Who uses Ubuntu, where and why? That's a question a lot of parties in the open source channel likely ask themselves. It's also one that's hard to answer, since public data on Ubuntu deployment is scarce. But it became a little less so recently with the publication of the results of Canonical's latest survey of Ubuntu server users. Read on for the highlights.

The stakes surrounding the deployment of Ubuntu on servers are high. They're a measure of Canonical's success relative to other competitors in the open source channel such as Red Hat and Novell. They also say something about the health of Linux as a whole. And last but not least, they reveal a lot about technology trends in the open source ecosystem and beyond -- where open source is being deployed, what kind of applications are popular and where the channel's momentum might lead it in the future.

Ubuntu Server Edition Survey 2012

On Feb. 14, 2012, Gerry Carr, Canonical's director of Communications, posted results of the latest survey of Ubuntu server users. This is the third time the survey, completed on a voluntary basis on Ubuntu's website, was taken. The last one was in 2010.

The full report is available here, but some notable highlights that help reveal how Ubuntu is being used -- and where it might be headed in the future -- include:
  • Traditional applications -- web, database and mail servers -- constitute the clear majority of current Ubuntu deployments. That suggests users trust Ubuntu for high-volume, mission-critical services, although it also means Ubuntu may still have room to grow when it comes to more novel server technologies, such as those related to the cloud.
  • And on that note, when asked whether Ubuntu is a viable platform for cloud-based deployments, a whopping 70 percent of users declined to answer (of the remainder, 27 percent answered yes and 3 percent said no). Canonical interprets that trend as evidence that many users currently lack sufficient experience working with the cloud, which may be true, though that still seems like a huge number of abstentions. Whatever the explanation, it seems clear that Canonical needs to work harder to encourage Ubuntu use for the cloud -- as indeed it has been -- to avoid being left behind in this growing segment of the IT world.
  • Interestingly, most Ubuntu servers are running on traditional desktop PCs. As the report pointed out, "the prevalence of the tower PC probably reflects the number of hobbyist and home users responding to the survey." I also wonder, though, if this suggests that Ubuntu may still lack as strong a presence in the server rooms of large organizations as some of its competitors.
  • More than half the respondents were located in Europe, and barely a quarter were in North America. According to the report, other data (which Canonical has not shared) suggests that in fact most Ubuntu server deployments are in the United States, with Europe overrepresented in the numbers released Tuesday because the survey was better publicized there. I'll take the report's word for that, but assuming it's true, the higher response rate among Europeans nonetheless may be linked to the stronger presence of Ubuntu among governments and other large organizations in Europe.
  • Although the data on users' perceptions of the compatibility of Ubuntu with other platforms is a little hard to make out, since it's presented only in a graph, it looks like most respondents believe Ubuntu works pretty well with other systems. In addition, there appears to be no major difference in its perceived ability to integrate with open source vs. proprietary platforms, suggesting that Ubuntu has done a good job of building bridges into the closed-source world -- no mean feat given the many hurdles to integrating open source technologies with proprietary ones.
  • Last but not least, KVM is now more popular than Xen as a virtualization hypervisor among survey respondents -- but both still lag behind VMware. And surprisingly, VirtualBox ranked relatively high given that it's mostly a desktop-oriented solution. (Perhaps this finding, however, also reflects the fact that many respondents are hobbyists who may be running server technologies alongside desktop ones).
All in all, the survey's use is a bit limited by the nature of the respondents, who seem to represent certain segments of the Ubuntu server community disproportionately. But nonetheless it helps to illuminate some key facets of Ubuntu Server Edition's future, with the takeaway points being the following: It has a lot of room to expand in the cloud; it's doing an impressive job of integrating with other parties in both the open and closed-source channels; and its popularity in the server rooms of large organizations may still lag beyond the appeal it enjoys among hobbyists.

Now, if only Canonical would take a survey of desktop users, that would be truly fascinating ...