Storage Research On Native XML Database

With the rapid development of information technology, the usage of XML in various fields is becoming more and more extensive, which requests effective storage and management. XML data is semi-structured, so it is difficult to store in the relational database directly. However, storage is the basis of database system and main factor impacting system efficiency. Native XML database provides a store-model for XML document, Which can be used for XML documents'efficient storage, so the study of the paper is very important.This paper mainly studies the storage of Native the XML database.Firstly,analyses the relationship between XML documents and XML database, meanwhile,does some study about the XML database storage strategy and its advantages. According to the storage granularity and the order of the database, this paper presents a XML documents segmentation algorithm. Secondly, in order to solve the shortcomings of SP and SL clustering algorithm, presents a new PSim-clustering algorithm. The method uses map segmentation techniques and making comparison of the clustering unit's similarity to form a path figure of the similarity.The method can make similar nodes stored in one page as far as possible, which reduces the query time and improves the query efficiency. Lastly, designs a prototype system of the Native XML database. sets up a buffer manager between the buffer and disk files, which manages the XML data exchange and reduces the I/O time. Certain experiments are made to test the validation of the system.