Last modified: 2016-05-10
Abstract
Web publication of the geospatial data requires the use of specialized server infrastructure to meet current publication standards (in particular, the Open Geospatial Consortium standards). Specialized server-side platforms such as Geoserver or Mapserver are used in this case. The deployment of these server-side platforms requires the use of a dedicated physical or virtual server, and needs a full administrative access to the server’s operating system. This approach is common and verified in many projects. However, it fails in the case when the amount of published data is small. In this case, the use of dedicated server becomes too costly.
The problem of Small Geospatial Data publication on the Web can be resolved partially using publically available cloud geospatial platforms (e.g., ArcGIS.com). Nevertheless, in the case of raster coverages publication almost all of currently available cloud geospatial platforms are not able to preserve raster coverages in initial form (form of scientific dataset, that remains applicable for analysis). Coverages is this case are converted into graphical representation that applicable only for visual study.
Taking into account this context, we study the issues of implementation of the lightweight servers for publication and manipulation with the geospatial data on the Web (Panidi et al., 2015). We implemented the Node.js-based engine for Web publication of the raster coverages. This engine implements the Open Geospatial Consortium Web Coverage Service Standard (WCS), and then fully compatible with currently used WCS clients. Additionally, the use of Node.js platform makes it possible to publish small raster datasets on virtual shared hosting. This approach expands the opportunities of geospatial data publication for small- and medium-scale projects, and reduces the cost of publication.