This message was deleted.

Druid has compaction ability so it can do that "combining and aggregating" itself. https://druid.apache.org/docs/latest/data-management/compaction

@James O'Claire check Druid rollups https://druid.apache.org/docs/latest/ingestion/rollup It allows aggregations at ingestion time (counts, min/max, sum, quantiles), so it might even eliminate your need for a Python processing. Pre-aggregating live data at ingestion time is a great use case of Druid.