Design and Implementation of the EASTWeb System

摘要

The EASTWeb system was designed to automate the retrieval, processing, and storage of satellite remote sensing data for public health officials and GIS researchers. This paper describes the motivating factors behind the design and implementation of the EASTWeb system and focuses on presenting how the system architecture was designed based on Parnas's information hiding modularization methodology. The paper uses a case study to show the ability of the design to account for large changes in the implementation of the system. The initial implementation of the EASTWeb system suffered from reliability and performance issues related to its reliance on ArcGIS, a commercial geospatial analysis tool. To solve these issues an open source geospatial library, GDAL, was used to replace ArcGIS. As a result of replacing ArcGIS, execution speeds were improved by a factor of five and there was a drop in error rates from 0.3% to less than 0.01%.