Re: Avoiding sub-query with group by

Or maybe those scans are not nested and so it's O(n), not O(n^2)?

Sorry, I am not experienced using "explain".

Andrew

>                              QUERY PLAN
> ---------------------------------------------------------------------
>  Hash Join  (cost=1.16..2.27 rows=1 width=24)
>    Hash Cond: ((public.tbl.val = x.v) AND (public.tbl.rmp = x.mr))
>    ->  Seq Scan on tbl  (cost=0.00..1.06 rows=6 width=16)
>    ->  Hash  (cost=1.13..1.13 rows=2 width=8)
>          ->  HashAggregate  (cost=1.08..1.11 rows=2 width=8)
>                ->  Seq Scan on tbl  (cost=0.00..1.07 rows=2 width=8)
>                      Filter: (rmp > val)