Efficient Processing of Branch Queries for High-Performance XML Filtering

In this paper, we consider the problem of filtering a continuous stream of XML data efficiently against a large number of branch XPath queries. Several approaches have been proposed, and many of them improve their run-time efficiencies by sharing some paths between branch queries. This paper further improves the run-time efficiencies by classifying and grouping semantically equivalent twig patterns, and identifying the common paths that are shared between these groups. Query structure matching is done at index compilation phase, and the paths shared between these groups of queries are processed once. Experiments show that our proposal is efficient and scalable compared to previous work.