If you want success with desktop virtualization, you need to reconsider VDI use cases. You're better off deploying VDI for a specific purpose -- only a few necessary users, perhaps -- than using it across the board.
Part of my job entails giving presentations to people that, more often than not, are just
We're led to believe that VDI should be in widespread use throughout organizations, and because of that we're led to design complex and expensive infrastructures -- and deploy equally complex and expensive VDI platforms. But instead of thinking of VDI as a sweeping end-all solution, you should treat it just as you would any other form factor in your environment. Often, physical desktops are just fine, and your top priority simply needs to be streamlining desktop management.
In essence: Stop putting VDI on a pedestal!
VDI success hinges on the use
VDI is just another form factor on which we deploy Windows. It's kind of like cheap beer -- pretty much the same thing, just in a different container. With beer, which container we use depends on the situation (cans for the beach, bottles for home, keg for the pub), and it's the same with desktops.
The de facto standard in a company is the desktop computer, but there may be a strategic reason to deploy a laptop in certain situations. Of course, along with the laptop come some concepts that are foreign in desktop environments, such as storing files locally rather than on the network, disk-level encryption and VPN connections. Still, the goal is to take the managed corporate Windows environment and make it portable. Laptops are a strategic choice to ensure continuity for all users -- and we are still able to manage those PCs efficiently.
More on VDI use cases
Knowing when VDI is right for you
Why VDI is a niche technology
VDI vs. RDS: Which should you use?
The same line of thought holds true for VDI use cases. Much like how laptops exist to accommodate unique requirements, virtual desktops satisfy a separate set of goals. I've written about goals for desktop virtualization before, but the general idea to reach VDI success is that you should only deploy it in specific situations that benefit from desktop virtualization (just like you should only use laptops in strategic situations).
A VDI project, of course, comes with its own set of differences. Storage, networking and virtualization hosts are all hot topics. Plus, which vendor tools are the best is always up for debate. At the end of the day, though, if a certain aspect of VDI addresses your need, there's no reason not to use it.
The bottom line is that while a VDI project is more expensive than the status quo, you're doing it to add features and value to your environment. If you try to do that at large scale by converting everyone to a data-center-based virtual desktop, the challenge is much more formidable than if you use it as a point solution.
Putting VDI on a pedestal can make for a paralyzing project. Knock it down a peg or two, and use it as a strategic desktop delivery mechanism based on nothing more than a different form factor.
This was first published in December 2012