The Research On Parallel Algorithm For XML Query

Recently, XML (Extensible Markup Language) has become new standard of data representation and exchange on the Internet and has been accepted in many fields, getting more and more attention. With wide adoption and development of XML as a description language for document and information, people will have more requirements for performance of XML query. Therefore, how to improve the query efficiency for XML has been increasingly important.Although great achievements of XML research have been made, there are many problems of XML query technology because of XML inherent characteristics and the differences from traditional relational data model. This paper focuses on XML queries, sums up the current status and developing, analyzes the methods of querying XML in detail. The developing, functions, structure, features and mainstream technology of XML database, including conventional XML-Enabled DBMS(XED) and Native XML DBMS (NXD), Storage modes of XML are disscused in XML database.It proposes a mixed XML storage structure, supporting bidirectional conversion storage structure between structurized data and nonstructured data, improving the efficiency of XML query.Many evaluation methods for XML queries are proposed recently. However, with the popularity of multiprocessor and multi-core processor computer system, these conventional approaches fail to use the advantage of multiprocessor and multi-core processor. In this paper, we discuss theory and of parallel computer and design a parallel paradigm using divide-and-conquer paradigm and working pool paradigm. By selecting XML processing model, designing parallel granularity and parallelism, we propose a new parallel algorithm for XML query using matching algorithm of tree, which largely improve the efficiency of XML query processing in shared memory multiprocessor and multi-core processor computer system and make a performance test.