XAncestor: An efficient mapping approach for storing and querying XML documents in relational database using path-based technique

作者:

Highlights:

摘要

XML has become a common language for data exchange on the Web, so it needs to be managed effectively. There are four central problems in XML data management: capture, storage, retrieval, and exchange. Even though numerous database systems are available, the relational database (RDB) is often used to store and query the content of XML documents. Therefore the processes of mapping from XML to RDB and vice versa occur frequently. Numerous researchers have proposed approaches to map hierarchically structured XML documents into the tabular format of a RDB. However, the previously developed approaches have faced problems in terms of storage and query response time. If the design of a RDB is inefficient, the number of join operations between tables increases when a query is executed, which affects the query response time. To overcome this limitation, this paper proposes a new mapping approach, known as XAncestor, which consists of two algorithms: an XML mapping algorithm (XtoDB) and a query mapping algorithm (XtoSQL). XtoDB maps XML documents to a fixed RDB with less storage space. XtoSQL translates XPath queries into corresponding SQL queries based on the constructed RDB in order to reduce the query response time i.e., the time taken to execute the translated SQL query. XAncestor is then developed as a prototype in order to test its effectiveness. The results of XAncestor are compared with those produced by five similar approaches. The comparison proves that XAncestor performs better than the previously developed approaches in terms of effectiveness and scalability. The correctness of XAncestor is also verified. The paper concludes with some recommendations for further work.

论文关键词:XML,Relational database,Model mapping approach,RDB storage space,Query response time

论文评审过程:Received 23 April 2016, Revised 3 October 2016, Accepted 7 October 2016, Available online 8 October 2016, Version of Record 9 November 2016.

论文官网地址:https://doi.org/10.1016/j.knosys.2016.10.009