Posted by : Amanda Stein Friday, April 12, 2013
Virtualization has been sweeping through corporate IT despite the global slump we’re in. And with businesses expecting to finally start ramping up in 2013, firms that have been holding back on those investments are likely to push through. It’s likely to be a desktop virtualization avalanche.
A lot of people are welcoming what promises to be a new age of storage efficiency, reduced cost and generally more control over their assets. But the trend toward virtual desktop infrastructure is going to require some adjustments in the way that teams think about data security and backup storage.
Comparing Desktops and VDI on Data Protection & Backup
Desktop virtualization is different than traditional desktops when it comes to protecting information. The traditional wisdom kept that you just didn’t back up desktops; if you want to protect user data, you store it on the network. The attitude toward laptop storage was even worse; if it was stolen, lost or damaged, users were simply out of luck.
In a virtualized environment, IT departments are forced to solidify their policies in regard to the machines being virtualized, and that means standardizing data protection and backup policies. But often, the default response is to simply back up everything, which is often overkill in the other direction.
Let’s think about that for a minute: If you have 500 users, each with a 40 GB desktop image, that’s 20 TB of stored data. What’s in all those terabytes? A few hundred copies of Office, hundreds of Adobe products and everything in users’ recycle bins, just to name a few.
Full backup in VDI is attractive due to its simplicity, though. If you lose a server, all you have to do is restore the full machine. But if you’re only backing up selective data, restoration requires that you track their full configuration as well, restoring the apps, then the data and then they’re up and running.
Agent-based vs. Storage-based Backup
Implementing VDI also requires some planning with regard to exactly how you’ll accomplish backups. An agent-based solution will simply run as a client inside each virtual machine, backing up data periodically. The alternative is a storage-based backup, which runs entirely separate from the VM with no client.
One clear advantage of agent-based backup over the alternative is the ability to do continuous backups of selected data on each machine. The backup client can run in the background, trickling changes to backed-up directories into storage instead of doing a large, one-time dump of data.
The benefit of having a client is that you can be very selective about the folders that are backed up, leaving out the operating system files. The alternative is to simply grab a full disk image. Again, we’ve run up against this dichotomy of sacrificing storage in the name of simplicity.
Clearly, there are some conceptual and operational differences in protecting VDI than traditional desktop boxes or even virtual servers. Those differences are going to require some planning and scrutiny if any company is planning on implementing a virtualized infrastructure.
What’s your VDI storage strategy? Let us know in the comments.
About the Author: Jared Jacobs has professional and personal interests in everything technology. As an employee of Dell, he has to stay up to date on the latest trends and breakthroughs in large enterprise solutions and consumer electronics buying trends. In his spare time he is tinkering with sound systems and other awesome gadgets he can get his hands on. He’s also a big Rockets and Texans fan.