This is the first post of our new VDI series by our senior technologist and resident virtualization expert, Chuck Mills. Take it away, Chuck!
The well-managed desktop has been the goal of IT for many years with the promise it would make their job easier and reduce total cost of ownership (TCO). The tools used to accomplish this feat have improved over the years, yet this is still an ongoing struggle for many companies. This better management solution is one of the benefits (along with many others) used for justifying Virtual Desktop Infrastructure (VDI).
During the early years of VDI, the thought was to replace as many desktops as possible using that promise of easier desktop management. After all, IT was very successful at doing this with servers why not conquer desktops next. But, as we all know, desktops have very different resource requirements along with many other differences that slowed the adoption of VDI. We are all aware of those differences and I’m not going through them here. Ok then, let’s focus on using VDI for the very specific use cases and that approach has provided good results to specific user populations. We have successful VDI solutions in areas like healthcare, call centers, shift-workers, contractors and many others. Those benefits include access to the desktop from anywhere, location based VDI (East – West coast VDI), faster deployment of new desktops, and security. These were all solid reasons to consider VDI and this is still true for today. Ongoing development of VDI solutions will continue to expand the use cases for virtual desktop solutions.
Wait, I’m getting ahead of myself. In reality when you look at a physical desktop there are several distinct sections that make it work and work pretty well on a single physical system. The individual installation for each user is what brings heartache to desktop management. Because each Operation System and their installed applications can be updated independently. This quickly produces non-uniform environment. This is where desktop management tools are needed and, as I have mentioned, have produced limited success.
Many early VDI projects used this individual model to create virtual desktops, which provided some benefits (security is the one I think of), but it did very little to fix the desktop management problem. In fact traditional desktop management tools don’t always cut it when it comes to supporting virtual desktops. This one virtual desktop for each user also creates tremendous resource requirements for the servers that held all the desktops. Now as we start discover all these issues we realize this can become very complex.
Ok, we could go on about the struggle around desktop management and complexity of VDI, instead, let work on ideas that can form an approach to solve these and many other issues. The approach needs to:
This series will provide solution options around approaching the desktop as a virtual solution and not a physical desktop “converted” to a virtual one. I will discuss ideas around the following areas.
I believe, for now, we can accept that VDI is not for everyone, but we should provide solid VDI solutions where they do fit. But, I also know there are some exciting developments going on that will expand VDI use cases. I know that VDI is complex and the approach I’m outlining, is to separate all the pieces of the desktop and re-construct them in a way that not only optimizes each section, but also takes advantage of current desktop virtualization technologies. The ideal solution should help IT keep things updated and secure without a negative impact on what the user needs, which are the things to be productive.