Rasdaman

Rasdaman is an open-source array database management system designed for storing, indexing, and querying very large multi-dimensional arrays, with a focus on raster data such as satellite imagery, climate data, and scientific sensor grids. It treats data as n-dimensional arrays where each array is defined over a coordinate domain and consists of cells holding values.

The system provides a declarative query language, RasQL, that lets users express operations on arrays, including subarray selection, slicing, masking, aggregation, arithmetic across arrays, mosaicking, and spatial-temporal analysis. Operations can be composed to support complex analytical workflows. Rasdaman supports storage of arrays in tiled chunks and uses indexing and compression to enable scalable performance on large datasets. It aims to support both ad-hoc querying and batch processing.

Rasdaman supports interoperability with standard data formats and geospatial standards. It can ingest and output data in formats such as NetCDF, HDF5, and GeoTIFF, and integrates with GIS tools and Web standards to enable sharing and visualization of raster data.

The architecture typically includes a server that stores and processes array data, a metadata/catalog service, and client interfaces (SQL-like RasQL and other APIs).

Use cases include climate and weather research, remote sensing, and earth observation where large time-series of raster data require efficient storage and frequent subsetting, analysis, and visualization.