Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 16 Jun 2013 10:20:00 +0200
From:      Attila Nagy <bra@fsn.hu>
To:        Mark Felder <feld@feld.me>
Cc:        freebsd-fs@freebsd.org
Subject:   Re: An order of magnitude higher IOPS needed with ZFS than UFS
Message-ID:  <51BD7530.9060000@fsn.hu>
In-Reply-To: <op.wykdduw834t2sn@markf.office.supranet.net>
References:  <51B79023.5020109@fsn.hu> <op.wykdduw834t2sn@markf.office.supranet.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 06/12/13 13:40, Mark Felder wrote:
> On Tue, 11 Jun 2013 16:01:23 -0500, Attila Nagy <bra@fsn.hu> wrote:
>
>> BTW, the file systems are 77-78% full according to df (so ZFS holds 
>> more, because UFS is -m 8).
>
> ZFS write performance can begin to drop pretty badly when you get 
> around 80% full. I've not seen any benchmarks showing an improvement 
> with a very fast and large ZIL or tons of memory, but I'd expect that 
> would help significantly. Just note that you're right at the edge 
> where performance gets impacted.
I'm aware of that, even thought about it, that's why I wrote the disk 
free percents into my mail. I see a completely normal write pattern, but 
a lot of reads, so I guess ZIL wouldn't help here (maybe L2ARC or more 
memory, if something doesn't fit into it).

BTW, I think having the allowed fill value this low (UFS has the 
breakdown point much higher) is of a really bad design...



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?51BD7530.9060000>