Re: EXPLAIN VERBOSE with parallel Aggregate - Mailing list pgsql-hackers

From Tom Lane
Subject Re: EXPLAIN VERBOSE with parallel Aggregate
Date
Msg-id 12585.1460737650@sss.pgh.pa.us
Whole thread Raw
In response to Re: EXPLAIN VERBOSE with parallel Aggregate  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: EXPLAIN VERBOSE with parallel Aggregate  (David Rowley <david.rowley@2ndquadrant.com>)
List pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
> I definitely agree that the current output is messed up, but I'm not
> sure your proposed output is much better.  I wonder if it shouldn't
> say something like:
> Output: serialfn(transfn(args))
> for the partial aggregate and
> Output: finalfn(combinefn(deserialfn(args)))
> for the finalize aggregate step.

> Or maybe just insert the word PARTIAL before each partial aggregate
> step, like PARTIAL sum(num) for the partial step and then just
> sum(num) for the final step.

+1 for the latter, if we can do it conveniently.  I think exposing
the names of the aggregate implementation functions would be very
user-unfriendly, as nobody but us hackers knows what those are.

> I think ending up with sum(sum(num)) is
> right out.  It doesn't look so bad for that case but avg(avg(num))
> would certainly imply something that's not the actual behavior.

Agreed.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Teodor Sigaev
Date:
Subject: Re: GIN data corruption bug(s) in 9.6devel
Next
From: "David G. Johnston"
Date:
Subject: Re: SET ROLE and reserved roles