2019 FOSS4G Bucharest Talks speaker: Carmen Tawalika


GRASS GIS in the cloud: actinia geoprocessing

Initially called GRaaS (GRASS as a Service), actinia has been developed to exploit GRASS GIS functionality via HTTPS REST API. GRASS locations, mapsets, vector and raster data as well as spatio-temporal data are available as resources to allow their management and visualization. With the existing (e.g. Landsat) and also recently emerging (Copernicus Sentinel) big geodata pools which are growing day by day, it is designed to follow the purpose of bringing algorithm to cloud geodata. It helps to prepare, analyse and provide a large amount of geoinformation without the need to have know-how of the data, of analysis, appropriate software (for automatisation) or scalable hardware resources.
Some of the features are persistent and ephemeral processing, user management to limit e.g. pixels, processes and duration of calculations as well as logging of API calls and used resources by each user. Another advantage is its easy deployment with docker. When deployed in a cloud based environment with OpenShift or docker-swarm, the integrated load balancer handles a cluster automatically.

Following the spirit of free and open software, we are proud to have emerged as an OSGeo community project and look forward to enlarge the actinia community.

How digging into the earth for the fibre roll-out took GRASS to the cloud

Deutsche Telekom AG (DTAG) revolutionises the planning process of fibre networks (FTTH) to
* increase number of connected households and industries dramatically, and
* shorten the time to market in general.

For this goal DTAG, advanced geoprocessing algorithms including artificial intelligence (AI) automate the conversion of surface information into cost optimised potential trenches used in the fibre build process. The surface information is gathered by terrestrial laser scanners and photos as well as extracted through a complex analysis from aerial orthophotos. This surface information then is used in conjunction with other (open) data as an input for the provisioning of potential trenches needed for FTTH. The processing of these orthophotos as well as the calculation of trenches are handled by the open source actinia cloud geoprocessing engine. Actinia is a new OSGeo community project, offering a REST API to GRASS GIS. The actinia processes are scaled and implemented as a service in the Open Telekom Cloud.

In this talk, we present the handling of large and complex input data in the FTTH-process. Focus is on the usage of actinia, and how actinia supports a core process of the largest German telecommunication provider.