Data Cubes in Cloud Platforms
One important feature of an EO cloud platform is to host satellite data. Usually vast amounts of satellite data, for example the whole archive of Landsat and the Sentinels which adds up to Petabytes of data. Users want to access this data efficiently. Therefore the storage system of an EO cloud platform needs to be optimized towards user requests. Data Cubes play an important role here. Organizing the raw files into virtual data cubes solves many issues for EO cloud platforms.
- Data Management:Â Different Satellites have different formats and are structured differently. This can easily create confusion because the different sources have to be treated differently. After organizing these files into data cubes the interaction with all different satellite data follows the same rules.
- Separated Collections:Â Every satellite missions images form an own data cube, also called collection, with its according metadata. Ideally cloud processing tools on the platform allow for easy data fusion of different collections.
- Flexible Subsetting:Â ()Â are developed to facilitate subsetting operations along the different dimensions and return only the required subset. Without worrying about the different raw files that are involved (e.g. area of interest is crossing a tile border).
Remember, a data cube is not an EO cloud platform. It’s only one part of it.