BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Desktops as a Service has been around longer than you might think, but there are a number of hurdles that have kept the promising technology from taking hold.
Today, it's mainly Windows licensing rules that prevent Desktop as a Service (DaaS) from really taking off, said Guise Bule, tuCloud Inc.'s CEO. TuCloud began offering on-demand hosted desktops back in 2009. Its enterprise DaaS offering can scale to hundreds of thousands of users, with a free version of its DaaS Engine for smaller deployments. It also has a community portal, where customers get activity points for engaging in the community, then cash them in to upgrade the DaaS Engine for free.
SearchVirtualDesktop spoke with Bule about the problems with DaaS licensing and how this market will look in 2014.
What do you think about Amazon and VMware entering the DaaS market?
Guise Bule: It's very, very interesting. … Amazon is taking what we call the hosted virtual desktop [HVD], and they're using [Windows] Server images. It's not Desktops as a Service. And they have to go that route because of [Windows desktop OS] licensing restrictions. … If VMware and Desktone don't copy Amazon and deploy Windows Server 2008 R2 images, they're not going to be able to compete in that market.
More on Desktop as a Service
Comparing DaaS product costs
More DaaS options emerge at VMworld
Explaining DaaS vs. VDI
At tuCloud … we're not about building public, multi-tenant clouds. The separation of identities on these multi-tenant infrastructures just isn't secure. We think it's a huge accident waiting to happen. The idea of clustering tens of thousands of businesses across a massive multi-tenant cloud gives me the willies.
You have to look at [Microsoft Service Provider Licensing Agreements (SPLAs)]. SPLA licensing is increasing dramatically over the next three years. [Editor's note: Microsoft has said it will increase Windows Server Datacenter edition licensing by about 38% in January 2014, plus another 13% in 2015.]
What that means is, in the lead-up to Microsoft launching Mohoro, they're squeezing HVD providers out of the market. They're making it increasingly expensive to license a server as a desktop. That tells me that Mohoro isn't going to be about Windows Server desktops; they're going to deliver Windows 8.
What might that mean for how DaaS licensing will work in the future?
Bule: This is the dilemma. … There's no way on earth Microsoft is going to skin Windows Server images to those desktops. It's their OS. You can argue that Microsoft can do whatever they want with their OS, but they'll be opening themselves up to strong legal action if they suddenly start selling desktops and keep everyone else from doing it. What Microsoft does, they have to let the rest of the world do.
The whole ecosystem that runs on Terminal Services and [Citrix] XenApp, they're holding their breath. … [Microsoft is] getting into the DaaS business, and they always have been. If they weren't, they never would have restricted us the way they have. What they've done is slowed the market until they're ready to enter the market. [Editor's note: Microsoft has declined to comment on its plans around DaaS.]
What do you see changing in the DaaS market next year?
Bule: The idea that you can build your own DaaS cloud.
Over the next four or five years, I'd say, [there will be] lots and lots of private DaaS clouds all scattered around. In highly regulated industries, multi-tenancy is a problem. You have to do single tenancy because the cost of making a multi-tenant infrastructure that separates things enough to satisfy those regulated industries about the way they handle data is prohibitive.
Private DaaS clouds being built -- that's the future of DaaS. The future should be multi-tenancy and public cloud; we're going to get there, but only once Microsoft changes the rules.
If Windows licensing gets fixed, does that mean the number of DaaS vendors will increase?
Bule: If the licensing gets fixed, it's going to completely blow up. The biggest thing for those MSPs [managed service providers] out there looking to build a DaaS business is choosing the right architecture -- architecture to scale.
If you look around at the cloud space, the big cloud players like Amazon and Google, they're not using SAN infrastructure; they're using grid distribution to achieve those economies of cloud. … If you can't scale, you're in trouble. … Grid distribution is the way forward; Desktone approaches that halfway, but it still relies on that SAN infrastructure. … Commoditized hardware, no shared components -- that's the model DaaS has to adopt if it's going to have a future.
Why didn't virtual desktop infrastructure pick up like people thought it would?
Bule: Ease of deployment and cost. It's too expensive because you need all the hardware for SAN. The only people who have made money in the VDI space in the past five years [are] storage providers.
The second factor is, it's really hard to configure these things. … They need to be easier to deploy; the learning curve needs to be flatter. [Citrix] Kaviza really led the way in that department with VDI-in-a-Box, which was revolutionary in the sense that it let nontechnical people sit down, get an ESXi server and quickly deploy desktops. We do that [appliance] model because … you should be able to download desktops and build it on a server without any special expertise.
So is 2014 the year of DaaS?
Bule: I've been asked this question every year since 1997. I've got one answer to it: There's never going to be a year of DaaS. The idea that there's suddenly going to be a boom and plateau -- that's not going to work. It's going to be a slow, gradual ascent. We're going to see the industry grow exponentially over the next 10 years.
We're not going to see that 'year of DaaS' until Microsoft changes the rules. … If they don't, I'll still be happy because we're doing nothing but building private desktop clouds for our customers. … So for me personally, it's a very sweet spot being in this space right now. I don't care at all if they change the licensing. If they do, though, I'll become a billionaire.