One of the hardest questions to ask in tech (and not only in tech) is "why?". There is a whole lot of activity getting done because of market momentum, or peer and vendor pressure. IT ends up doing things because they seem like a logical next step from what was done in the past, or it's what most of their peers are doing. At this point it's important to keep the stereotypical mother's retort in mind: "if they all jumped off a bridge, would you do that too?".
The momentum trap, where IT just keeps moving forward along its existing path, is particularly hard to avoid at times of transformation. The move from mainframes to distributed systems is an example of a past transformation where IT teams that missed the wave struggled to catch up. Cloud computing represents that kind of transformation.
In this spirit, it was interesting to hear from TheInfoPro (part of the 451 Group) that 83% of their survey respondents are facing significant roadblocks keeping them from moving to the next phase in their journey. Those roadblocks were:
I have been having an exchange with Chris Evans of Architecting IT on this very topic of roadblocks to cloud adoption. Basically, what I see is a lot of self-referential conversations within IT, focusing narrowly on technology and not on what the technology is supposed to achieve.
Because the past focus was on virtualising more services and increasing density of virtual servers, much of the industry is focused on continuing those efforts. Now that is certainly a valid effort, especially since it was only in 2010 that we achieved crossover, with more virtual servers than physical ones, and that future is still quite unevenly distributed. But is this what users outside the IT department need or care about?
It is at this point that the roadblocks come in. Careers have been built on virtualisation and on datacenter management before that, so resistance is often found around that radical switch to listening to what users want instead of just telling them what they could have, like in the good old days. This is why we see the plateau in public cloud adoption rates that TheInfoPro report flags. New-build IT from startups and the like is going to the public cloud as a matter of consequence, but old-school enterprise IT teams are not comfortable with that model and what it means for them.
The problem for the foot-draggers is that the users are doing an end-run around them. There is an oft-quoted maxim that the Internet interprets censorship as damage and routes around it, and users are doing just that to what is known as the Department of No. IT teams are playing catch-up, trying to ban public cloud services and force people to use internal would-be equivalents.
Fortunately for IT, there is another old saw that applies here: "if you can't beat 'em, join 'em". Don't try to block public cloud services, make them available to users in a way that is controlled, with placement policies to ensure correct matching of business services and infrastructure options. Instead of objecting that there is no approval process for public cloud, include those resources in your existing approval and tracking process. And don't go too far in the other direction either; there are all sorts of reasons why you might still need your own datacenter as well as all the public cloud you can eat. There are still lots of happy mainframe customers around...
Bottom line, sit down with a clean slate and try to figure out what the business needs, then how IT can deliver it. If you want to see how you and your cloud plans stack up, take the free survey developed by CIO Magazine and CIO.com, and get a report tailor-made to your situation.