From: | Bruce Momjian <bruce(at)momjian(dot)us> |
---|---|
To: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
Cc: | Jeff Janes <jeff(dot)janes(at)gmail(dot)com>, Jeremy Harris <jgh(at)wizmail(dot)org>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Minor performance improvement in transition to external sort |
Date: | 2014-04-16 23:38:53 |
Message-ID: | 20140416233853.GS7443@momjian.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Thu, Apr 10, 2014 at 06:03:15PM +0100, Simon Riggs wrote:
> On 6 February 2014 18:21, Jeff Janes <jeff(dot)janes(at)gmail(dot)com> wrote:
> > On Tue, Feb 4, 2014 at 2:22 PM, Jeremy Harris <jgh(at)wizmail(dot)org> wrote:
> >>
> >> The attached patch replaces the existing siftup method for heapify with
> >> a siftdown method. Tested with random integers it does 18% fewer
> >> compares and takes 10% less time for the heapify, over the work_mem
> >> range 1024 to 1048576.
> >
> >
> > Thanks for working on this.
>
> +1
>
> Your patch isn't linked properly from the CF manager though.
>
> If you like patches like this then there's a long(er) list of
> optimizations already proposed previously around sorting. It would be
> good to have someone work through them for external sorts. I believe
> Noah is working on parallel internal sort (as an aside).
>
> There's also an optimization possible for merge joins where we use the
> output of the first sort as an additional filter on the second sort.
> That can help when we're going to join two disjoint tables.
Where should this be recorded? TODO? Commitfest manager?
--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com
+ Everyone has their own god. +
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2014-04-16 23:48:06 | Re: [BUG FIX] Compare returned value by socket() against PGINVALID_SOCKET instead of < 0 |
Previous Message | Tom Lane | 2014-04-16 23:35:58 | Re: [COMMITTERS] pgsql: Include planning time in EXPLAIN ANALYZE output. |