Re: priority on a process - Mailing list pgsql-general

From scott.marlowe
Subject Re: priority on a process
Date
Msg-id Pine.LNX.4.33.0305161347340.6494-100000@css120.ihs.com
Whole thread Raw
In response to Re: priority on a process  ("Chris Palmer" <chris.palmer@geneed.com>)
List pgsql-general
On Fri, 16 May 2003, Chris Palmer wrote:

> Scott Marlowe writes:
>
> > Also, if your machine has memory to spare, look at increasing
> > the effect cache size.
>
> Doesn't Linux automatically grow the fs cache/buffer to most of
> available memory? The OP said he was using Linux.
>
> On e.g. OpenBSD you would have to set the size in the kernel config, but
> Istr Linux doesn't need that. Granted, I haven't followed Linux closely
> since the 2.2 days, and I know there has been some VM hilarity in the
> "stable" 2.4 series. So maybe someone can update my old understanding...

Yes, effective_cache_size is the postgresql setting that tells the
postmaster we have "about this much kernel cache" on average.

If it's set low, then postgresql assumes the kernel isn't caching much, if
it's higher, then it assumes it's more likely for data to be "in memory"
and makes index scans more likely than seq scans.

Sorry, I should have pointed out I was talking out about a postgresql
configuration parameter and not a linux one...


pgsql-general by date:

Previous
From: DeJuan Jackson
Date:
Subject: Re: Rule on tabel
Next
From: Tom Lane
Date:
Subject: Re: priority on a process