Talos and KubeCon 2019

Thu Dec 12 2019
Timothy Gerla

The Talos Systems team and I attended KubeCon 2019 in San Diego a few weeks ago. It was my first KubeCon and I was stunned by the scale and scope of the conference. It reminded me of some of the recent AWS re:Invent shows: the crowded hallways and the jam-packed show floor seemed out of proportion for a software platform that's barely 5 years old. It underscored the massive and rapid adoption of Kubernetes, and the breadth and depth of projects being built on top of Kubernetes was astonishing.

Since Thanksgiving came right after the conference, I got to thinking: how would I describe Kubernetes, and the work that my company does, to my friends and family who are not in the infrastructure industry? I experimented with a couple of approaches, but I think I found one that works. Before I go into my analogy, I should first say that I tend to be a bit cynical about technology trends that come and go. Sometimes that means it's hard for me to step back and see the growth and improvements in maturity of our industry since I have been a part of it. When I first encountered Kubernetes, I found it to be like a lot of software: seemingly overly complicated, and difficult to wrap your head around. But as I learned more about the kinds of problems that people were solving with Kubernetes, and learned a bit more about the abstractions and architecture, my cynicism faded.

So, back to the analogy I have been using to describe this stuff. If you go back 15 years and look how web applications were run, you might have seen a server under someone's desk, running some business-critical software. It runs fine for months, until someone accidentally kicks the power switch and takes the server down. (Ask me how I know!) After one or two failures like this, you could probably count on the server being moved to a rack in the closet. That's better. Then a power outage hits your building. Fine, we'll set up a separate server in the other branch office a few time zones away. More complicated, failure-resistant architectures begin to emerge. Then comes cloud computing, which makes it easy to expand the footprint of your infrastructure worldwide, and let someone else worry about the power strips under the desk.

Throughout this time, our industry learned lessons, solved problems, learned about different ways to fail, and built countless tools to improve the lives of operators, developers, and users. We created layers of abstraction to make systems more resilient, faster, and more secure. Before Kubernetes, a lot of these abstractions had to be built by hand. Architectures and system design previously built piece-by-piece, app-by-app, have now become a language, codifying the lessons learned over the early years of the internet and through countless hours of system design and troubleshooting.

How does our new operating system for Kubernetes, Talos, fit into this story? We think that the rapid adoption of Kubernetes marks an important turning point for the way that our industry manages systems. The new layers of container orchestration and new kinds of abstractions mean that the role of the operating system has changed. Kubernetes is a new model for running applications, and deserves a host operating system that is designed from first principles to fit into this new model. We are building an OS with a modern API, integrating with Kubernetes components like Cluster API, and our goal is to be the best OS on which to host Kubernetes.

Are you interested in helping us? We would love your feedback. See below for some ways to get started.

Try Talos

If you are interested in trying Talos, check out our Getting Started guide! You can set up a test cluster in a local Docker environment in just a couple of minutes.

Get Involved

We’d love to talk to you about the project. Join our community to chat with us on Slack, join the mailing list, and check out the source code at GitHub.

Talk to Us

Do you have questions? Are you interested in learning more about our services and support offerings? Contact us and we’ll get back to you right away.