Coding And Index Techonlogy Research For Dynamic XML Data

With the development of our social informatization level, the requirement of real-time information is also increasing day by day. How to find the latest information from the huge amount of information has attracted the attention. The XML as a popular data format for data representation and exchanging information on the web, many experts and scholars were already doing a lot of successful work in XML data coding and index. But most of the technology does not support the dynamic XML data. When data update too frequently, there will be need for data re-encoding or reconstructs index. With the increasing of the scale of XML document, the cost of this kind of operation is sometimes unbearable. Good coding technique and index scheme can provide strong support for efficient XML query processing and data update. Therefore, this paper makes a beneficial exploration in coding technique and index scheme for dynamic XML data.This paper first makes a thorough analysis on the structure of XML document characteristics and encoding technology, and put forward a new kind of coding scheme for Dynamic XML Data LSVP (A New Labeling Scheme using Vectors Based on Polar coordinate system for Dynamic XML Data). In the scheme, each XML node is assigned a unique vector Numbers. Because between any two vector can be inserted into innumerable vector, there is enough space for new node insert, update, and delete operation without any effect of coding of old node. At the same time, it is also quickly to figure out the relationship between any two nodes through their unique labels. So when the XML data update very frequently, it can greatly improve the efficiency of the node coding.On the basis of LSVP coding scheme, this paper also presents a new Index scheme-ISDX (Index Scheme for Dynamic XML Data).With the use of the key technology of traditional relational database, XML data information is stored in the path of the two-dimensional list in the scheme. When conducting query operations, it can help skip large number of irrelevant nodes through the search of the summary of the path in the paper.At last, through experiments, compared with XISS index, ISDX index has big advantages not only in build time, but also in query efficiency etc.