nodes. A group of experiments concerning the search coverage rate, the search latency, the load of index nodes and the network disturbance of the ROIN framework are carried out. A series of experimental evaluations and comparisons are made between ROIN, PCIR, Pastry, and GlusterFS. Upon the experimental results and evaluations, we conclude that applying invert index and selecting a suitable number of nodes with high reputations as index nodes will bring about a robust and efficient distributed network that supports fuzzy search for resources.