A cloud computing implementation of XML indexing method using hadoop

Abstract

With the increasing of data at an incredible rate, the development of cloud computing technologies is of critical importance to the advances of researches. The Apache Hadoop has become a widely used open source cloud computing framework that provides a distributed file system for large scale data processing. In this paper, we present a cloud computing implementation of an XML indexing method called NCIM (Node Clustering Indexing Method), which was developed by our research team, for indexing and querying a large number of big XML documents using MapReduce. The experimental results show that NCIM is suitable for cloud computing environment. The throughput of 1200 queries per second for huge amount of queries using a 15-node cluster signifies the potential applications of NCIM to the fast query processing of enormous Internet documents. © 2012 Springer-Verlag.