VDI (virtual desktop infrastructure) has long been predicted to be a growth area. This year, the technology has started to edge towards adoption in newer segments, segments in which it has not had significant traction in the past. But will this continue to be the case? Is VDI the computing environment of the future, or just a stepping-stone to another environment altogether?
It’s worth noting that when technologists refer to VDI, they primarily mean Windows-based VDI. The major VDI vendors—Citrix, VMware, and Microsoft—don’t yet support Linux VDI guests. This is not to say it can’t be done. Leostream and Virtual Bridges (amongst others) can deliver Linux-based VDI, but they remain in the minority for now.
What makes organizations lean towards VDI in the first place? In the main, it is for application compatibility. Windows has a huge ecosystem of applications. Some of these applications won’t run on other platforms, like Android and OSX, and some of them won’t even run on other flavours of Windows, such as Remote Desktop Services server instances.
Windows VDI also offers the centralized management and insulation from local hardware failure that virtualized solutions bring. In addition to this, there are other drivers. Users can use whatever device they choose (including their own). If your Mac users want to join in, for instance, you can simply deliver the Windows desktop to them via technologies such as Citrix Receiver or the VMware View Client. VDI also offers “instant” remote access in the event of a disaster. VDI desktops can work from any device with very little required pre-planning. But the main value-add concerns the user’s applications. There are probably very few business or consumer applications in the world that don’t provide a version that runs on a standard Windows desktop client OS.
So, the main reason many companies deploy Windows VDI is to have the ability to run software in their virtual environments that is tied, for reasons of compatibility or support, to the Windows client OS. If that restriction were taken away, would their need to be a VDI customer also be taken away?
Is the industry, then, heading to a place where any OS becomes simply a hypervisor for the business applications? Will VDI adoption decline as Windows desktops become less critical? Will the next migration that businesses undertake be the last “big one,” as the desktop becomes increasingly compartmentalized?
On one hand, growth in SaaS appears likely to contribute significantly to a decline in VDI adoption over the next five years. Software such as Dropbox is a case in point. Available on a wide variety of platforms, either as a native application or through a browser, it offers market penetration that is almost entirely independent of the OS in use. If all user applications were available in the way Dropbox is, VDI investment would be unnecessary—users would simply connect up to their applications from whatever device they happened to be using at the time.
However, with this in mind, it is not prudent to underestimate how heavily entrenched Windows is on the corporate desktop. Enterprises have decades of investment in legacy applications, and in all likelihood these legacy apps will be around for a considerable time. The only way to break this pattern and capture corporate users is to do it in greenfield deployments, and how many of those are happening at a given time?
But still, one of the bigger barriers to wider Windows VDI implementation—and certainly one that stops it from competing on a level playing field with RDS—is that of the licensing. Microsoft’s arcane VDI licensing terms continue to be a major deal-breaker on a variety of levels. With the breadth of Windows applications out there, licensing cannot help but be a very big deal for VDI. In addition, using Windows Server (like Amazon is doing) to provide a Windows 7 “experience” does not help those businesses with applications that don’t work on server operating systems and for which shims have never been written.
The only thing that probably isn’t subject to change is the need to deliver secure, managed environments and applications. This will be a constant. Right now, VDI and RDS are arguably the only ways to do it across all platforms.
So where indeed will VDI be in five years’ time? Will it be at the heart of business attempts to reach the “promised land” of seamless application delivery to myriad devices?
I believe that it will be significantly changed from where it is now. The base needs of users will not change—people still need to work with applications through which they retrieve, update, and store data—but persona management, application virtualization, disk management, and other surrounding tech also will evolve. SaaS will push forward, and cloud-based solutions—whether private or public—may start to showcase what they are capable of delivering.
With these things in mind, I believe that VDI will become just one of a number of workspaces that users can utilize in their daily jobs. Depending on the user’s requirements and application compatibility, it could become the primary workspace. If the VDI solution is delivered competently and with the right supporting technologies, many users may well find it the most productive workspace, but some will certainly choose other options.
And with Amazon WorkSpaces on the horizon and the ink drying on VMware’s acquisition of Desktone, much of the next five years’ worth of VDI may well turn out to be in the form of DaaS. Although for DaaS to achieve widespread adoption—which probably means “true” VDI with Windows 7 or 8.x—something will surely have to give on the Microsoft licensing front.
Thanks to Simon Townsend, Chief Technologist at AppSense, and Kevin Gallagher, CIO at Channel 4 Television, for sharing their VDI views, which contributed to this article.