Re: BBU Cache vs. spindles - Mailing list pgsql-performance

From Bruce Momjian
Subject Re: BBU Cache vs. spindles
Date
Msg-id 201010211659.o9LGxkV01781@momjian.us
Whole thread Raw
In response to Re: BBU Cache vs. spindles  ("Kevin Grittner" <Kevin.Grittner@wicourts.gov>)
Responses Re: BBU Cache vs. spindles
List pgsql-performance
Kevin Grittner wrote:
> Bruce Momjian <bruce@momjian.us> wrote:
>
> > With a BBU you can turn off full_page_writes
>
> My understanding is that that is not without risk.  What happens if
> the WAL is written, there is a commit, but the data page has not yet
> been written to the controller?  Don't we still have a torn page?

I don't see how full_page_writes affect non-written pages to the
controller.

full_page_writes is designed to guard against a partial write to a
device.  I don't think the raid cache can be partially written to, and
the cache will not be cleared until the drive has fully writen the data
to disk.

--
  Bruce Momjian  <bruce@momjian.us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

  + It's impossible for everything to be true. +

pgsql-performance by date:

Previous
From: "Kevin Grittner"
Date:
Subject: Re: BBU Cache vs. spindles
Next
From: Greg Smith
Date:
Subject: Re: BBU Cache vs. spindles