The Four Biggest Pitfalls to Avoid in VDI

At Strategic Integrators, we’re working with many clients around the country who are either in the midst of, or are planning for, a VDI deployment. 

We were recently in Seattle talking with a client — a large Fortune 1000 company that has facilities and operations around the world. For this firm there are many obvious benefits to VDI not the least of which are ease of desktop management and security. 

During our conversation, however, they asked us: “what’s the biggest mistake that you see other people making in VDI?” In response, I thought it would be helpful to briefly cover what we see as the Four Biggest Pitfalls to Avoid in VDI.

Pitfall #1: Poor Planning

Poor Planning is the biggest mistake we see people making by a wide margin. It’s easy, I think, to believe that it will be as simple to virtualize a desktop infrastructure as it was to virtualize a server infrastructure. The reality is, though, that there are a lot more moving parts to a desktop environment than there are to a server environment, and that means that you just have to plan, and plan considering every possible detail and dependency. 

For example, if you were building a house, or doing a remodel as my wife and I did a few years ago, your success would hinge on a thorough plan in combination with a contractor who was good at what he did. For our remodel, we had an awesome architect who put together a terrific plan. That plan provided the guidance our contractor needed to produce a great result. (And how many remodels have you heard of that go really well?) If having a plan for a construction project is important, it’s that much more critical for a VDI project because it’s so much more complex, so much more technical and thus so much more can go wrong if your plan isn’t spot on. 

Your VDI project plan should start with a comprehensive assessment of your overall IT environment …of applications, of workloads, of software, of existing hardware as well as available resources. It’s not until you account for all of this (and much more) that you can put together a VDI project plan that has a chance of working well from the start. 

The great news is that we believe developing such a plan is very doable and, again, nothing will help ensure the success of your VDI project more than thorough planning. For sure, the planning you do now for your VDI project will pay incredible dividends in the months and years to come. 

Pitfall #2: Planning Without the End (User) in Mind

This second pitfall to avoid — Planning Without the End (User) in Mind is an easy one to neglect. It’s crucial that one of the primary objectives of a VDI deployment be a rich end user experience. It’s so easy to get caught up in a VDI plan with all the details and complexities of a desktop infrastructure discussed above. But if the end user community doesn’t have at least as good an experience — if not an improved one — at their desktop than there’s going to be serious push back which we’ve seen doom many VDI projects. 

The issue is that every end user has specific requirements, needs, and workloads that they need to get their job done. So even the most technically well-executed VDI project can fail if it negatively impacts those end users’ ability to do their job — or even gives the perception that it negatively impacts the end user. 

Encouragingly, the solution is a fairly simple one … be sure to involve those end users (or representative end users) in your VDI planning process from the very beginning. As you do, not only will you get a thorough understanding of those specific needs, requirements, and workloads that you will have to maintain in the new VDI environment, but I think, at least as important, you’ll get their buy-in. One of my favorite axioms is, “people will support that which they help to create.” So this will pay an important return, not only in the planning process but in the rollout and deployment of your VDI project because it won’t be your project at that point, it will be their project as well, and your chances of success will increase dramatically. 

The good news is we believe that this is absolutely an attainable goal and that your end users can be thrilled with their improved experience as well as increased desktop mobility. So start with the end (user) in mind and who knows … they may end up building a tasteful shrine in your honor somewhere in the office!

Pitfall #3: Failure to Put Together the Right VDI Project Team

We discussed above the importance of getting end-user input in the VDI project planning process. Pitfall #3 — Failure to Put Together the Right VDI Project Team — is a corollary of that point. It’s crucial to put together a comprehensive and representative VDI project team. 

I think it’s easy to see why this issue trips up so many people. You have a virtual desktop project so the natural thought is “let’s get our virtualization team to do it.” But as I wrote about previously, a VDI environment is so much more complex than a virtualized server one. So in addition to your virtualization team (or representatives from that team) — which you’ll absolutely require — you’ll also need folks who have expertise and experience in compute, in storage, in networking and in desktops themselves. Finally, as we discussed above, you’ll need representative end users as well. 

If you put together a representative and comprehensive VDI project team, they will produce a much more comprehensive VDI plan. It will be a plan that takes into account all of the different requirements across the continuum of that environment and provide the specialized expertise and knowledge needed to address those requirements. 

And, as before, the chances of project success will go way up. Jack Welch, who was the amazingly successful CEO of GE for 20 years, used a phrase that I love: “I want every brain in the game.” I think that’s a wonderful thing to keep in mind as you put together a VDI team … that you want every brain at the table, or at least a representative brain from virtualization and compute and storage and networking and desktops and the end user community. If you do you’ll not only increase your chances of a solid, comprehensive plan to start your VDI project, but you’ll also (as noted above) get buy-in from across your IT organization. So if you have representatives from the different facets of your organization helping develop your VDI project plan, they will become your advocates throughout the enterprise when the time comes for VDI roll-out. This will assuredly help increase the level of buy-in and ultimately successful adoption of your new infrastructure.

Pitfall #4: Failure to Provide Sufficient Performance

The fourth pitfall to avoid in your VDI project is, a Failure to Provide Sufficient Performance for your VDI environment. There are two area which are particularly important here. The first is network bandwidth. For example, network bandwidth is an especially important consideration for your WAN links. If your WAN links can’t provide sufficient bandwidth for your VDI environment — or the latency is too high — you’ll run into significant performance issues. 

Even more important than network throughput capability, though, are storage performance requirements. I think by this point everyone knows it’s key to plan for log-in or boot storms and yet it’s not as simple a matter as just throwing SSD or Flash storage into a VDI project and thinking that you’ll have a great result. This is definitely not a case where one size fits all. Instead you have to find a storage solution that’s intelligent and is architected and optimized for VDI. There are a lot on the market today that really aren’t and will either underperform or you’ll pay far too much for the performance you ultimately require. But if you can find and deploy a storage platform that’s optimized for VDI it will pay real dividends in the months and years to come. 

As always, I hope this is of some help. Please let me know if you have any questions or comments below. All the best!