News Stay informed about the latest enterprise technology news and product updates.

Will Microsoft enter the DaaS market fray with Mohoro?

Microsoft loses out big time if people move off Windows, so it would do well to join the DaaS market.

Microsoft has a lot to lose if it doesn't join the Desktop as a Service market.

I've written a lot about Microsoft licensing, focusing on the fact that no matter how you slice it, Microsoft is making our lives miserable. This is especially true for DaaS providers, since Microsoft doesn't allow them to share compute resources (called multi-tenancy) with multiple clients. Each piece of compute hardware (and, if you ask some people, storage) must be designated for a single company. That results in excess processing capacity going to waste that could be used to realize vast economies of scale. Not only is it inefficient, it's also maddening that Microsoft arbitrarily restricts something that is technically possible.

Ultimately, it means that DaaS costs more than it has to if you want to deploy Windows client operating systems. To combat this, DaaS providers are using single-user Remote Desktop Session Host (RDSH) servers, which aren't governed by the same asinine restrictions because you simply need to buy a per-user RDS client access license for each user. Plus, there is a service provider licensing program for RDSH servers, which means they can share any resource they want among their clients. This practice is catching on at service providers and within organizations, and the very fact that we have to do this is the catalyst of the #fixvda movement.

Musing Microsoft's Project Mohoro

Imagine, then, the reaction of DaaS providers to hear rumors of Microsoft's Project Mohoro. Mohoro is expected to be Microsoft's own DaaS offering, hosted in Azure. Rumors circulated back in May, and while there hasn't been any more information emanating from Redmond, the DaaS market is shaping up so that if Microsoft doesn't have an offering for its own OS, it will be left behind.

Citrix has one. VMware has one. Dell, well, theirs was based on the Desktone platform that VMware now owns, so we'll have to wait and see on that front. The point is that Microsoft, the company with the most to gain from a DaaS tool as well as the most to lose if people move off Windows, needs to be in this landscape.

More on the
DaaS market

What customers need to know about DaaS

How are DaaS and VDI different?

FAQ about Desktop as a Service

But what will Mohoro look like? Will it be RDSH-based, using some version of Windows Server? If so, which version of Server? Some use cases still require 16-bit applications. Will you be able to run Windows Server 2003 R2 or Windows Server 2008 R1? Despite those questions, the RDSH method is the easiest, because running Windows 7 or 8 would be complicated due to the licensing restrictions Microsoft itself imposes.

In July, Microsoft amended its licensing agreement for Azure to expand the use of Remote Desktop Services (RDS) in Azure virtual machines beyond admin tasks. RDS is an umbrella term that includes RDSH (terminal services) and Remote Desktop Virtualization Host (VDI), but Microsoft was careful to note that client versions of the OS are still not able to be used due to the multi-tenancy restrictions.

All this adds up to RDSH being the most likely candidate. When it was first rumored, Mohoro was called RemoteApp as a hosted service, which would indicate seamless Windows apps delivered from Azure. Still, all the news related to this came out between May and July. There are new indicators that DaaS's stock is rising, so could there be more to this?

Microsoft licensing's gotta give

As long as Microsoft plays by its own licensing rules, no new complaints should arise aside from partners squawking about stealing customers. If, however, Microsoft decides that since it owns the OS it can make up its own rules, you can expect an IT donnybrook to break out. The idea most talked about would be that Microsoft can escape its own licensing limitations by creating another version of Windows that is virtually identical to a client version of Windows and license it only to itself. That theory likely draws most of its supporters from a pool of people that are known to wear tin foil hats, but it is possible.

What it means, though, is that if Microsoft is going to play in the DaaS market and not just deploy applications, something will have to give. Microsoft will either have to fully back the single-user RDSH method (which simply has to cost them money compared to selling client OS licenses), make changes to Windows client OS licensing, or do something more drastic like creating an OS for its own use only.

Rumors of Project Mohoro persist, so I expect it to happen, but exactly how disruptive it will be remains to be seen. If it's just RDSH from the cloud, it will be simply another deployment option. If it's more than that, it will be very interesting to see how things shake out in the evolving DaaS market.

This was last published in November 2013

Dig Deeper on Virtual desktop infrastructure and architecture



Find more PRO+ content and other member only offers, here.

Join the conversation


Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.

How do you think Microsoft will build its DaaS offering?
Microsoft is going to make its DaaS offering very narrow and suitable to round out its own offerings and solve a specific point problem. There is too much risk to the Windows business to make an excellent DaaS solution exactly the way customers want it, and in so doing they would be handing more client business to Apple and Google. There are also many other products with potentially competing capabilities in the Microsoft server and tools portfolio. You won't see a broad, flexible offering like Desktone coming from Microsoft. This will be an app delivery solution, or it will at least start that way.
Microsoft has shown a willingness to unfairly compete against SPLA partners already, as evident by their Office 365 plan pricing. For example, the Office 365 Small Business Premium plan provides the desktop versions of all core Office apps for $13.25 per user per month, which undercuts SPLA subscribers COST for the Office Pro Plus suite (by itself) by roughly 40%.

To add insult to injury, the Office 365 Small Business Premium plan also includes "free" hosted Exchange, web conferencing and IM, file sharing via SkyDrive Pro, etc. An SMB that is looking for cloud hosting of Office apps, along with cloud storage, is going to legitimately balk at the price an SPLA partner could offer on a comparable package. For SPLA partners, we currently need to provide hosted Desktops, hosting of third-party apps, and other cloud services in order to differentiate ourselves from the Office 365 offering.

If Microsoft decides to become a DaaS provider, they are sure to once again offer plans that far undercut the COST to SPLA partners for Windows Server and RDS licenses, not to mention the huge disparity with Office licensing costs already discussed.

In short, if SMBs are able to get Cloud Desktops, along with hosting third-party legacy apps via VMs running under Azure and accessible from the Cloud Desktop, then SPLA partners simply won't be able to compete.
Regardless of stupid licensing and windows versions, remote apps simply make the most sense. Microsoft strategy is "devices and services" and RemoteApps fit in nicely with this strategy.

In the end you'll have a nice rich client device like a surface, a laptop or even an XBOX with new-style metro apps running side-by-side with web applications. To make the transition easier, Microsoft can seamlessly integrate legacy applications from the cloud.