CrateDB is not only a pure time series database; it also accommodates a variety of data types such as JSON, geospatial, vectors, full-text search, and binary large objects (BLOBs).
- Relational and Time Series Data: CrateDB uses a simple relational representation of time series data, allowing easy querying with standard SQL, including window and aggregation functions.
- Document/JSON Data: CrateDB supports JSON data, which is ideal for importing/exporting time series data. It allows direct insertion of JSON payloads with new attributes automatically indexed.
- Full-Text Search: CrateDB handles textual data effectively, offering robust full-text search capabilities for exact matches, typos, phrases, and synonyms.
- Vector Search: CrateDB supports float vector data type for columns, accommodating up to 2048 dimensions. It uses a Hierarchical Navigable Small World graph for efficient indexing of vectors.
- Geospatial Data: It handles geospatial data, enabling tracking of changes over time in a geographic area, vital for numerous applications.
Want to read more?
