Hi, partitioning by a generated column using date_trunc was already possible in version 3, but for downsampling you would probably be looking at using date_bin which was introduced in 4.7 so upgrading would be best if that is an option for you.
Take a look at Resampling time-series data with DATE_BIN - Tutorials - CrateDB Community
and CrateDB and Apache Airflow: Implementation of a Data Retention Policy - Tutorials - CrateDB Community.
2 Likes