Join us in February 2024 for an exclusive Cloud Native Copenhagen meetup, where we'll explore two compelling talks about the problems (and maybe solutions) around scale.

Agenda

16:30 Door open, finding seat, enjoying snacks and networking

17:00 Welcome by the Cloud Native Copenhagen

17: 05 Welcome by our host Onomondo

17:15 Talk 1: Beyond Planet Scale: Is Kubernetes Ready for the Edge Computing Challenge?

18:00 Break with food and drinks

18:45 Talk 2: How to scale to N number of environments using ArgoCD and Flux Image Automation

19:30 Break with snacks, networking

20:00 Thank you for today

TALK 1: “Kubernetes is the answer to planet scale, and also for 1-node edge" by Sofus, Team lead and Trainer at Eficode

Kubernetes describes itself as planet scale, and much of the fuzz around it is just that, making it scale.

But what about scenarios where you are in a remote location?

Where you do not have direct connection to your cluster, and where scalability is basically just that one-node cluster?

Would kubernetes also fit the needs there?

In this hands-on talk I will go into the world of “edge kubernetes” and showcase how a toolstack consisting of open source software can help tackling the problems encountered in having compute in rural and remote areas with flaky internet connection.

TALK 2: “How to scale to N number of environments using ArgoCD and Flux Image Automation” by Dag Bjerre Andersen, Site Reliability Engineer at Visma e-conomic

Most talks and articles on how to structure your Kubernetes manifests use simple examples of a company with a production, staging, and testing environment.

But what do you do if you have 20 staging/testing environments, each with 50+ applications, and each application on each environment needs to support different image versions?

The more I think about it, the more questions come up. How do we structure the code? How do we ensure we do not end up with 1000+ manifests? How do we update the image tags when new image versions need to be tested? How easy is it to add/remove environments? How maintainable is the solution?

This is not a trivial problem, and there are many less-than-desirable solutions. I would like to show you how my team chose to solve it. Hopefully, this will serve as an inspiration for others when they are designing their systems.