From: | David Rowley <david(dot)rowley(at)2ndquadrant(dot)com> |
---|---|
To: | James Sewell <james(dot)sewell(at)lisasoft(dot)com> |
Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Parallel Aggregate |
Date: | 2016-03-14 04:20:53 |
Message-ID: | CAKJS1f8zy0jZxKa=T0VAy3Sm1xL=SjwE2VgSqmeOHA9erYf=mQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 14 March 2016 at 17:05, James Sewell <james(dot)sewell(at)lisasoft(dot)com> wrote:
>
> Hi again,
>
> I've been playing around with inheritance combined with this patch. Currently it looks like you are taking max(parallel_degree) from all the child tables and using that for the number of workers.
>
> For large machines it makes much more sense to use sum(parallel_degree) - but I've just seen this comment in the code:
>
> /*
> * Decide what parallel degree to request for this append path. For
> * now, we just use the maximum parallel degree of any member. It
> * might be useful to use a higher number if the Append node were
> * smart enough to spread out the workers, but it currently isn't.
> */
>
> Does this mean that even though we are aggregating in parallel, we are only operating on one child table at a time currently?
There is nothing in the planner yet, or any patch that I know of to
push the Partial Aggregate node to below an Append node. That will
most likely come in 9.7.
--
David Rowley http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
From | Date | Subject | |
---|---|---|---|
Next Message | David Rowley | 2016-03-14 04:45:55 | Re: Combining Aggregates |
Previous Message | James Sewell | 2016-03-14 04:18:58 | Re: Parallel Aggregate |