| From: | Peter Geoghegan <pg(at)heroku(dot)com> |
|---|---|
| To: | David Rowley <david(dot)rowley(at)2ndquadrant(dot)com> |
| Cc: | Andres Freund <andres(at)anarazel(dot)de>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Memory prefetching while sequentially fetching from SortTuple array, tuplestore |
| Date: | 2015-12-13 21:30:02 |
| Message-ID: | CAM3SWZTM5jz_=cpRqXYxtbsoyNw8VyzKFPYfstRetAuvk5pqOg@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Mon, Nov 30, 2015 at 1:04 PM, Peter Geoghegan <pg(at)heroku(dot)com> wrote:
> Perhaps we can consider more selectively applying prefetching in the
> context of writing out tuples.
It may still be worth selectively applying these techniques to writing
out tuples, per my previous remarks (and the latest "Using quicksort
for every external sort run" revision, which has a specialized version
of this patch). I see real benefits there across a variety of
situations on different hardware, and I'm hesitant to give that up
without further analysis.
However, I think that this patch is over as an independent piece of
work -- there is too much of a mixed picture. I'm going to mark this
patch "returned with feedback".
--
Peter Geoghegan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jeff Janes | 2015-12-13 22:43:55 | Re: Using quicksort for every external sort run |
| Previous Message | Tomas Vondra | 2015-12-13 19:38:05 | Re: PATCH: add pg_current_xlog_flush_location function |