You know, I’m not a fan of VDI. That’s not because it isn’t theoretically cool but because it practically sucks!
All vendors, especially storage ones, promote a recipe for a cheap VDI infrastructure but, when it comes to real life, limits, constraints (and Microsoft licenses costs) are as many (and irritating) as a swarm of bees and the only thought that comes to your mind is to stay well away from it.
There are (a few) exceptions
I also know that there are exceptions to what I’ve just said (so please don’t jump up to say that I’m totally wrong and that you have the right solution since the beginning!).
Some startups did a very good job on VDI but in most of the cases they are niche players and in other cases they have very interesting but vertical solutions.
And now DaaS!
A few months ago some brilliant marketing guy forged a new term DaaS (Desktop as a Service), this is a VDI delivered from the cloud.
The concept is interesting. And I mean theoretically interesting as it was in the case of VDI!
Suddenly, as you can imagine, we started to hear about Public and Private DaaS… that’s another bandwagon to jump on!
Don’t get me wrong but Private DaaS sounds a lot like a coat of paint on VDI, doesn’t it?
On the other hand Public DaaS has it’s own practical problems too.
Even if Microsoft agrees to sell its desktop licenses at a very low cost to cloud providers (and please let me know why Microsoft should do that??!?) there are still doubts about the overall efficiency of this kind of architecture.
There are many possible scenarios on how implementing DaaS but I’ll try to point out the firsts two issues that come to my mind:
1) if you still have on-premise legacy applications, how can you afford to have virtual PCs on the cloud and thin clients on the desks? (bandwidth, security, latency issues are only the beginning).
2) Which kind of thin client will you adopt for DaaS? a Web browser? a proper thin client? a PC? a tablet? Each one of these has its own problems (sometimes licenses) and management issues could be countless!
And I’m sure you can add much more than that if you look at your current infrastructure.
Bottom line
Desktop virtualization has failed to be broadly adopted when it was the right time simply because its advantage is not as clear as in the case of servers virtualization. that’s the point!
Next generation desktop applications/clients are developed with web browsers and tablets in mind.
End users want tablets for doing their job.
Tablets go in the hands of average users, the same users that are theoretically suitable for VDI.
So, to recap, Desktop (and, as a consequence, VDI) market is shrinking in favor of tablets… why should DaaS change this?
BTW, please stop using the term “private DaaS”!
The days when every enterprise thought VDI would cure their operational and budget concerns around EUC are long over. The expectation that the UX is going to be on-par with a distributed device model is no longer relevant either. The enterprise has shifted its focus to deploying VDI in a very targeted way. Security, call centers, and even for BYOD is some scenarios. ISVs are rapidly innovating in this area, and the MSP market is searching for a solution to sell to their consumers. Ultimately, DaaS isn’t where it needs to be. Yet. Give it a chance. Give MS a chance to get all their dominoes stood up as well. They will open the license up when they are ready to offer DaaS on MS Azure public cloud. Note the RDP 8 functionality and mobility support as of yesterday? Should also reference stuff like the VMWare acquisition of Desktone, and the Virtual Bridges DaaS release. All happened in the last week. The reason this stuff is moving is because it makes money and companies are buying it. Further, the argument that the desktop days are over in favor of tablets is only true for the form factor. The operating system behind them is just portable tablet friendly one. Like Win 8 and Chrome.