New kubernetes usage survey info, with IBM 5151 retro-dream charts. Scroll to the end to chill out your ear holes on the smoove waves.
Here’s an analysis of the most recent CNCF survey I did. It’s been on the shelf for awhile since we were looking for somewhere to publish it, on InfoQ it turns out. As you can guess, I liked making those charts.
Speaking of kubernetes surveys, we’re about to publish the 2022 State of Kubernetes survey. This is the forth one VMware has done (Heptio did the first in 2018, since acquired by VMware, as you’ll recall). I’ve written analysis and blog posts for the past few years (here’s 2021’s and 2020’s), so I’m finishing one up right now. Here’s a draft of the post, on the question about benefits people are getting (“realizing”) from using kubernetes:
Good for Ops, Could be Better for Dev
When it comes to cloud operations, kubernetes has clear benefits as the responses in the chart above show.
However, the two developer related benefits are actually declining. Back in 2020, respondents reported quicker release cycles and application modernization as benefits, but both of these have decreased since then. What’s going on here?
To speculate, once again I’d cite the growing use of kubernetes, both for more developers but also for more complex applications. As more application developers target kubernetes for deployment, you’ll encounter more people who are not interested in, nor skilled, at tinkering around with infrastructure. They want to focus on their application code. Indeed, the most interesting discussion in the kubernetes space now is around “developer experience,” which is to say, making kubernetes easier for application developers.
The good news is that the kubernetes ecosystem is working on this problem. For example, check out what we’re doing in the Tanzu Application Platform. We’re pulling together many open source components in the kubernetes world with a laser focus on improving developer productivity and experience.
I haven’t written the text on this chart yet, but here’s where people are running kubernetes:
I need to go double check this, but I believe “hybrid” in the survey means using private (on-premises) and public cloud.
Need more smooth sailing? Check out the 9 minute 43 second version!