Re: Auto Vacuum optimisation

[wenhui qiu]

Hi Japin
> Worth a try.
> However, I don't think we need to based on {vac,anl}_scale_factor when
using
> the sqrt algorithm.
  Here a new path, I removed the log2 algorithm and the debug2 log and the
enumeration type is also kept consistent, based on {vac,anl}_scale_factor
when using  the sqrt algorithm ,The main purpose is also to give DBAs the
flexibility to make small adjustments,Mainly referencing another discussion
as well, trying to stay in line with them

On Thu, Nov 28, 2024 at 1:46 PM Japin Li <japi...@hotmail.com> wrote:

> On Wed, 27 Nov 2024 at 15:51, wenhui qiu <qiuwenhu...@gmail.com> wrote:
> > HI Japin Li
> >      Thank you for you reply, Do you think that removing the log2
> algorithm because it will be more frequent,I think sql
> > server chose sqrt algorithm after their long-term verification, I think
> we need to retain sqrt algorithm,The relevant
> > documents are here(
> >
> https://learn.microsoft.com/en-us/sql/relational-databases/statistics/statistics?view=sql-server-ver16),I
> will provide a
> > new patch later with your  reply
> >
>
> Worth a try.
>
> However, I don't think we need to based on {vac,anl}_scale_factor when
> using
> the sqrt algorithm.
>
> --
> Regrads,
> Japin Li
>

0002-Add-guc-to-optimize-the-autovacuum-tigger-algorithm.patch
Description: Binary data