Research On Data Query Optimization Based On XML Database

XML language has been a standard of expressing and exchanging data on Internet. How to inquire XML projects becomes more important, because XML projects appear more. As a result of many advantages of relational database, such as, mature technologies, extensive applications, great managing data capabilities, high data safety degree, and so on, XML data query can utilize mature relational database technologies to realize.This thesis analyzes the flaws of HTML appearing with the development of Internet, such as interacting badly, describing badly, extending difficultly; introduces the characters of XML, which is widely used on Web, and analyze its superiorities; forms structure projecting methods and model projecting methods of XML store methods on the basis of the relational database, compare some major store methods; studies several XML database query technologies, and analyze XML query ways.Query rewriting technology based on view is an important technology of the query optimization. Half-structure query rewriting methods receive correct query rewriting plans by visting each query rewriting selection plan, without reducing the exponent query rewriting selection space, so the actual implementing algorithm efficiency is low. This thesis studies how to reduce the exponent query rewriting selection space, and brings out a new method to solve the problem, with drawing some tactics in the relational query rewriting bucket algorithm into original query rewriting, that is to say, in half-structure database using the relationships among child targets and the self-characteristics of half-structure data query. Promising algorithm integrity, we limit rewriting space from query aspects and view aspects, to acquire less query rewriting selection plans, and to enhance the efficiency of implementing algorithm. At last, prove the viewpoint brought out by comparing the two algorithms by time complexity and space complexity.