| From: | Andres Freund <andres(at)anarazel(dot)de> |
|---|---|
| To: | pgsql-general(at)postgresql(dot)org |
| Cc: | "Henry C(dot)" <henka(at)cityweb(dot)co(dot)za>, "Toby Corkindale" <toby(dot)corkindale(at)strategicdata(dot)com(dot)au>, "luv-main" <luv-main(at)luv(dot)asn(dot)au> |
| Subject: | Re: Poor performance of btrfs with Postgresql |
| Date: | 2011-04-21 13:03:58 |
| Message-ID: | 201104211503.59149.andres@anarazel.de |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Thursday, April 21, 2011 12:16:04 PM Henry C. wrote:
> > I've done some testing of PostgreSQL on different filesystems, and with
> > different filesystem mount options.
>
> Since Pg is already "journalling", why bother duplicating (and pay the
> performance penalty, whatever that penalty may be) the effort for no real
> gain (except maybe a redundant sense of safety)? ie, use a
> non-journalling battle-tested fs like ext2.
Don't. The fsck on reboot will eat way too much time.
Using metadata only journaling is ok though. In my opinion the problem with
btrfs is more the overhead of COW, but thats an impression from several kernel
version ago, so...
Andres
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tiruvenkatasamy Baskaran | 2011-04-21 13:05:35 | Re: Which version of postgresql supports replication on RHEL6? |
| Previous Message | Pavel Stehule | 2011-04-21 12:31:57 | Re: [HACKERS] Defining input function for new datatype |