...
The time series are stored in NetCDF-files. Other data types are stored in xml or regular text files. In some cases, for example for security reasons, it is not possible for the FEWS FSS to write directly to the archive data folder. In that case the FEWS FSS should write to an intermediate folder and an copy script should copy the data files in the archive data folder.
To facilitate searching in the archive a catalogue is used. The catalogue is based on Elastic (https://www.elastic.co/) , which is an open source tool for indexing data. The catalogue is populated by the harvester.
The harvester scans the data folder to discover which data sets are available in the archive and updates the catalogue accordingly. THREDDS https://www.unidata.ucar.edu/software/thredds/current/tds/ , an open source software tool, is used to make the data accessible by FEWS and other third party applications. By using THREDDS the data can be downloaded by using HTTP. It addition THREDDS also supports protocols such as OpenDAP and WMS.
...