Automatic Parallelization of XQuery Programs

XQuery is a functional language with implicit parallelism. It is an important approach to improve the efficiency of XML query by taking full advantage of multi-core environment in the parallel implementation of XQuery language. In this paper, we propose an implementation method for parallelizing XML query represented by XQuery programs automatically. According to the features of its functional language, an XQuery program is divided into a number of tasks that can be executed in parallel. Then, on the basis of the running cost evaluation, three kinds of parallelism are applied to different tasks and they are data parallelism, task parallelism and pipeline parallelism. Under the guidance of a novel scheduling strategy, the execution of the XQuery program is parallelized automatically. The experiments show that this approach improves the efficiency of the execution of XQuery programs and the computing resources of multi-core computer are used efficiently.