Best practice for fast file level restores from a VM file server

[Willie_Eckersli]

Hi,

I am running 2010 with the VM agent installed and I'm trying to find an answer on how to provide fast file level restores. My windows file server has RAWS installed and has a 2 TB hard drive. I currently backup my VM to tape and wish to continue this practice for disaster recovery. Unfortunateley when trying to restore a single file I have to first restore the complete VMDK to a staging area which takes 16 hours.

I'm thinking maybe a solution would be to backup to disk and then stream this copy of the backup to tape. Hopefully this can be achieved which ideally would give me in a disaster recovery scenario of being able to restore from tape with the added bonus of being able to restore individual files from the copy held on disk.

Is this the sort of thing anyone else is doing in which case what is the steps, or is there a better solution?


Thanks

Paul

[Ken_Putnam]

You can do this with a policy (B2D2T, that is)   By managing the OPP of the media set that writes to disk, and only doing Overwrite jobs, you can manage how much space on the B2D volume is used (how many copies of disk data that you keep)

[Willie_Eckersli]

Is there a built in policy that can be edited or do I have to create one from scratch, in which case is there any kb articles that describe the steps to set this up?

Thanks in advance

[Ken_Putnam · Accepted solution]

See the 2010 Admin Guide, page 532 and following