Research On XML Data Query Optimization Technique

With the rapid development of the Internet,XML has become data storage and exchange standards on the Internet. However, XML documents are tree structures to represent, each time the data query must start from the root of the tree search to read through each layer of the label name of the layer by layer the search until you find the required node, so for data storage and query, we still need to spend a lot of space costs and time costs, enabling people to query the data when faced with many difficulties and how to quickly and efficiently inquiries,processing of XML documents, has become an important research topic in the field of the database .This paper researches XML algebra, as well as XML based on XML Schema of the XPath query technology and based on MP code of XPath query technology to get the purpose of query optimization. The main researches are as follows:Firstly,This thesis is based on the theory of algebra. By introducing the data model and XML language combine to minimize the number of scans of the original XML document, using heuristic algebraic query rewriting method, the use of the operation of relational algebra operations to rewrite XML document to achieve physical optimization purposes.Secondly, Due to the flexibility of XML data, in the XPath query time, there are many uncertainties. So, we propose an optimization method, XPath expression Analysised behind determined based, use XML documents given by Schema information and XPath finding, establish tree directed graphs of two informations, use the product of maps to optimize, through the Xmark resulting data validation before and after optimization of the time ratio is greater than 1, and prove that optimize purposes.The third, As XPath query is a hierarchical search, each time need to start from the root, to search for a specific axis of each node, and then for each node in the node test done, resulting in a lot of time cost waste. Therefore, if we can make a good tree structure for the encoding mechanism, completely record the tree structure and properties, retain the relationship and the order between the nodes in the document, use this encoding mechanism to XPath queries specific axis so that no further root start finding, you can effectively query the desired nodes.Major propose a MP code method, and prove that the encoding of each node are the only one, and each node in the encoding of the node contains the path information, and Document Order of the node. With XPath based, this study used qualitative coding method to provide nine kinds of axes of search algorithms, as well as the encoding node stored in the Code Book , when users query data, only through the mathematics division and factorization operations, can be quickly find the desired data, effectively reduce the query cost.