Container Orchestration & Kubernetes covers: Designing Distributed Systems. Understand Kubernetes architecture, deploy applications to K8s, manage services, scaling, and configuration. Know when K8s is the right choice and when simpler tools (Kamal, Docker Compose) suffice.
Deploying Rails on K8s vs Kamal. When each makes sense. K8s shines when you have many services, need auto-scaling, or are part of a platform team. Kamal shines when you have a single Rails app and want simplicity.
| K8s Concept | Rails Equivalent |
|------------|-----------------|
| Pod | One Puma process |
| Deployment (3 replicas) | 3 Puma workers behind a load balancer |
| Service | Internal routing (like Nginx upstream) |
| Ingress | Nginx reverse proxy with SSL |
| ConfigMap | credentials.yml.enc or env vars |
| CronJob | Whenever gem + cron |
| HPA (autoscaling) | No direct equivalent (manual scaling) |
| Helm chart | Kamal deploy.yml |
This course unlocks once you've finished its prerequisite. Open prerequisite →
I am going to say the thing that Kubernetes evangelists will not: K8s is overkill for most Rails applications. If you have one app, a few background workers, and a Postgres databa…
Every time you've bolted Nginx in front of Puma, put PgBouncer between your app and Postgres, or run a log-shipping process next to your web server, you were implementing a distri…
Work through each item before the checkpoint.
Deploy a full Rails stack to Kubernetes with a Helm chart — the app at 3 replicas, PostgreSQL as a StatefulSet, Redis, Sidekiq workers, and an Ingress with TLS. CI spins up a real…
4 lessons. Read in order; spiral back when you need to. By the end you'll have used the core ideas twice — once on the abstract, once on something you'll meet at work next week.