I’m no fan of Virtual Desktop Infrastructure, but a friend says it has useful qualities
Anyone who works within a large organization will be well aware of what Virtual Desktop Infrastructure (VDI) is. For those unaccustomed, essentially VDI is a technology whereby on any device, anywhere, a user can log in and access a virtual representation of a desktop computer. It’s a way to use the applications, user setup and security of a fixed physical device without needing to use a fixed physical device.
That’s the good part. But alas VDI has a negative side: poor user experience, often laggy processing and a “one size fits all” approach that doesn’t really cover the multitude of form factors that end users actually need.
So, when I heard awhile ago that a new player in the VDI space picked up some funding, I wrote a post that articulated my incredulity that an investor would actually see the space as attractive. A friend of mine, a particularly studious chap who, given a science background, has a preference to empirical data and research-based findings, called me out and sent me an email critiquing my commentary about VDI.
Benefits of VDI
His opinions were formed from some testing he has been doing of Amazon Web Services’ Workspaces product. Within the context of a startup he’s been working with, he’s been trialing Workspaces. And based on that experience, he offered up some thoughts in a kind of devil’s advocate position, as to why VDI is actually useful:
- VDI makes security and Bring Your Own Device (BYOD) an easier proposition, especially with temporary employees. They can use pretty much any machine, and, in the case of AWS, they just have to download the Workspaces app and then they can connect to the remote desktop. By doing so, they can be fully functional, without being able to download any sensitive data.
- VDI moves (the majority of) IT costs from capex to opex.
- Admittedly there is no functionality without an internet connection. But if you have enough bandwidth for the streaming, it makes the users’ internet connection irrelevant because the actual machine is in a data center with a massive pipe.
- In the case of Workspaces, if the rest of your infrastructure is on AWS, there is the potential to integrate nicely without having to use virtual private networks (VPNs).
- “Zero Clients” have been cheap to obtain, with the promise of minimal maintenance, to enable true “hot-desking” without needing a typical PC and the associated hassles. Alongside that there is the ability to upgrade specifications that the user needs remotely. When a VDI vendor releases new operating systems, faster machines, etc., the company won’t have to change any end user terminals.
All the points that my friend raised are indeed valid. VDI does solve some of the problems around access, a desire for opex-based spend and off-loading responsibility to a third party. But I still can’t agree that VDI is a good way forward for many organizations. Even putting aside the fact that, at least in my experience, VDI has been sub-optimal in terms of speed and performance.
My distaste for VDI has a more fundamental basis: a desire to see organizations truly arm themselves for the future rather than take a small step that, if anything, further avoids the real existential issues they face. As I said to my friend in response, he raised some good points—VDI does offer some benefits in terms of remote access, reduced IT overhead and simplicity.
But my concern is that by going down the VDI route, many organizations avoid having to think about what mobile/social/local/big data/insert your buzzword means for their business. To use a trite example, by deploying VDI, would the cab companies have gotten any closer to sensing the threat that Uber raises and responding accordingly?
I can concede that VDI does introduce some benefits and efficiencies to an organization, but in doing so, it perhaps only avoids the inevitable and allows those same organizations to divert their glaze from a huge elephant that’s sitting in the room.
