Re: PG 7.0 is 2.5 times slower running a big report - Mailing list pgsql-general

From Alfred Perlstein
Subject Re: PG 7.0 is 2.5 times slower running a big report
Date
Msg-id 20000526034107.S28594@fw.wintelcom.net
Whole thread Raw
In response to Re: PG 7.0 is 2.5 times slower running a big report  ("Matthias Urlichs" <smurf@noris.de>)
Responses Re: PG 7.0 is 2.5 times slower running a big report  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-general
* Matthias Urlichs <smurf@noris.de> [000526 02:47] wrote:
> Hi,
>
> Bruce Momjian:
> > As I remember, Linux turns off file system prefetch if a seek is done.
>
> Correct (still so for Linux 2.4).
>
> > BSD file systems turn off prefetch only if prefetched blocks remain
> > unused.
>
> "Unused" is decided when?
> Do any, all, or some percentage of prefetched blocks need to be unused?

The statement:

> > BSD file systems turn off prefetch only if prefetched blocks remain
> > unused.

is false for FreeBSD afaik.

FreeBSD tracks how often sequential access happens and will retry it
as the file's access trend indicates.  It can flip back and forth from
readahead and non-readahead.

--
-Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org]
"I have the heart of a child; I keep it in a jar on my desk."

pgsql-general by date:

Previous
From: Giles Lean
Date:
Subject: Re: createdb -- alternate locations
Next
From: Bill Barnes
Date:
Subject: Windoze port of version 7