Alex,
Are you using a Clariion or a Symetrix? There is a big difference in how
these are configured.
I mainly have experience setting up Clariions.
A Meta is probably a Meta LUN – made up of multiple LUN's. The Meta-LUN
allow you to add drives to LUN without having to reconfigure the OS.
As for performance, it seems that you do have an I/O issue. This is
probably due to not enough spindles. You really need more disks. The
Meta-LUNs and SAME just try and spread the load across all of your available
spindles.
If you are using 120 Disks in a RAID 1+0 configuration with 256 GB drives,
you probably have 10k speed spindles. This is slow in this day and age.
You can do approximately 100 IOPS per disk for a total of 12000 IOPS. If
your applications require more than that, you have an I/O issue. SAME or
Meta will not solve the underlying issue of not enough disks.
You really need to size your SAN these days based on I/O's per second
(IOPS) not space. The disks are getting larger, but not necessarily
faster. Check to make sure your cache is turned on. I have seen huge
performance hits from the Clariion cache being disabled.
If you have a symetrix, everything is a slice across all disks in the
box. That is just the way it works. Of course you have a much larger cache
with the Sym too, but that's another discussion in itself.
I hope this helps,
Bryan
Bryan Thomas
Senior Consultant and Practice Manager
Performance Tuning Corporation
www.perftuning.com
------------------------------
*From:* oracle-l-bounce@xxxxxxxxxxxxx [mailto:
oracle-l-bounce@xxxxxxxxxxxxx] *On Behalf Of *amonte
*Sent:* Thursday, February 01, 2007 2:52 PM
*To:* oracle-l@xxxxxxxxxxxxx
*Subject:* How many of you use S.A.M.E?
Hi
Anyone here use Stripe.All.Mirror.Everything methology? The one claimed by
Mr Loaiza 6 years ago?
Not long ago I posted to list several I/O problems I have with Storage. I
was told yesterday that the EMC we have is using Meta Devices, some sort of
SAME. I was told by the EMC guy that the SAN has 120 disks with RAID 1+0,
256GB disks. In the factory it is configured with a internal stripe, 256GB
disk is divided into slices of 8GB. So we have 120 disks each disk with 32
slices.
With that configuration you take slices from several physical disks to
form LUNs. He syas that this is what they call Mete Devices and usually they
do that only when the customer is in a hurry and needs the disks badly and
dont have time to make a better planning. He further said that this sort of
configuration he has seen response times of over 120ms and it is not
unusual. The porblem with our disks is that sometimes we get like 80ms
responsetime and most of time beteen 25 and 60.
I wonder, is Mete Device a SAME configuration? Sure it is deadly easy to
configure and maintain but then we are suffering performance problems. The
EMC box is used by over 10 databases, from OLTP to DWH and some of them
Hybrid configuration. So the chance that a disk is used by all 10 databases
is quite easy. Wont the disk head go crazy when 10 databases is asking for
data in several sectors in the same disk? May the chance of that is quite
small however I would not be surprised that a disk is being used by 3, 4
databases concurrently.
TIA
Alex
