Kubernetes creators launch startup, take on Docker

Craig McLuckie and Joe Beda, key creators of the container orchestration framework Kubernetes at Google, have partly pulled back the curtains on a new venture: Heptio, a company that offers an ecosystem of services and products around Kubernetes.

They have yet to make public what that business will look like and what its product line will consist of. But other ventures that have sprung up around related technologies in the field may offer some clues.

Think globally, act locally

A TechCrunch article published today provided a profile of what McLuckie and Beda had in mind for Heptio: a “strong need for a Kubernetes-focused company to emerge that wasn’t tied to an adjacent business,” per McLuckie.

McLuckie believes Kubernetes users want to employ the framework without having to throw in their lot with a particular cloud or infrastructure arrangement. Google itself has been working toward a vision of a powerful cloud built entirely from open source components, but at the end of the day, it still requires buying into Google’s vision of how the pieces are assembled, held together, and monetized.

A couple of posts on the Heptio website provides slightly more context. The first, a general announcement about the company, says little other than that the company will “bring Kubernetes to enterprises in order to accelerate software development, increase infrastructure efficiency, and reduce the complexity of managing software at scale.”

Cloud native and not-so-native

But another post that discusses the term “cloud native” is more enlightening. “Cloud native” is a term McLuckie has attached to the Cloud Native Computing Foundation (CNCF), his initiative to develop open source platforms for managing containerized applications.

With the CNCF now established to set standards and handle evangelism for containerized apps, McLuckie and Beda are turning their attention to the lessons learned from creating such items and applying them to an enterprise application architecture, wherever it lives.

“You don’t have to run in the cloud to be ‘Cloud Native’,” writes Beda in the blog post, but he notes “there are very few examples of [the ‘cloud-native’] philosophy being applied outside of technology early adopters.” In other words, the cloud-native philosophy has not been put to use on a scope that would benefit small to medium-size business or modest enterprises.

Sound familiar?

So far, it seems the long-term plan for Heptio is to create the tooling, packaging, and support base for Kubernetes that Docker created for its version of container technology.

Docker caught on not because it promoted an entirely new idea, but because it commodified existing concepts and technologies that hadn’t previously been commodified, wrapped in a useful metaphor. Kubernetes is more of an original creation, but its roots are in a project that was developed for Google internally. There’s always room to increase its accessibility.

The bullet points listed in the second blog post sound like wish-list items for achieving that goal: enable nimble teams, reduce heavy lifting, make infrastructure more reliable, provide insight into applications, enhance security, and better employ what’s already running.

Thing is, they’re all in line with the promises already made by Docker, Mesosphere DC/OS, and other enterprise-aimed projects. If Heptio wants to draw a paying audience, it must show that Kubernetes can achieve all those goals in a manner nothing else can.