Research Of The Index Based On Temporal XML Document

Recently, with the development of computer technology and network application, data transmission through network becomes more and more frequent. But different people use different methods to denote the data, which make it difficult to transmit the data. As a kind of data presentation method, XML is actually becoming the standard. The wide usage of XML makes its research important in the field of database.The researches of XML focus on two fields: storage and query. Now there are a lot of researches about these two fields, but the researches on special XML documents are very few. The Temporal XML document is such an example. Now the research on this document is still the same as the traditional XML document, which means the usage of traditional methods to solve the problems on the Temporal XML documents; hence limits the performance efficiency of the Temporal XML document.In this paper, we put forward a storage model of the XML document, which is a typical NXD (Native XML DBMS) storage model. The basic idea is to separate the elements and the character data. The element schema is a tree structure based on the data block, and the character data are stored by cluster. Using this structure, we can manipulate the document more efficiently.In view of the special feature of the Temporal XML document, we propose a new data model on it. XPath data model is a common data model on traditional XML document. Based on this data model, we add the idea of timing; hence adapt it to the Temporal XML document.Based on this storage model, we put forward a special index structure aimed at the Temporal XML document. We extract the element nodes from the document and give rise to a directed acyclic graph (DAG) structure. And the character data are put in the index table and stored in the secondary storage. In the whole index structure, we also take notice of the difference between the Temporal XML document and the traditional XML document, i.e. the time slice property. We put the time slices as a property into the index table, and when we manipulate this special XML document, we can get rid of the traditional methods. This special index structure aimed at Temporal XML document can improve the efficiency greatly. Based on this structure,we also propose algorithms on data insertion, deletion and query.