Research on Data Grid-based Geospatial Information Collaborative Sharing System
|School||University of Science and Technology of China|
|Course||Safety Technology and Engineering|
|Keywords||geospatial information data Grid catalogue service virtual geospatial data information sharing Globus Toolkit SOA|
Geospatial data are playing more and more important roles in many research areas (geology, meteorology, agriculture, global climate change, disaster monitoring and controlling, etc) and social/economic activities. Currently many public and private sector organizations have collected huge amount of numerous types of geospatial data using techniques like remote sensing, in-situ measuring, and computer simulation. Those large amounts of geospatial data are archived in different data centers distributed around the world. But there are still many problems associated with the effective use of geospatial data.Geoscience research is multi-disciplinary. It needs to analyze a large amount of data and information. Geospatial data are usually large in volume and geographically distributed. They are collected from multiple sources and archived in different data models and formats. So there is a need for a mechanism to effectively manage and utilize large amounts of distributed geospatial data, so that the burden to search, integrate, and preprocess data for geospatial computation can be relieved. Also, most of archived geospatial data are raw or standard data, which can not be used by most users directly. What users need are high-level data that contain necessary geospatial information and knowledge. A mechanism that enables users to utilize existing geospatial data and services resources to create customized high-level geospatial data can greatly promote the utilization of geospatial resources.This paper explores the application of Grid technology in the geospatial domain. It leverages Grid technology and related geospatial domain technologies and standards to design a geospatial information collaborative sharing system based on data Grid. The main research topics of this paper are:1) Based on the characteristics of geospatial domain, the 5-layer Grid architecture was extended for the geospatial data Grid.2) Based on the OGC CSW Specification, ebXML, and related geoscience standards (ISO 19115 I/II, ISO 19119, ECS, etc), we designed and implemented a geospatial information catalogue service to provide the capabilities of registeration, management, and utilization of large amounts of distributed geospatial data and services.3) By integrating Grid technology and OGC Web Services technology, we designed a geospatial data Grid system, which enables the integration and management of large amounts of distributed geospatial data. This system also provides users a common, interoperable mechanism to acquire distributed geospatial data in a customizable, dynamic way.4) Based on the SOA technology, we proposed the concept of virtual geospatial data. Virtual geospatial data provides geospatial domain experts and common users an effective way to communicate and share knowledge in a collaborative manner.5) We designed a virtual geospatial data system, which provides the capabilities of creation, instantiation, and materialization of virtual geospatial data.6) By using Globus Toolkit, we implemented and deployed a Grid-based geospatial information collaborative sharing system in a real wide-area network environment to validate the concepts and architectures we proposed in this paper.