It does not mean don’t try
In part 1 I talked about how application delivery has evolved and expressed my general disappointment with the relative traction of application virtualization and streaming.
To be clear I am not saying don’t try, as I believe a more holistic approach to application delivery offers the maximum flexibility to change, minimize capital and operating costs. This enables IT to have more service level control while delivering the experience users want. Virtualized applications enable remote access and provide superior TCO, performance & security for both legacy and new applications in the Windows world. While optimization enables high performance, security and superior TCO for both web-based and browser style apps.
However in many cases many users just want all of their applications delivered with a desktop experience. Users for years have been trained to use the Desktop Shell as an application session manager. For them it still makes a lot of sense to provide a desktop experience-in many cases they will demand it. I recall when I joined Citrix and was given access to the Web Interface with all of our applications. I didn’t want to use it, because coming from Wall Street I was used to a desktop centric experience. In my travelers meeting customers I’ve found the preference to vary widely, but I will say for those customers that are new to Citrix, they tend to prefer the desktop centric experience. I’ll bet however as the new app models retrain the masses, we’ll see a swing back to app centric as evidenced by the advent of the AppStore concept. Pragmatically speaking, when thinking strategically it’s helpful to have both options at ones disposal.
Irrespective of an app or desktop centric experience, I don’t believe the desire or need to run local applications offline is going away. In the below diagram it appears that desktop virtualization is only geared towards the more task orientated users, and that power users are better served with local apps. As a reminder to those who did not read part one, these diagrams are an older Citrix document that I think is useful in present day strategic discussions. What we know today is that desktop virtualization is applicable to many user types including the ones that create knowledge. However this is where a lot of the debate in the industry starts on costs, VDI desktops vs. Shared hosted virtual desktops. The reality is the stateless desktop must continue to be enabled for broader use case penetration. For some however the diagram as is will still make sense, and that is ok. Remember the goal is to build a strategy and map to use cases.
If you decide to head down the local app path, then you will still want to find ways to deliver and simplify management for a distributed set of users. So why not stream as many of those apps as possible. Granted that may not still capture 100% of ad-hoc apps, but design for the mass use case and manage the exception has always been my philosophy.
Let’s go back to the thought in Part 1, discussing the problem of standard images and the associated lifecycle management.
Now let’s consider the impact of a holistic delivery model superimposed on top of this model. In effect the number of unique disk images is minimized helping simplify the problem. Not only that, but one can start to deliver highly efficient desktop computing environments and still empower users all within the same strategic framework.
Imagine if we had to support the image management model across multiple device types. Especially in today’s world of many devices per user. The question becomes can you scale your current approach? In fact I doubt the current approach of distributed computing can even address the problem.
A lot of customers wrestle with the decision of refreshing their Pcs when considering desktop virtualization vs. buying a brand new thin client or cheaper PC. One way to think about this strategically could be to start to buy powerful PCs on an annual basis and then repurpose them every year to a different tier of user as requirements evolve. This is illustrated in the next few slides. This may be great in theory but hard to manage in practice in a dynamic organization. However the salient point here is that having a strategy helps you plan for these types of scenarios so you can make an informed decision.
I’ll pause here and leave the conclusion for the final part 3. However I hope for at least some this helps as a starting framework for your own strategic desktop conversation that ignores the hype. As I said in part one, my experience has shown that the most successful customers are very thoughtful.