On Thu, 23 Sept 2021 at 08:21, Michael Lewis <mlewis@xxxxxxxxxxx> wrote: > select station, array_agg(distinct(channel)) as channels > FROM( > SELECT station,channel FROM data GROUP BY station,channel > ) AS sub > group by station; Since the subquery is grouping by station, channel, then there's no need for the DISTINCT in the aggregate function. Removing that should remove some tuplestore overhead from the aggregate node. David
