Performance Analysis of Complex Query Processing over DHT-based P2P System

Abstract

Distributed Hash Table (DHT) is a promising approach for building a large data management platform. But it cannot support complex query processing due to its one-dimensional index structure. Complex query processing is an essential role in resource discovery services, location aware services, and file sharing applications. In this paper, complex query is handled and analyzed by building a multidimensional indexing scheme over DHT (MIDHT). It is built with two mechanisms: (1) conversion of onedimensional key-based DHT to multi-dimensional keys-based DHT and (2) generation keys for the users’ desired complex query (KeyGenerator). MIDHT is designed and implemented using PlanetSim simulator. It is analyzed under various complex queries. In addition, the analysis results demonstrates that compared with the state-of-the-art approaches, m- Light and LIGHT, MIDHT can save number of DHTlookups and bandwidth consumption in range query performance.