| From: | David Rowley <david(dot)rowley(at)2ndquadrant(dot)com> |
|---|---|
| To: | Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Amit Khandekar <amitdkhan(dot)pg(at)gmail(dot)com>, Ashutosh Bapat <ashutosh(dot)bapat(at)enterprisedb(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: modeling parallel contention (was: Parallel Append implementation) |
| Date: | 2017-05-05 02:40:43 |
| Message-ID: | CAKJS1f-rFjZkYHDe1as2zuCxLTjN2i4-JCXHLuDZpCH6F6NkGw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 5 May 2017 at 14:36, Andres Freund <andres(at)anarazel(dot)de> wrote:
> I wonder how much doing the atomic ops approach alone can help, that
> doesn't have the issue that the work might be unevenly distributed
> between pages.
I wondered that too, since I though the barrier for making this change
would be lower by doing it that way.
I didn't manage to think of a way to get around the wrapping the
position back to 0 when synch-scans are involved.
i.e
parallel_scan->phs_cblock++;
if (parallel_scan->phs_cblock >= scan->rs_nblocks)
parallel_scan->phs_cblock = 0;
--
David Rowley http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andres Freund | 2017-05-05 02:45:33 | Re: modeling parallel contention (was: Parallel Append implementation) |
| Previous Message | David Rowley | 2017-05-05 02:36:58 | Re: modeling parallel contention (was: Parallel Append implementation) |