I can achive 140MBps to individual disks until I hit a 1GBps system ceiling
which I suspect 1GBps may be all that the 4x SAS HBA connection on a 3Gbps sas
expander can handle. (just a guess)
Anyway, with ZFS or SVM I can't do much beyond a single disk performance total
(if that) I am thinking
140MBps.
> What do you think?
That is about right per disk. I usually SWAG 100 +/- 50 MB/sec for HDD
media speed.
-- richard
>
> Thank you again for your help!
>
> --- On Thu, 7/29/10, Richard Elling wrote:
>
> From: Richard Elling
> Subject: Re: [zfs-discuss]
On Thu, Jul 29, 2010 at 6:04 PM, Carol wrote:
> Richard,
>
> I disconnected all but one path and disabled mpxio via stmsboot -d and my
> read performance doubled. I saw about 100MBps average from the pool.
>
> BTW, single harddrive performance (single disk in a pool) is about 140MBps.
>
> What d
Thu, 7/29/10, Richard Elling wrote:
From: Richard Elling
Subject: Re: [zfs-discuss] ZFS read performance terrible
To: "Carol"
Cc: "zfs-discuss@opensolaris.org"
Date: Thursday, July 29, 2010, 2:03 PM
On Jul 29, 2010, at 9:57 AM, Carol wrote:
> Yes I noticed that threa
> Yep. With round robin it's about 80 for each disk for ascv_t
Any ideas?
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Yes I noticed that thread a while back and have been doing a great deal of
testing with various scsi_vhci options.
I am disappointed that the thread hasn't moved further since I also suspect
that it is related to mpt-sas or multipath or expander related.
I was able to get aggregate writes up t
I'm about to do some testing with that dtrace script..
However, in the meantime - I've disabled primarycache (set primarycache=none)
since I noticed that it was easily caching /dev/zero and I wanted to do some
tests within the OS rather than over FC.
I am getting the same results through dd.
Vi
Yes because the author was too smart for his own good and ssd is for Sparc, you
use SD. Delete all the ssd lines. Here's that script which will work for you
provided it doesn't get wrapped or otherwise maligned by this html interface:
#!/usr/sbin/dtrace -s
#pragma D option quiet
fbt:sd:sdstrat
> You should look at your disk IO patterns which will
> likely lead you to find unset IO queues in sd.conf.
> Look at this
> http://blogs.sun.com/chrisg/entry/latency_bubble_in_yo
> ur_io as a place to start.
Any idea why I would get this message from the dtrace script?
(I'm new to dtrace / open
Good idea.
I will keep this test in mind - I'd do it immediately except for the fact that
it would be somewhat difficult to connect power to the drives considering the
design of my chassis, but I'm sure I can figure something out if it comes to
it...
--
This message posted from opensolaris.org
You should look at your disk IO patterns which will likely lead you to find
unset IO queues in sd.conf. Look at this
http://blogs.sun.com/chrisg/entry/latency_bubble_in_your_io as a place to
start. The parameter you can try to set globally (bad idea) is done by doing
echo zfs_vdev_max_pending/W
Hello Karol,
you wrote at, 29. Juli 2010 02:23:
> I appear to be getting between 2-9MB/s reads from individual disks
It sounds for me that you have a hardware failure because 2-9 MB/s
are less than dropping.
> 2x LSI 9200-8e SAS HBAs (2008 chipset)
> Supermicro 846e2 enclosure with LSI sasx36 e
Hi Robert -
I tried all of your suggestions but unfortunately my performance did not
improve.
I tested single disk performance and I get 120-140MBps read/write to a single
disk. As soon as I add an additional disk (mirror, stripe, raidz) , my
performance drops significantly.
I'm using 8Gbit F
Actually writes faster then reads are typical fora Copy on Write FC (or Write
Anywhere). I usually describe it like this.
CoW in ZFS works like when you come home after a long day and you ust want to
go to bed. You take of one pice of clothing after another and drop it on the
floor just where
On Jul 29, 2010, at 9:57 AM, Carol wrote:
> Yes I noticed that thread a while back and have been doing a great deal of
> testing with various scsi_vhci options.
> I am disappointed that the thread hasn't moved further since I also suspect
> that it is related to mpt-sas or multipath or expande
Yes I noticed that thread a while back and have been doing a great deal of
testing with various scsi_vhci options.
I am disappointed that the thread hasn't moved further since I also suspect
that it is related to mpt-sas or multipath or expander related.
I was able to get aggregate writes up t
This sounds very similar to another post last month.
http://opensolaris.org/jive/thread.jspa?messageID=487453
The trouble appears to be below ZFS, so you might try asking on the
storage-discuss forum.
-- richard
On Jul 28, 2010, at 5:23 PM, Karol wrote:
> I appear to be getting between 2-9MB/s
>Hi r2ch
>The operations column shows about 370 operations for read - per spindle
>(Between 400-900 for writes)
>How should I be measuring iops?
It seems to me then that your spindles are going about as fast as they can and
your just moving small block sizes.
There are lots of ways to test for
> Update to my own post. Further tests more
> consistently resulted in closer to 150MB/s.
>
> When I took one disk offline, it was just shy of
> 100MB/s on the single disk. There is both an obvious
> improvement with the mirror, and a trade-off (perhaps
> the latter is controller related?).
>
>
Hi r2ch
The operations column shows about 370 operations for read - per spindle
(Between 400-900 for writes)
How should I be measuring iops?
--
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail
How many iops per spindle are you getting?
A rule of thumb I use is to expect no more than 125 iops per spindle for
regular HDDs.
SSDs are a different story of course. :)
--
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-dis
I appear to be getting between 2-9MB/s reads from individual disks in my zpool
as shown in iostat -v
I expect upwards of 100MBps per disk, or at least aggregate performance on par
with the number of disks that I have.
My configuration is as follows:
Two Quad-core 5520 processors
48GB ECC/REG ra
22 matches
Mail list logo
