We've discussed third-party hosted cloud-based desktops-as-a-service quite a bit in the past. My general consensus has been that while they make sense in certain cases, the logistical and technical limitations mean that third-party hosted desktops will probably be a minority for the near future. (I said they weren't ready yet as recently as March.) But what's interesting is that the past few months have seen an explosion of offerings in this space.
And I'm not talking about those interesting-yet-quirky browser-based desktops (here and here) or the future desktop (small "d")--I'm talking about gold old-fashioned monolithic Microsoft Windows-based (Big "D") Desktops, accessed from a third-party provider via a remote display protocol.
Just take a look at all the companies who now have offerings here: (Seriously, half of these are new in the past 30 days.):
- Desktone (using vSphere, some custom software, & supporting HDX)
- IBM (using Virtual Bridges VERDE)
- Rackspace (using XenDesktop)
- WiPro WDaaS (XenDesktop)
- Dell vDaaS (XenDesktop or View)
- CSC (XenDesktop)
- Maybe Microsoft? (With Azure-based desktops, possibly connected to via RemoteFX?)
- Plus lots of others: Nasstar, ThinHost, Nivio, dinCloud,
(By the way, Desktone has a trademark on the term DaaS, which is why you see all these weird things like vDaaS and wDaaS.)
So what do you think? Is the third-party cloud-based hosted desktop market ready for an explosion? Or is this industry primed for an implosion?
My personal view is that we're not yet ready in any kind of major way. I just don't think the technology is ready enough for most users to be far away from the place where their desktops run. Remote protocols over Internet connections just don't support the unlimited graphics, peripherals, and speeds that we need today. Sure, you could argue that these cloud-based desktops would be a good fit for a subset of desktops--like only the physical desktop (i.e. non-laptop) users with simpler needs, but that's the argument that I'm making. These things are great for simpler environments and to play with (and they make for good DR sites), but they're not for everyone today.
On the other hand, if you can come up with a good reason as to why your desktop should run in a datacenter, then I'm all for getting that desktop the hell out of your datacenter and into a cloud provider. Let them deal with all the mechanical crap so that you can just focus on your apps and data. Yeah, I know that moving the desktop off-site requires a lot of work, like figuring out authentication and data storage and maintenance. But I truly believe that the actual hosting should be done offsite somewhere. And of course if you want to move your desktop into a datacenter, then you're probably motivated by something other than saving money. (After all, if you just wanted to save money, then why would you take what could be done on a $300 piece of hardware and move it into a datacenter where you have to pay for cooling, heating, bandwidth, virtual desktop licenses, etc.
To be clear, though, I still feel that the limitation right now is technical. And I feel that we'll ultimately overcome it enough so that 100% of Windows desktops are running in the cloud. But I don't we're ready for that right now. So why are all these companies here all of the sudden?