Virtual desktop infrastructure — otherwise known as VDI — is, without a doubt, a hot topic.
When we first started down the VDI path in January 2011, I agreed to document and share our experiences — for better or for worse — to help showcase the benefits of desktop virtualization. And as a result, I’ve talked and written *a lot* on VDI over the last year. I was a guest blogger on VMware’s End User Computing site (see posts here: VDI Right on the Money; Out on a Limb With Virtual Desktops; and 10 VDI Lessons from the Real World) and have contributed to the eCIO Forum on the subject; I’ve hosted a desktop virtualization workshop and presented in several webinars and conferences (my BrightTalk session was ranked one of their “Top 6 IT Infrastructure Webinars of 2011″ — how cool is that?!?); and fielded emails and calls on an almost weekly basis from colleagues looking for recommendations or advice as they begin their own VDI projects.
The topics I’ve covered, and questions I frequently receive, are generally along the lines of:
- Is it really as easy as vendors make it out to be? (Not always, but what ever is?)
- Will we save money doing VDI? (Depends)
- How can we best prepare? (Plan, test, revise, plan, test, and plan some more)
- Should we even be considering this, especially if the savings are unclear? (Absolutely!)
Most institutions, including my own, are looking towards VDI to improve their desktop management capabilities, save PC refresh dollars, expand access to institutional software, and reduce their carbon footprint — and roughly in that order. These are all great goals, and certainly reason enough to implement virtual desktops. But they are also reflective of incremental (however large) improvements, rather than true innovation.
What do I mean? Well, we’re using virtual desktops to replace physical desktops, but in many (most? all?) implementations, the model we’re implementing against is still the same. Students still do work in computer labs — physical ones with virtual machines or virtual ones, but labs nonetheless. Faculty and staff still have one office with one client connecting to one “desktop”. VDI makes it a lot better, but….
With VDI, so many other things are possible.
Employees, for example, don’t need to be tethered to one device or location, or even one desktop. We could work from our own device(s) or provided ones, or both. We could use one desktop or several, depending on the task we need to accomplish and the software associated with it. We could structure flexible work environments that support the formation and re-formation of cross-functional teams as projects change.
For students, what if we looked at it from a class or individual level, instead of by lab? Perhaps we could provision desktops by student, so that an entering freshman receives her own virtual desktop that automatically updates with software she needs, based on course enrollment. Or we could enable faculty to teach — in real time — with students working remotely “in the field” on laptops and tablets and other mobile devices, but still view and share student work across the class.
We are over a year in to our VDI implementation, and are just now starting to scratch the surface of VDI’s true potential for innovation. Imagine if we started over, designed our ideal environment first, and *then* applied VDI to it…what could we do? What would *you* do?