Re: UPDATEDs slowing SELECTs in a fully cached database

From: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
To: lars <lhofhansl(at)yahoo(dot)com>
Cc: Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, Ivan Voras <ivoras(at)freebsd(dot)org>, pgsql-performance(at)postgresql(dot)org
Subject: Re: UPDATEDs slowing SELECTs in a fully cached database
Date: 2011-07-14 23:03:43
Message-ID: CAMkU=1z2oriZAv4D2hb55+PX=Q5j8dfxpaZpHhGuYic168B18A@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Wed, Jul 13, 2011 at 3:41 PM, lars <lhofhansl(at)yahoo(dot)com> wrote:
> On 07/13/2011 11:42 AM, Kevin Grittner wrote:
>>
>> So transactions without an XID *are* sensitive to
>> synchronous_commit.  That's likely a useful clue.
>>
>> How much did it help the run time of the SELECT which followed the
>> UPDATE?
>
> It has surprisingly little impact on the SELECT side:

If your fsync is truly fsyncing, it seems like it should have
considerable effect.

Could you strace with both -ttt and -T, with and without synchronous commit?

Cheers,

Jeff

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message chris 2011-07-15 06:34:24 Hardware advice for scalable warehouse db
Previous Message Jeff Janes 2011-07-14 21:45:43 Re: UPDATEDs slowing SELECTs in a fully cached database