Generating Efficient Execution Plans for Vertically Partitioned XML Databases
Experience with relational systems has shown that distribution is an effective way of improving the scalability of query evaluation. In this paper, the authors show how distributed query evaluation can be performed in a vertically partitioned XML database system. They propose a novel technique for constructing distributed execution plans that is independent of local query evaluation strategies. They then present a number of optimizations that allow them to further improve the performance of distributed query execution. Finally, they present a response time-based cost model that allows them to pick the best execution plan for a given query and database instance.