Good points, some comments:
- You have to (by default) allocate 2x + deltas for LUNs. For a
100GB LUN, you'd need 240GB of space (by default--I know there are workarounds with ONTAP 7.2 and later). With an NFS mount (which is nothing more than exporting a FlexVol), you only need to account for the 20% Snapshot reserve...
Be careful! The reason you need 2x + deltas is to simply ensure that you maintain enough space in the volume to guarantee that you'll be able to overwrite data if you decide to overwrite the entire LUN (say, if you format the drive while a snapshot was there, or maybe re-indexed or defragged the NTFS volume in a VMFS filesystem?) Just because you choose a 20% snap reserve in a NAS volume doesn't mean that you won't blow through that in this case in NFS, and end up with the same problem with "out of space"- you'll do a DF and see that your snapshot usage is like 1000% while your regular fs usage is only 40% or something like that. And since netapp is all about preserving data in all cases, you won't be able to do anything until you delete snaps. Bottom line is that you don't get a totally free lunch with NFS, although the thin provisioning DEFINITELY helps. Don't forget flexclone, if you're using VMWare and NOT using Flexclone you're missing out.
Also, if you don't do snaps, you don't need the size of 2x + deltas, but then again why would anyone do that?
- You can't resize iSCSI LUNs. With NFS on a FlexVol, you can
resize to your heart's content because WAFL is controlling the filesystem--not the host.
Sure you can! The problem is that VMWare doesn't like it when you do. Windows and Linux couldn't care less, they see more space to extend their partitions. VMWare hasn't gotten there yet.....I don't think. I know it blew up ESX 2.5.
- It's open, meaning that your VMDKs aren't locked into the
proprietary VMFS file system. This could potentially simplify backups and restores.
Totally agree, good point. One other thing you need to consider is that you'd still likely keep each VMWare Guest VM in their own volume and export each one, to be granular with snapshots + replication. It's more work but well worth it.
Glenn @ Voyant
-----Original Message----- From: owner-toasters@mathworks.com [mailto:owner- toasters@mathworks.com] On Behalf Of Scott Lowe Sent: Wednesday, July 11, 2007 10:44 AM To: Davies,Matt; toasters@mathworks.com Subject: NFS vs. iSCSI for VMware (was "Re: List still active?")
The #1 complaint I have about using iSCSI relates to how LUNs are handled in the NetApp world:
- You have to (by default) allocate 2x + deltas for LUNs. For a
100GB LUN, you'd need 240GB of space (by default--I know there are workarounds with ONTAP 7.2 and later). With an NFS mount (which is nothing more than exporting a FlexVol), you only need to account for the 20% Snapshot reserve.
- You can't resize iSCSI LUNs. With NFS on a FlexVol, you can
resize to your heart's content because WAFL is controlling the filesystem--not the host.
- It's open, meaning that your VMDKs aren't locked into the
proprietary VMFS file system. This could potentially simplify backups and restores.
As Glenn @ Voyant already mentioned, you also gain thin provisioned disks by default and more knowledge/history/experience with NFS than with iSCSI.
Thanks, Scott Lowe ePlus Technology, Inc. slowe@eplus.com
On Jul 11, 2007, at 9:34 AM, Davies,Matt wrote:
Scott,
Any chance you could expand on the advantages of NFS over Iscsi ?
NFS
isn't an area I have any experience of....
Cheers
Matt
-----Original Message----- From: owner-toasters@mathworks.com [mailto:owner- toasters@mathworks.com] On Behalf Of Scott Lowe Sent: 11 July 2007 14:05 To: Forest Leonard; toasters@mathworks.com Subject: Re: List still active?
Forest,
When it comes time to configure VMkernel for VMotion, then I'd definitely recommend keeping it separate from the virtual machine network. As it stands right now, you don't even need a VMkernel NIC configured because it sounds like you are using Qlogic iSCSI HBAs
and
only have a single ESX Server. Since the Qlogic cards handle the iSCSI traffic and there is no VMotion, there no current need for a VMkernel NIC (unless you want to use NFS from the FAS to provide additional storage for VMs--which, by the way, works pretty well and has some nice advantages over iSCSI, IMHO).
Regards, Scott Lowe ePlus Technology, Inc. slowe@eplus.com
On Jul 11, 2007, at 8:33 AM, Forest Leonard wrote:
Interesting stuff. I only have one ESX server so I haven't gotten to the Vmotion configuration yet.. That should be later this year.
I just configured a RDM to run a Virtual server on. I found an article where you want to create the type as NTFS if it is a RDM for a windows host. Not sure if I am going to use this going forward. I actually don't know if you can migrate into a RDM. It doesn't look like it.
I am only using 2 NIC's on my Vmware server.. and a QLOGIC card for my ISCSI access.. I may need to look at bulking up my NIC configuration. It looks like I may gain some performance if I separate my VMKernal from my Virtual Machine network?
Thanks, Forest
-----Original Message----- From: Davies,Matt [mailto:MDAVIES@generalatlantic.com] Sent: Wednesday, July 11, 2007 8:18 AM To: Forest Leonard; ggwalker@mindspring.com; toasters@mathworks.com Subject: RE: List still active?
Sounds like we are both at the same stage.
We are not using RDM's yet, however when it comes to exchange we
will
have to, or I may just stick with using the Microsoft Iscsi initiatator from within the VM, not exactly supported, but I know of other
people
doing it.
We have migrated 8 machines so far, into just one datastore and have not seen any performance problems at all, although most of the machines have very low IO requirements.
We are using a script to snapshot and then replicate using snapmirror, it works very well, however our Virtual Centre server is also a VM which was causing a few problems with the snapshots on the VMware side
not
being removed, but moving this to a separate datastore seems have cured the problems, even SQL doesn't seem to have a problem.
The script is the one written by Evan Battle, that is in the newest netapp docs on VMware. I did have a few problems with ssh to the filer, but we are now using rsh and it seems to be ok.
I don't know how you have setup your virtual switches on the ESX side, but I got some best practice information out of VMware on that subject.
Each of our ESX IBM3550 host servers have 6 Nics, connected as follows.
2 Nics for service console VMKernel for Vmotion, load balance using Virtual port ID
2 Nics for Virtual Machine network, load balance using Virtual port ID
2 Nics for ISCSI (Service Console and VMKernel), load balance using IP Hash
Hope this helps....
-----Original Message----- From: Forest Leonard [mailto:fleonard@rvigroup.com] Sent: 11 July 2007 11:06 To: Davies,Matt; ggwalker@mindspring.com; toasters@mathworks.com Subject: RE: List still active?
Hey Matt... I am actually doing the exact same thing on a FAS 270..
I
have migrated about 7 servers so far.
Are you using RDM's (remote device mappings) for the Virtual Machines? I actually just created 2 200GB LUNs on the netapp to use as
Datastores
and have not had any performance issues.
Just wondering what your experience with RDM's are.. I added one into a virtual machine.. It just lets you map a LUN directly to a Virtual machine.
Has anyone out there used the netapp script to capture a VM
snapshot?
Thanks, Forest
From: owner-toasters@mathworks.com on behalf of Davies,Matt Sent: Wed 7/11/2007 1:00 AM To: ggwalker@mindspring.com; toasters@mathworks.com Subject: RE: List still active?
Still working by the looks of things.
busy in process of migrating all our physical severs to VM, stored on a iscsi lun on a FAS 270.
For those that want to know we are using the software iscsi
initiator
within ESX and have not had any problems so far.
cheers
matt
-----Original Message----- From: owner-toasters@mathworks.com owner-toasters@mathworks.com To: toasters@mathworks.com toasters@mathworks.com Sent: Wed Jul 11 01:03:49 2007 Subject: List still active?
I've noticed that I'm still subscribed, but have received no email since July 4th. Everyone didn't trade their NetApp gear for something
else
while I was out of town did they?? J
This e-mail (including all attachments) is confidential and may be privileged. It is for the exclusive use of the addressee only. If you are not
the
addressee, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error,
please
erase all copies of the message and its attachments and notify us immediately at help@generalatlantic.com mailto:help@generalatlantic.com. Thank You.
#####################################################################
#
## ############# Note: If you have received this message in error, please notify the original sender immediately by telephone (203)975-2100 or by return e-mail, and delete the message, along with any attachments from your computer. If you have received this message in error, you are hereby notified
that
any dissemination, distribution or copying of this e-mail, and any attachments thereto, is strictly prohibited.This note is also to notify the recipient of this email that it has been scanned for all known viruses and attacking techniques. Thank you.
#####################################################################
#
## #############
#####################################################################
#
############### Note: If you have received this message in error, please notify the original sender immediately by telephone (203)975-2100 or by return e-mail, and delete the message, along with any attachments from your computer. If you have received this message in error, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments thereto, is strictly prohibited.This note is also to notify the recipient of this email that it has been scanned for all known viruses and attacking techniques. Thank you. http://www.rvigroup.com
#####################################################################
#
###############