Re: Advice needed: query performance deteriorates by 2000% within 1 minute

From: Michael Lewis <mlewis(at)entrata(dot)com>
To: Peter Adlersburg <peter(dot)adlersburg(at)gmail(dot)com>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Pgsql Performance <pgsql-performance(at)lists(dot)postgresql(dot)org>
Subject: Re: Advice needed: query performance deteriorates by 2000% within 1 minute
Date: 2022-03-02 01:44:48
Message-ID: CAHOFxGrhHE055kjkRaj2xcFQRH=vXHRW9XVM=RN0Tj8scL8=7w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

And I would absolutely crank up autovacuum and analyze settings. Turn up
the cost limits, turn down the cost delays, decrease the scale factor.
Whatever you need to do such that autovacuum runs often. No need to
schedule a manual vacuum at all. Just don't wait until 20% of the table is
dead before an autovacuum is triggered like the default behavior. The cost
to gather new stats and do garbage collection is rather minimal compared to
the benefit to queries that rely on the data in many cases.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Michel SALAIS 2022-03-02 07:28:34 RE: Simple task with partitioning which I can't realize
Previous Message Michael Lewis 2022-03-02 01:39:40 Re: Advice needed: query performance deteriorates by 2000% within 1 minute