Font Size: a A A

The Research On Query For Native Xml Database Based On Xquery

Posted on:2009-03-30Degree:MasterType:Thesis
Country:ChinaCandidate:Y Y LiFull Text:PDF
GTID:2198330332981903Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Extensible Markup Language is defined by World Wide Web Consortium of the XML Working Group. Because of its extensible, structural, platform independent and self-descriptive characteristics, XML is fast emerging as the defacto standard for data representation and exchange, and more and more fields use XML to store, exchange and promulgate information.In the domain of storage and query XML data, there are two ways:one is expanding the corresponding function in the existing relational database system or object-oriented database system, and enable them to manipulate XML data, which kind of database known as XML Enabled database; the other is a new kind database-Native XML database (can be abbreviated as NXD), which considers fully the characteristics of XML data. But comparing with traditional database system, Native XML database is quite new and immature, particularly the aspects of query methods, standardization, quotation integrity and so on, all these aspects are a hot topic in database field at present.In order to standardize XML query, W3C have designed a kind of new query language -XQuery. It is consistent with the original XML structure and can be seemed as the "SQL" in XML data management system.The main topic of this paper is about query techniques of Native XML database that using XQuery language. First of all, we carry out in-deep research on XQuery language and the characteristics about Native XML database, pointing out that the differences between Native XML database and XML Enabled database in dealing with XML data, and summeding up the XQuery query, and exploring the commonly used method in XML query processing, and then we design an XQuery query engine for Native XML database. Usually, in Native XML database the mainly method of XML query is pattern-matching or navigation. XQuery query engine that we have designed make use of Data Manager and Index Manager to manage tree-based navigation and pattern matching separately. In the process of dealing with query, we integrate Unnest algorithm and ZigZag Join algorithm and bring forward comb(U+Z) algorithm which combine two query methods to implement query. In this way, it will produce many query plans, and we can utilize cost model to estimate the cost of query plan and choose the best query plan. Through test verification experiment, the results show that using comb(U+Z) algorithm to carry on query, the mixed query plan is better than original only using the sole one query method sometimes, and the method of the cost model which estimate the cost of query plan to choose the best plan is correct. Finally, we carry on the research on analysis the major XML data model. In view of XPath data model not support metadata, we extend the XPath data model by adding Meta attribute.
Keywords/Search Tags:XQuery, Native XML database, XML query processing, data model
PDF Full Text Request
Related items