XML Query Optimization

点击翻译按钮获取中文摘要

XML is becoming the principal medium for data exchange over the Web, and for information integration in general. Increasing amounts of public and private data are described in XML while more legacy sources (e.g., relational databases) offer public XML views. The feasibility of many applications that have emerged with the growth of XML on the Web requires new and complex query optimization techniques. The goal of this research project is to develop a "chase & backchase" optimization method for XML queries. Based on chasing with constraints and incorporating cost-based optimization, the method brings together strategies such as use of indexes, use of materialized views, semantic optimization and join/scan minimization, allowing optimizations that depend on non-trivial interactions between these strategies. Particular attention is given to the challenges posed by XML document order and by regular path expressions in queries. This project is expected to result in a theoretical foundation and a practical framework for defining and using indexes, materialized views and complex constraints in XML query processing systems. The practical framework will be demonstrated through a publicly available software prototype appropriate for teaching about XML query systems and for supporting related research projects.