Date: Tue, 22 May 2007 16:59:43 +0200 From: Solon Luigi Lutz <solon@pyro.de> To: freebsd-current@freebsd.org Subject: Re[2]: ZFS committed to the FreeBSD base. Message-ID: <204195092.20070522165943@pyro.de> In-Reply-To: <m1myzxrzda.fsf@binarysolutions.dk> References: <20070407191517.GN63916@garage.freebsd.pl> <20070407212413.GK8831@cicely12.cicely.de> <20070410003505.GA8189@nowhere> <20070410003837.GB8189@nowhere> <20070410011125.GB38535@xor.obsecurity.org> <20070410013034.GC8189@nowhere> <20070410014233.GD8189@nowhere> <4651BD6F.5050301@unsane.co.uk> <20070522083112.GA5136@hub.freebsd.org> <4652B15D.5060505@unsane.co.uk> <20070522090947.GA3005@garage.freebsd.pl> <m1myzxrzda.fsf@binarysolutions.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
KVS> Pawel Jakub Dawidek <pjd@FreeBSD.org> writes:
>>> I may reinstall at a later date as this is still very much a box to play
>>> with, but I gather there is no great gain from going 64 bit other than
>>> not having to play with PAE if you've got lots or RAM.
>>
>> I expect there is a huge difference in performance between i386 and
>> amd64. I'm currently setting up environment to compare ZFS on
>> FreeBSD/i386, FreeBSD/amd64 and Solaris.
KVS> I've had precious little time to do more testing on our amd64-setup, but
KVS> it seems that vm.kmem.size_max is a 32-bit uint, so we can't really use
KVS> much RAM for ZFS.
These are the figures for the following out-of-the-box 7.0 amd64 smp system:
Athlon X2 3800, 1GB Ram, Asus M2N-SLI (modified), 24x 500GB (Samsung Spinpoint),
Areca ARC1280 running RAID-6.
ZFS
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
8000 89555 85.6 188475 63.3 114048 36.9 95865 97.2 460375 64.0 41.2 0.3
Test was done on a 10TB volume.
Since the machine is running in semi-production mode, I can't perform any UFS
tests anymore. The systems is running stable, and extensive
checksumming after 8TB data transfers didn't reveal any errors,
neither did a filesystem-stress-test. Checksumming with 'cfv' runs at a sustained
data rate of 290MB/s
Thanks for the good work!
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?204195092.20070522165943>