Study of index techniques for DOM-based XML databases

As an international standard of data exchange over the Web, XML has penetrated virtually into all fields of Internet applications. How to store and query the XML data with database quickly and precisely is an important research topic. Indexing techniques for XML data are very important to enhance XML database query processing. In this paper, several index structures, including path join index, reference index and value index, for DOM-based XML data query processing are proposed in order to solve the problem of the poor performance of conventional tree traversal approaches. Comparisons and analyses among the different processing approaches for complicated RPE with predicates and/or reference relations are conducted. The detailed performance evaluation results obtained by using typical benchmark experiments are also given to analyze the space utilization, support for queries and maintenance overhead. The results show that the new index techniques can effectively enhance the query performance in XML databases.