Improving multidimensional point query search using multi-way peer-to-peer tree network

Abstract

Nowadays, peer-to-peer (P2P) networks are widely accepted in multidimensional applications like social networking, multiplayer games, P2P e-learning, P2P mobile ad-hoc networks, etc. Various P2P overlay networks combining multidimensional indexing (MI) methods are preferable for efficient multidimensional point or range search in a distributed environment. However, point query search in existing P2P has limitations viz.: 1) either does not give support to MI or uses replications to support MI; 2) point query search cost is limited to O(log2N). Hence, traditional MI techniques based on the multi-way tree structure (having larger fan-out) can be employed to enhance the multidimensional point query search capabilities. Based on our observations, a hybrid model combining m-ary (m fan-out of the tree, > 2) P2P tree network and MI based on space containment relationships is preferred to reduce point query search performance bound to O(logmN) using single overlay network. The present paper shows how this model improves the search performance of the point queries in O(logmN) steps, independent of the dimensionality of objects.