Oriented Soa, Xml Data Compression Technology Research And Application

With the rapid development of computer technology in recent years, XML has become data representation, storage and exchange standard on Internet. At the same time, the advent of electronic commerce and the business needs in firms have led to the booming of service-oriented architecture system(SOA), these factors make SOA become a hot discussion topic. The essence of SOA is a architectural patterns in order to organize services, it connects services through unified, neutral, well-defined interfaces and contracts, and its implementation is independent with hardware platforms, operating systems and programming languages, so it can make services interact each other in a universal way. Besides, XML is self-descriptive and cross-platform, these characters make XML correspond with the requirements of SOA enviroment. Therefore, XML has a good application that we can apply XML as data transmission format and information carrier in SOA architecture.However, the greatest flaw of XML is its redundancy, especially a large amount of repeated structural information cause the volume of XML data bulky, take up too much storage spaces and network bandwidth, reduce processing efficiency, increase the difficulties of SOA message transmission, result in SOA data transmission bottlenecks. To solve the problem, we can make use of Gzip to compress transmissive data through configuring the servers parameters in SOA system, Gzip is a universal text compression technology, not designed for XML specifically. While taking XML data itself characteristic and SOA application enviroment into consideration, this paper mainly research the compression method which is storage-oriented and exchage-oriented and suitable for XML data, and come up with own improvements on the basis of previous studies.Firstly this paper describes SOA basic concepts, achievement ways and existing SOA data transmission problem which are resulted in XML redundancy, and propose that XML data compression is an effective means to solve the problem. Secondly, this paper introduces XML basic concepts, related processing technology and its redundancy, and introdoces two classical universal data compression technologies, which are respectively Huffman coding and LZ77 algorithms, these basic compression algorithms are a foundation of all compression techniques and compression tools. In addition, this paper describes compression model XMill working principle, it separates structual information from data contents. For structural information, XMill uses dictionary coding to compress, for data contents, XMill classifies the data contents into the corresponding containers in accordance with labels these tags belong to and use Gzip to deal with them. If users can participate in the compression process, XMill can invoke more suited compression method to compress data comtainers based on user's choices. so compression rate can be increased. Finally In SOA application enviroment, the transmissive data must comply with the specification of XML Schema document, Therefore, based on XMill the paper use XML Schema into XML data compression, Schema contains all the XML document element type definition. When the data are put into containers, we can read data type values from XML Schema and choose the appropriate compression algorithm to deal with data, rather than still using Gzip.If some special data types do not match with compression algorithm successfully, we still use Gzip,The method can be independent on user participation, treduce user burden and realise automation to a certain extent. The improved model is named XS-XMLComp (Based on XML Schema of XML data Compression Technology). One of its innovations is to use auxiliary XML Schema. documents, and its compression ratio is better than Gzip used in SOA.At the last, this paper carry out the simulation experiments, compare XS-XMLComp with other several compression models and contrast several performance indicators including compression ratio, processing speed and efficiency, environmental adaptation respectively. This paper commentates the advantages and disadvantages of various models through the analysis of results, and confirms that XS-XMLComp compression model has feasibility and validity, make the compression effects increased to a certain degree.