From: | "scott(dot)marlowe" <scott(dot)marlowe(at)ihs(dot)com> |
---|---|
To: | Robert Creager <Robert_Creager(at)LogicalChaos(dot)org> |
Cc: | pgsql-performance <pgsql-performance(at)postgresql(dot)org> |
Subject: | Re: Enabling and Disabling Sequencial Scan |
Date: | 2003-06-02 15:23:11 |
Message-ID: | Pine.LNX.4.33.0306020921070.11997-100000@css120.ihs.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On Fri, 30 May 2003, Robert Creager wrote:
> On Fri, 30 May 2003 14:46:12 -0600 (MDT)
> "scott.marlowe" <scott(dot)marlowe(at)ihs(dot)com> said something like:
>
> >
> > level cache is. On my machine it's about 800 meg. It's measured in 8k
> > blocks, so 100,000 * 8k ~ 800 meg. The smaller this is, the more
>
> My 'Cached' usage is 1.7Gb. I've hit the kernel mailing list, and the
> one response I got said don't worry about it :-(
Oh, yeah, just a bit on that. as far as the kernel developers are
concerned, the buffer / cache is working perfectly, and they're right, it
is. What they probably don't understand if your need to tell postgresql
how much cache/buffer is allocated to it.
so don't worry about the kernel, the linux kernel really is pretty good at
caching disk access.
From | Date | Subject | |
---|---|---|---|
Next Message | scott.marlowe | 2003-06-02 16:34:43 | Re: Degrading performance |
Previous Message | scott.marlowe | 2003-06-02 15:20:28 | Re: Enabling and Disabling Sequencial Scan |