Re: Set access strategy for parallel vacuum workers

From: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
To: Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
Cc: PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Peter Geoghegan <pg(at)bowt(dot)ie>
Subject: Re: Set access strategy for parallel vacuum workers
Date: 2021-04-08 03:24:58
Message-ID: CAA4eK1JGvsuW9Q+mgzhMqnxSZu8g-=Un4MfUNuYVfozU_UKZ3g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, Apr 8, 2021 at 8:52 AM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> wrote:
>
> On Thu, Apr 8, 2021 at 12:17 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> >
> > On Wed, Apr 7, 2021 at 5:11 PM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> wrote:
> > >
> > > On Wed, Apr 7, 2021 at 7:00 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> > > >
> > > > During recent developments in the vacuum, it has been noticed [1] that
> > > > parallel vacuum workers don't use any buffer access strategy. I think
> > > > we can fix it either by propagating the required information from the
> > > > leader or just get the access strategy in each worker separately. The
> > > > patches for both approaches for PG-13 are attached.
> > >
> > > Thank you for starting the new thread.
> > >
> > > I'd prefer to just have parallel vacuum workers get BAS_VACUUM buffer
> > > access strategy. If we want to have set different buffer access
> > > strategies or ring buffer sizes for the leader and worker processes,
> > > the former approach would be better. But I think we're unlikely to
> > > want to do that, especially in back branches.
> > >
> >
> > Fair enough. Just to be clear, you prefer to go with
> > fix_access_strategy_workers_11.patch, right?
>
> That's right.
>

Okay, I'll wait for a day or two to see if anyone else has comments or
suggestions.

--
With Regards,
Amit Kapila.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2021-04-08 03:27:04 Re: Feature improvement: can we add queryId for pg_catalog.pg_stat_activity view?
Previous Message Alvaro Herrera 2021-04-08 03:22:35 Re: Autovacuum on partitioned table (autoanalyze)