Based On The Index Technology Of Xml Query Optimization Research

Nowadays XML data are widely used in different fields, the data and documents scale are on the increasing and growing larger. There are so many XML data with the complicated structure on the network. How to manage and query complex XML data effectively has become a critical issue at present. In the past10years, many researchers and scholars put forward a variety of query optimization methods from different views. Encoding and indexing technologies are two the most common and important methods in all query optimizations. It has been proved that the coding and indexing are fully applied, and can achieve and meet the need of query optimization well. Therefore, combined with coding and indexing technology to explore query optimization for the XML data is still a subject worthy of further studying.The paper has proposed a new method to encode XML document node. It introduces the trigeminal liked list storage structure of binary tree for encoding, while binary tree traversal-based coding is used formerly. The natural numbers are expressed as the serial numbers of the node encoding. The paper establishes the relevant indexing model, which uses the appropriate key words as index terms, and takes CSBTT(Coding Schema Based on Binary Tree Traversal) as XML document tree node encoding.On the basis of traditional interval encoding, the paper has designed the joint structure index model, which is composed of DTD(Document Type Definition) structure index, XML document index and content index. The DTD and content index use the inverted table as basic unit, while the XML document index uses B+tree as basic unit. In the processing of queries, only through one entrance that other index entry can be found to complete and meet the needs of comprehensive and multiple queries.Theoretical and experimental results indicate that CSBTT has a small storage space and high query efficiency, and supports dynamic data updating operation. The index has the advantages of small space, fast response speed and high query recall. The paper has built an index model with joint structural, which use the interval encoding as the document tree node coding. The index model has high query efficiency, can satisfy the multiple document query demand and solve the XML document hybrid optimized query (text query and structure query) and other practical problems.