OpenStack makes production gains, Fuel rises
Not so fast.
In fact, the overall trends were largely consistent with what we've been seeing in the past, a fact which the Foundation chalks up to a suggestion that "the user survey has achieved a stable cross-section of the community". More than that, however, is the fact 64% of OpenStack deployments are now running in production, rather than proof-of-concept or dev-test environments, showing a maturing of the technology.
Not that there weren't ANY changes, of course.
The total number of OpenStack clusters was up 15% to 405, and the number in production grew by 5%, which is consistent with the anecdotal evidence of OpenStack's growing adoption. Of those clusters, the vast majority -- 72% -- were private clouds, either on premise or off.
Interestingly, while "reducing cost" was, as always, the most common primary motivator for adopting OpenStack, that wasn't entirely the full story. When looking at respondents' top 5 choices, it seemed that if reducing cost wasn't their first concern, then they weren't worried about it at all. On the other hand, virtually all respondents -- 97% -- were concerned, to some degree, about standardizing the toolkits with which they're working with clouds in a hybrid environment.
In addition, containers were once again the most popular emerging technology on people's minds, and Kubernetes, which was nearly tied with Cloud Foundry six months ago, now outpaces it by a wide margin. (This may be related to a large drop in respondents who said they were interested in PaaS, from 54% to 35%.)
Other significant changes include the Deployment Tool category, as Fuel jumped from number 5 to number 3 to become the top purpose-build installer (behind Puppet and Ansible), and an indication that deployments are much more likely to include more than one type of guest operating system as the Ubuntu (#1) and RHEL (#3) both made gains while CentOS (#2) remained largely unchanged.
You can see the entire report here.