From: | Ants Aasma <ants(dot)aasma(at)eesti(dot)ee> |
---|---|
To: | adrien(dot)nayrat(at)anayrat(dot)info |
Cc: | hexexpert(at)comcast(dot)net, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Skylake-S warning |
Date: | 2018-10-04 07:49:40 |
Message-ID: | CA+CSw_tuynrscN+VkKPSKTufETkcugreZf1BUvQnhfaLVpm6yA@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Thu, Oct 4, 2018 at 9:50 AM Adrien Nayrat <adrien(dot)nayrat(at)anayrat(dot)info> wrote:
>
> On 10/3/18 11:29 PM, Daniel Wood wrote:
> > If running benchmarks or you are a customer which is currently impacted by
> > GetSnapshotData() on high end multisocket systems be wary of Skylake-S.
> >
> >
> > Performance differences of nearly 2X can be seen on select only pgbench due to
> > nothing else but unlucky choices for max_connections. Scale 1000, 192 local
> > clients on a 2 socket 48 core Skylake-S(Xeon Platinum 8175M @ 2.50-GHz) system.
> > pgbench -S
>
> Could it be related to :
> https://www.postgresql.org/message-id/D2B9F2A20670C84685EF7D183F2949E2373E66%40gigant.nidsa.net
> ?
Unlikely. I understood from Daniel's email that profiling shows a
different hot-spot. In the cited .NET issue the problem was mostly due
to issuing PAUSE in a loop without attempting to grab the lock. In
PostgreSQL it's called only once per retry attempt.
Regards,
Ants Aasma
--
PostgreSQL Senior Consultant
www.cybertec-postgresql.com
Austria (HQ), Wiener Neustadt | Switzerland, Zürich | Estonia,
Tallinn | Uruguay, Montevideo
Facebook: www.fb.com/cybertec.postgresql
Twitter: www.twitter.com/PostgresSupport
From | Date | Subject | |
---|---|---|---|
Next Message | Amit Langote | 2018-10-04 07:53:02 | Re: partition tree inspection functions |
Previous Message | Amit Langote | 2018-10-04 07:49:31 | Re: partition tree inspection functions |