| With the development of science and technology, data grow rapidly, people are at the age of submerged by data. If people want to find what they want from the huge amounts of data, more time will cost than before. However distributed technology can be applied to build a distributed system for querying original data stored in different machines. At the same time, due to the size of metadata is far less than the original data, so you can through the management of metadata to query the raw data.XML is an extensible markup language, it uses tree model to represent data, and XML document is usually expressed as twig pattern to query. At the same time, keyword on XML document retrieval is widely studied because of its intuitive, friendly query interface. In order to better integrate the domain of data management and information retrieval for XML document query technology research, the W3C proposed XPath, it is a kind of language used to locate a part in the XML document content, based on the XML tree structure, which can find a node in the tree data structure. In this paper, aiming at the problem of huge amounts of data query, will extract metadata from massive data in XML format and store into HBase, turning it into metadata query problem of huge amounts of data. Through to the metadata query, the user can quickly query what they want.To solve the above problem, this paper developed a distributed system based on XML metadata management system. It is a system based on B/S structure, using XML-RPC as communication protocol between the modules. After login system, the user can upload their own XML metadata file by FTP, input an XPath query language to query, and they can also add, delete or modify the data on the result. |
PDF Full Text Request