I just had a Hyper-V situation dumped into my lap and I'm new to the Hyper-V environment so I need a little advice.
2 Physical win2008 R2 servers running Hyper-V
24gb ram on each server
Server1 has 3 VM's running each allocated 4Gb ram
Server2 has 8 VM's (5 running) each allocated 4Gb Ram. (Note: 3 of the VMs were turned off because they were locking up the system probably due to not enough ram available however I can't be sure as this was done before the project was handed to me)
All VM's and their data are installed on the Host systems, no SAN involved.
Backups of the VMs are currently non existent. There is a Drobo-fs device on the network which has space allocated for backups.
1) There isn't enough physical memory on the host systems to handle 11VM's with 4Gb ram each. Each host server will need at least 48Gb of ram to run all 11 VMs simultaneously.
2) I'll need a backup solution such as Veeam to facilitate backing up the VM's and their data while they are running.
1) Have all 11vm's available 24/7/365 to run simultaneously
2) Have daily backups of all VMs and the data on their VHD available for disaster recovery
1) Upgrade the host hardware with at least 48Gb of Ram each
2) Divide the VM's between the two hosts
3) Install a backup solution such as Veeam to take daily snapshots/backups of the VMs and store them on the Drobo-FS
4) should one of the hosts go down, the VM's that were running on it should be able to be recovered to the other host from their last backup and continue running there.
1) What if anything is wrong with the above?
2) Is this just a completely improper use of Hyper-V for this situation or should this work OK?
3) Any other suggested solutions?