In 2009, we conducted a long rolling review of Virtual Desktop Infrastructure solutions spanning almost the entire year. I'm a tough reviewer to please, so while I was excited to see how easily the benefits of server virtualization could be transferred to the desktop, I was a little disheartened by how much further VDI has to go in order to garner more mainstream support. When you're talking about the benefits of server virtualization on a large scale, the cost savings is easy articulate, but when you're doing desktop virtualization, you're messing with the comfort zone of your users. As a result, cost savings isn't always the overriding issue. If desktop virtualization saves money but reduces productivity, what's the point?
That's the question that engineers, designers, and game developers (among others) are asking when they're considering running graphically intensive CAD or graphics apps on a virtual desktop. The fact is that first generation virtual desktop solutions were designed to serve out a basic virtual desktop running office and client/server or web apps, and they did a great job at that task. However, to change this en masse requires that the complete desktop computing environment needs to remain consistent. That means transparent use of I/O between the client hardware and the virtual desktop, and CAD, streaming media apps and even computer games should run as well or better in a virtual desktop environment as they do in a thick client environment.
According to Harry Labana, one of Citrix's divisional CTO's, 2010 will be a breakout year for desktop virtualization. Of course, most of the desktop virtualization vendors were predicting 2009 as a breakout year, and that reality never came to pass. While I'm still a major skeptic on desktop virtualization actually picking up major steam in 2010, if Citrix and VMware can deliver on some of the streaming, multimedia and I/O enhancements they're promising, you can be sure that VDI will take off along the same trajectory as server virtualization.