I'm learning Apache IoTDB. I find that there's a concept/feature named tag/attribute. As far as I know, the data model of influxdb is all based on tag, while this concept is not that estimated in IoTDB. What's the difference between them?
1 Answers
Good question! From my perspective, IoTDB's tag is just an assistant feature to help manage timeseries while Influxdb's tag is the key concept to define timeseries. Influxdb uses tags to uniquely coordinate one timeseries via concatenating measurement, tag-key, tag-value, field. Comparatively, IoTDB uses the full path from the tree root to the leaf node to uniquely represent one timeseries. I think the node names on a full path in IoTDB are equivalent to the tag in InfluxDB. IoTDB's tag is not designed for timeseries identification, but for timeseries batch process. IoTDB supports query timeseries with the same tag via specific query SQL.
Related