openEO Client Side Processing
The openEO Platform team is happy to announce the release of a new functionality part of the openEO Python Client, called Client Side Processing. This feature allows every Python user to run openEO process graphs in their environment, without the need of a direct connection to the openEO Platform servers.
If you want to try it out, you can find the installation instructions and some examples in the documentation page.
Why
We are constantly trying to understand what could be useful to Earth Observation (EO) data users. In fact, our focus is not just on improving the cloud infrastructure where the data is stored and processed, but we are also concerned about aspects that might be handy for prototyping an EO processing chain.
STAC support
With the load_stac process, also used in the video example above and in the docs, it is possible to load data from public STAC Catalogues and process it in the usual openEO way.
SpatioTemporal Asset Catalog (STAC) is an open specification that evolved from different organizations coming together to increase the interoperability of searching for satellite imagery. More info at stacspec.org
Fast Prototyping
The fast interaction of the Python environment allows to test many combinations of processes, implemented with xArray and Dask. This allows the best interoperability and scalability. The source code of the processes implementation is available on GitHub in the openeo-processes-dask repo, where anyone can contribute designing and implementing new useful processes based on the openEO specifications.
We think that this new feature will ease the process of learning the concepts and processes of openEO, before using them in a cloud environment.
Next steps after successful prototyping
To make use of the scalability of the cloud environment you can register and try openEO Platform for free at this page. You can find the same link also under the main website: openeo.cloud. Additional resources for research projects can be sponsored by the European Space Agency (ESA) via the Network of Resources (NoR), more info here.
Consider also reading this inspiring blog post about scaling to a continental area an openEO workflow using openEO Platform.
