Best practice, or least administration, to setup SAN for hosting VMs (non bootfroSAN). How many Luns, etc.?
I'm firing up a PROD farm using an LSI array with Brocade FC fabric. VMware, both their engineers and documentation, evade from answering what is a typical setup for supporting 4 hosts (vMotion) and approximately 100 VMs. I don't want to hold anyone's feet to the fire, I just need to know what is typical and the easiest to support. My last position, I supported two SAN environments with three hosts each (no vMotion) and they dedicated a single LUN at 220% (iSCSI Netapps) per VM for snapshots. If VMware has a 128-LUN limitiation, this could pose an issue in fairly short order.
Is 4 LUNs OK? 8? How would I (you) label them and categorize them?
Thanks much,
Carl
Who is Participating?
Experts Exchange Solution brought to you by
"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.
Experts Exchange Solution brought to you by
Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.
Start your 7-day free trialVMWare previously recommended an optimal LUN size of 256GB. I don't know if this recommendation has increased but I have luns upto 1TB. This is not an issue at all if your SAN is capable.
I would suggest that standard LUN size of 512GB is very acceptable and flexible. If you using Netapp you do need to keep your snapshot reserve in mind when creating your volumes but what percentage you leave is up to your unique requirements.
Also we ensure each LUN never has less than 50GB or about 10% of available space or you may run the LUN of of space during VCB's or snapshots.
ESX 3.5 update 2 now supports live cloning so long term snapshots are less of a concern.
You mentioned your running over a FC environment, therefore it must be 2Gb or better. Due to the large cluster side of the VMFS your IO is surprisingly low for the workload.
My final word on this would be, Look at your guests, if they have disks which are close to 250GB then use a 512GB LUN if not use a 256GB LUN size.
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
From novice to tech pro — start learning today.
Experts Exchange Solution brought to you by
1. I try to keep the LUNS 300-500GB in size to keep them managable.
2. Add 500MB to LUN creation overhead, add 5% to each ntfs for vmdk overhead, add memory of each virtual machine for swap file and about 10GB for snapshots
3. I try to balance low I/O and medium I/O machines a LUN and try to keep the same OS on each LUN so that we can have single insance memory page sharing which can cutdown paging by 20%
4. Large virtual machines I keep on their own LUNs, usually over 250GB since the VMFS lun needs to be a different block size
5. You can setup LUNs per VM function, subnet, vlan, etc.. this personal preference
6. Setup a LUN for ISOs and templates, it will make is easy to keep track of them
Just a few notes from recent projects, there are many ways of doing this.