Remoting desktops and browsers is getting easier. We can argue the specifics, but we can all agree that with Windows Virtual Desktop on its way, and after years of advances in VDI, DaaS, the cloud, and infrastructure, all types of desktop virtualization are more attractive than they used to be.
At the same time, we’ve had a revolution in devices and endpoint management: We have Windows 10 modern management and automated device provisioning; Chromebook and Macbook management has gotten easier; and the march towards SaaS/web/cloud apps that have clients for multiple platforms continues.
What this means is that if you’re embarking on a VDI, DaaS, or other remoting project, you’re probably excited about all the advances in that space. However, it would be a mistake to not consider the latest advances in endpoint technology, as well.
Years ago, Gabe gave a BriForum session called “How to get the benefits of VDI without doing VDI.” I remember that it was a very popular session, so think of this as the updated version.
Where to consider UEM and modern devices
Let’s look at some typical VDI/DaaS/remoting use cases. (To save time, I’m just going to call it all VDI for now.)
Use case #1: Say you want to use VDI to make it easier to manage Windows. Sure, you know that Windows is hard to manage no matter what, but having it in a datacenter will make it easier. Today, you’ll also have to consider the latest in laptop management. We have modern management, MDM, third-party products that build on top of MDM, AutoPilot, content delivery networks for distributing software, and more. And remember, you don’t have to get into all of this at once, since modern management is spectrum.
Use case #2: Using VDI to support unmanaged endpoints (i.e., contractors, BYOD, home users, etc.). Again, all the advances in laptop management could mean that once you get past the capex of buying the hardware, it might just be easier or cheaper to send these users a corporate-liable laptop, instead of spending the money to deliver a VDI desktop. This echos some trends I’m seeing in mobility. Sometimes, instead of trying to figure out how to securely do work and personal stuff on a single device, it’s easier to just have two devices.
Use case #3: The same thing goes if your plan was to mail thin clients, monitors, and keyboards to home users as a way to make remote endpoint support easier. Again, you have to weigh that against all the ways that laptop management has changed. With AutoPilot, Apple DEP, and other automated provisioning systems, you can cut out a lot of imaging and setup work, so it’s just as easy to send them a laptop.
Use case #4: For years, I’ve been interested in the idea of using a secure remote browser service to wrap a bubble around SaaS web apps. This use case is still emerging, but again, it’s in competition with just drop shipping the user a corporate Chromebook.
Use case #5: Using VDI to deliver Windows apps to non-Windows devices. Here, I do believe that there are more and more users who can get by without Windows apps. For example, you can use Microsoft Office almost anywhere now. (Yes, I know that many users will still need Office on Windows, but plenty don’t.) Chromebooks are also much more capable than ever, and even people like Brian Madden (the person) are arguing that they could be the ultimate enterprise client device.
As a bonus, as you have less of a need for Windows devices to run Windows apps, keep in mind that all these non-Windows devices are getting easier to manage and secure, because enterprise mobility management is pretty mature these days.
As another bonus: If you only need Windows apps occasionally, VDI in the cloud can make this a lot more practical. This is a point I made a few weeks ago: Five years ago, you weren’t likely to build a whole VDI or app publishing environment just so people could switch to fancy Macbooks because they think they’re cool—you’d tell them to take a hike. But now that Macs are way more mainstream in the enterprise, and DaaS means that (in theory) you could cost-effectively publish some apps that might only get used a few times a quarter, this is a viable scenario.
Still many important VDI uses cases!
Of course, there are still plenty of use cases where VDI has unique benefits, for example, keeping data off the endpoint; putting apps right next to data or servers; or renting a cloud workstation by the month instead of shelling out the capex for one that sits under a desk.
And it goes without saying that you could carefully consider the examples I listed above and still come out in favor of VDI. This isn’t all absolute, and everyone’s reasons are different.
But you have to weigh everything against the current state of endpoints and unified endpoint management; the costs of UEM versus VDI; and the value of having a local experience versus the remaining complexities of VDI.
The key point is that if you’re getting excited about the developments of VDI and getting ready for a new project, be sure to also catch up on all the advances that endpoints have made at the same time.