Rearch And Application Of XML Compression Query Algorithm Based On LFB Storage Structure

With the unceasing development of network and distributed technology,information on many servers throughout the world can be shared. Under theManagement and Analysis System of China Mobile Communications Corporation,Hunan Branch, data from all prefectures and cities need to be gathered and thenstatistically analyzed to get valuable information. But the problem lies in the fact thatdifferent areas store data by using different formats such as WORD, EXCEL, PDF andso forth. Thus, those different formats should be uniformly converted into the sameone in order to integrate the data of “The Isolated Islands”. This problem has to besolved in this subject.XML technology, which has a semi-structured data model and can access data inany format, can solve this problem very well. However, big redundancy ischaracteristic of the XML data. First, redundancy can result in an inordinate waste ofstorage space. Second, redundancy can increase the I/O time of data searching andprocessing, and then searching speed will definitely decrease. At present, compressionis a valid way to minimize the size of XML documents, but compressed XMLdocuments need to be decompressed before doing such operations as validation andsearching, during which time and space costs are excessively high in some application.Therefore, how to synchronize the effective compression of XML documents andoperations like searching on compressed documents is the problem that needs to besolved based on XML data exchange. To this end, this paper lucubrates on someaspects as following:First, it describes the structures and technical features of some existing XMLcompression methods in details, analyses the advantages and disadvantages of varioustechniques, and compares the properties of various compression methods. Second, itstudies X-path, X-query and traditional branch and by value search methods, carriesout a detailed comparison and analysis on these methods, and expatiates on the LFBstorage structure. Third, with the assistant information from LFB, it brings forwardthe branch and by value algorithms based on LFB storage structure to seek the optimalexecution order of query. By comparing the X-path, X-query and traditional branchand by value search methods, it illustrates the advantages of the branch and by valuealgorithms based on LFB storage structure and it shows a good scalability of those algorithms mentioned in this paper through theoretical analysis. At last, it introducesthe establishment of a test platform by making use of Myeclipse IDE. First, theplatform builds a tree like XML model to generate XML. Subsequently, XML data aresimulated and analysis and comparison are made for the results of the simulation. Atthe end the results illustrate that the algorithm mentioned in this paper has higherefficiency of query execution.