The case for Virtual Tape Libraries in the SMB space

Due to new VTL offerings which include integration of De-Dupe and Single Instance File copy technologies, VTLs have been getting a lot of press coverage lately, particularly in various trade magazines, and industry websites. Recently new code allowing for only single instances of files to be saved to virtual tape, or disk have made VTL investments more cost effective.

VTLs with compression and data de-duplication improve capacity utilization and lower expenditures on disk, they also can hold more data before moving it to tape.

However there are a few factors to be considered when looking at VTL technologies. To most software backup applications, the VTL appears to be just another tape drive. So the VTL performance is still limited by the speed of the backup software and the server that is running the backup software. So implementing a VTL as just a tape library replacement solution is not ideal. The VTL must support a network backup solution scenario meaning agents need to be deployed which allow for disk to disk backup over the network.

That begins to increase the cost to the SMB client. Where each server you would like backed up requires an addtional license fee, so now not only are you paying for the disk, you are paying for the agents, and you are also paying for the tape that will eventually archive your data.

So your initial VTL implementation costs looked good at 30k, however to get the real performance gains have jumped to 45k.

OK so all SMBs have their pain points, backup windows definitely being one of them, however there are instances where VTL definitely make sense however there are other cases where tape investments still make sense.