From: | Tim Ellis <Tim(dot)Ellis(at)gamet(dot)com> |
---|---|
To: | Curt Sampson <cjs(at)cynic(dot)net> |
Cc: | stigmata(at)blackangel(dot)net, pgsql-admin(at)postgresql(dot)org |
Subject: | Re: memory strangeness (fwd) |
Date: | 2002-07-08 18:23:04 |
Message-ID: | 20020708112304.733fbe30.Tim.Ellis@gamet.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
On Sun, 7 Jul 2002 13:06:00 +0900 (JST)
Curt Sampson <cjs(at)cynic(dot)net> wrote:
> > On this particular system, though, it's doing
> > nothing except PG. 384 MB of RAM, I can give PG 160 of it, which
> > leaves me with some 170 MB of idle RAM.
>
> No, that's not idle RAM; that's buffer cache. (FreeBSD, like most
> modern Unix operating systems, will use any spare physical memory
> to cache blocks read from the disk.) If you allocate 160/170, you
> now have pretty much maximized your chances that postgres and the
> operating system will be buffering the same data, and made your
> memory as minimally effective as possible.
Is there not a method of reading a disk block and instructing the OS not
to buffer it, or am I hallucinating? (My main RDBMS experience experience
is on Solaris, and I could swear that major RDBMSs on that OS will ask the
OS not to buffer disk I/O on behalf of the database)
--
Tim Ellis
DBA, Gamet
From | Date | Subject | |
---|---|---|---|
Next Message | becks | 2002-07-08 18:39:07 | disk space |
Previous Message | Bruce Momjian | 2002-07-08 18:13:06 | Re: PostgreSQL data -> Oracle |