Research And Implementation Of The GML Data Storage And Index Machenism

Geographic Markup Language is an encoding specification which is established by the OpenGIS and based on XML for the geographic information. It is vendor-neutral and platforms-neutral. It provides a unified framework for the modeling, transport and storage of geographic information including both the spatial and non-spatial properties of geographic features. Its appearance offers nice solutions for the sharing and interoperation of spatial data, and meanwhile, it also breaks a new path for the development of WebGIS. GML is an XML encoding for spatial data, so it has the characters such as text-based, extensible, self-description and expressing the complicated data model in the nested way, and it is also massive, complexity and un-orderable like the spatial data. Presently, GML has been the international standard for the encoding and exchange of spatial data. With the massive GML spatial data unceasing emergence, it brings us new challenges: how to effectually manage and manipulate these GML format data and how to build first-rate index mechanism to efficiently retrieve the GML document with larger data. Integrating the technology of XML database and traditional spatial database, this paper delves into the storage and index mechanism of GML spatial data. The main research contents of this paper are: (1) Firstly, the characteristics of the spatial data in the GML specification, the GML schema and the modeling rules of the GML spatial data are analyzes, and then utilizing the open-source, new generation of the XML parser VTD-XML, one universal GML parsing engine is designed and developed, which can parse any GML schema and GML document. (2) In this paper, the primary storage patterns of the current XML data are analyzed in detail, and then by combining the concepts of the native XML database and of the traditional spatial database, the feature-based storage granularity and the feature-based and object-oriented memory model of the GML spatial data are presented. (3) On the basis of the open source native XML database OrientX, a system framework of the native GML database is designed and its corresponding prototype is accomplished in the C#.net development environment. (4)The GML data simultaneously has the XML data characteristics and the spatial data characteristics, so it should be both of them organic synthesis. This paper discusses the common XML index technology and the traditional spatial data index. Considering the enhancement of the query processing efficiency, it suggests one unified indexing model based on spatial index for GML data, and designs a concrete GML index based on the QR-Tree.