Research On Labeling Scheme Of XML Containment Join For Update-supporting

As XML has become the de facto standard of data storage and information exchange on Internet, people have got used to applying the XML document to store, manage vast amounts of information on Internet.Consequently Web is gradually transformed into a large XML document database. How effectively store and query XML document has become a hot research field in XML database.In order to effectively support the XML query, especially the structural query, a number of labeling scheme of XML documents is proposed at present. But in actual application XML document data is frequently updated, which takes a lot of time cost of re-encoding, seriously affecting the efficiency of XML query.In this paper, an improved Dewey-labeling scheme for update-supporting (IDSU) is proposed to balance the query efficiency with update efficiency in XML documents, based on thorough analysis of the existing XML document labeling scheme. by extending Dewey coding scheme, it gives each node of XML document tree a unit doublet , IDSU_id means the encoding nodes, the initial form is same as Dewey encoding; Depth indicates the node the number of layers in the document tree,which is used to speed up the structural join operation.The labeling scheme can not only support structure queries efficiently, but also quickly and accurately determine the relationship of any two nodes in the XML document tree among father/son, ancestors/descendants and brothers, meanwhile, for the situation of frequent update and re-encoding, apply a new update rules calld"code addition"to avoid code adjustment after update operation, effectively supporting update of XML document data.To validate the effectiveness of the proposed IDSU labeling scheme, we arranged a large number of comparative experiments from the encoding time, encoding space to query and update performance, the comprehensively and comparatively analysis between these labeling scheme and existing labeling schemes.The experiment results show that, IDSU has good advantages in the XML data updating, meanwhile, effectively support data updates and completely avoid re-encoding, further, there is also a good performance in structural query.The proposed IDSU labeling scheme is an effective attempt in solving the problem of data update of the XML labeling scheme, and give an important reference significance of further research and practical applications.