Dynamically expanded VHD and fail over cluster.

I’m looking for solution how to use dynamically expanded VHD disks on Hyper-V fail over cluster. I’m going to set up system on DELL MD3000i iSCSI SAN. Problem is, as far as I know it is impossible to put a few virtual machines on one LUN in fail over cluster, so basically there is no point to use dynamically expanded HDD on fixed size LUN. Is any way to set up dynamically expanded LUN on MD3000i? Is any other option to achieve safe space goal in fail over cluster?

Thanks for your help

Find more posts tagged with

Free for TechExams community: Cybersecurity salary guide

Compare cert salaries and plan your next career move

Button

Comments

astorrs

You're basically out of luck. the MD3000i is very basic and can't extend a virtual disk (what it calls a LUN) and doesn't support MetaLUN type functionality (like a CX3 series would). Given the limitations of failover in Hyper-V (this wouldn't be an issue in ESX), I don't see any option other than to allocate your space as you think you'll need it all at the beginning. I don't think the MD3000i even supports LUN migration (check with Dell) - if it did you could always move the files to a larger LUN if need be down the road.

Sorry to ruin your day.

PiotrIr

Many thanks for your reply.

You ruin my day but I've got answer so I'm happy enought.

PiotrIr

And what about DELL AX4-5i, will it work?

astorrs

PiotrIr wrote:

And what about DELL AX4-5i, will it work?

You could use an AX4-5i with dual controllers and using 300GB 15K SAS drives, even with only Navisphere Express you'll still get the ability to do MetaLUNs and LUN-to-LUN migrations. Have a look at my response to your other post here for more info on why that's a good thing: http://www.techexams.net/forums/viewtopic.php?p=263165#263165

You won't get thin provisioning with the AX4 and it has way less cache than the CX4 series, but it may meet your needs. Call your Dell account manager and get him to do a conference call with his pre-sales storage engineer and work with them to confirm what you will need. Have you done any baselining of the 80-100 VMs that you want to migrate/deploy? You should be able to provide IOPS required, read/write percentage, and the pattern(s) of I/O (random, sequential, etc) - those are things they are going to want to know.

The storage is the backbone of your virtual infrastructure, you want to take the time to plan and implement this right or you'll be building an inverted pyramid...

PiotrIr

Astorrs many thanks for your advices.
It looks you have huge experience with storages and it also looks I have no experience with SANs. Basically you are right with this pyramid so I this is reason I’m posting so many questions on this forum.
My problems are:
1. I have no idea how to get IOPS required, read/write percentage and the pattern(s) of I/O (random, sequential, etc). So I would really appreciate if you will give me some kind of idea how to get this information. I already build test environment for 3 machines on one server…
2. My next problem is I have no DELL account manager so I need compare this information with documentation. Is any way to do this?
Once again many thanks for your priceless help.