Re: shared_buffers/effective_cache_size on 96GB server

From: Claudio Freire <klaussfreire(at)gmail(dot)com>
To: sthomas(at)optionshouse(dot)com
Cc: Bruce Momjian <bruce(at)momjian(dot)us>, Julien Cigar <jcigar(at)ulb(dot)ac(dot)be>, pgsql-performance(at)postgresql(dot)org
Subject: Re: shared_buffers/effective_cache_size on 96GB server
Date: 2012-10-10 18:24:42
Message-ID: CAGTBQpbR35EDtEBcUepJCzU03px8=rHZXUvwCCbKnadFtyDfTA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Wed, Oct 10, 2012 at 3:18 PM, Shaun Thomas <sthomas(at)optionshouse(dot)com> wrote:
>> Why does nobody every mention that concurrent access has to be taken
>> into account?
>
>
> That's actually a good point. But if you have one giant database, the
> overlap of which tables are being accessed by various sessions is going to
> be immense.

That's why I said "several huge indices". If regularly accessed
indices are separate, and big, it means they don't overlap nor do they
fit in any cache.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Shane Hathaway 2012-10-10 18:25:54 Re: Ways to speed up ts_rank
Previous Message Shaun Thomas 2012-10-10 18:18:49 Re: shared_buffers/effective_cache_size on 96GB server