Peter Viertel wrote:
Who says ufs only goes to 2TB? Nobody told my 5TB one it can't exist.
;-)
I seem to recall that going over 1Tb required many patched on Solaris
9. Going above 2Tb was not possible with regular LUNs. I think I read
that it was possible over some kind of LVM (DiskSuite ?).
For the record - ufs on a vxvm volume can only go to 2TB. Ufs on a
metadisk can be considerably larger - up to around 16TB if you're crazy
enough to try it. Performance is almost comparable to zfs when you
treat it carefully - eg you keep the filecount down, you tune to
maxcontig to match raid strip widths on the storage, and you only delete
data in large tranches..
I had two 2Tb LUNs on an AX150 and performance was rather poor,
especially when I was trying to stage while running a backup). About a
year ago (March 2007), I asked about this problem and you replied with
the following:
"Co-incidently I am looking at sequential I/O issues on a Sun Fire 6900
which comes down to UFS - benchmarked UFS at 20Mb/s, ZFS at 210Mb/s on
the same device (DMX-3 LUN). Yes - 10 fold improvement and it wasnt
filesystem caching."
Care to share your ZFS tuning ?
Zfs gives you more chances to use your hardware to its best potential -
but you can still make it go really slow if you don't treat it well.
Try running it at 98% full for a while with lots of random file adds and
deletes..
This isn't meant to criticise zfs or ufs, or say that vxfs or qfs are
'better', but the reality is that you can make any of them work well
with care, and you can make all of them perform dreadfully with a little
effort also.
Overall IMHO - zfs and qfs are better positioned for the future of
storage than the other filesystem types, and both should be considered
if you're looking to deploy an adv_file solaris system.
Sun came out with a little whitepaper this month on using EBS on a
thumper here;
http://www.sun.com/bigadmin/sundocs/articles/sfx4500_ebs_storagenode.pdf
It doesn't spend any time at all going into zfs performance though...
-----Original Message-----
From: EMC NetWorker discussion
[mailto:NETWORKER AT LISTSERV.TEMPLE DOT EDU] On Behalf Of Yaron Zabary
Sent: Tuesday, 29 April 2008 11:21 AM
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Subject: Re: [Networker] recommendations for networker server upgrade
ZFS seems attractive for adv_file devices. It supports
multi-TB file
systems (UFS is limited to 2Tb). Performance seems OK,
although I hoped
for better results. When I was running at Solaris 9, I had to use UFS
which had miserable performance.
NOTICE
This e-mail and any attachments are confidential and may contain copyright
material of Macquarie Group Limited or third parties. If you are not the
intended recipient of this email you should not read, print, re-transmit, store
or act in reliance on this e-mail or any attachments, and should destroy all
copies of them. Macquarie Group Limited does not guarantee the integrity of any
emails or any attached files. The views or opinions expressed are the author's
own and may not reflect the views or opinions of Macquarie Group Limited.
To sign off this list, send email to listserv AT listserv.temple DOT edu and type
"signoff networker" in the body of the email. Please write to networker-request
AT listserv.temple DOT edu if you have any problems with this list. You can access the
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
--
-- Yaron.
To sign off this list, send email to listserv AT listserv.temple DOT edu and type
"signoff networker" in the body of the email. Please write to networker-request
AT listserv.temple DOT edu if you have any problems with this list. You can access the
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
| 