Kubernetes
koo-ber-NET-eez
An open-source system for automating the deployment, scaling, and management of containerized applications.
Kubernetes (K8s) is an orchestration platform for running containerized applications at scale. Docker runs containers on a single machine. Kubernetes runs containers across a cluster of machines, handling scheduling, scaling, networking, and recovery automatically.
When you tell Kubernetes "I want 5 instances of my web service," it finds machines with available resources, starts the containers, monitors their health, and replaces any that crash. If traffic increases, you tell Kubernetes to scale to 20 instances. If a machine dies, Kubernetes moves the containers to healthy machines.
Kubernetes is powerful but complex. It has a steep learning curve, requires significant operational expertise, and adds overhead that small teams may not need. Most companies under 50 engineers do not need Kubernetes. Managed services like AWS ECS, Google Cloud Run, or even Heroku handle container orchestration with far less complexity.
Examples
A company runs microservices on Kubernetes.
The cluster runs 40 microservices across 20 nodes. Each service has a deployment manifest specifying replica count, resource limits, health checks, and update strategy. Kubernetes manages all of it: scheduling, networking between services, load balancing, and automatic restarts.
Kubernetes auto-scales during traffic spikes.
The horizontal pod autoscaler watches CPU usage on the API service. When CPU exceeds 70% across all instances, it adds more pods. During a product launch, the service scales from 5 pods to 40 pods in 3 minutes. After the traffic subsides, it scales back down.
A startup adopts Kubernetes too early.
A 5-person startup spends two months setting up Kubernetes for their single application. They now have a complex cluster to maintain but the same application that could run on a $20/month PaaS. The engineering team spends 25% of their time on infrastructure instead of product.
In practice
Read more on the blog
Frequently asked questions
When does a company need Kubernetes?
When you have multiple services that need automated scaling, self-healing, and complex networking. Most companies reach this point around 20-50 microservices. If you have a monolith or a handful of services, managed platforms (ECS, Cloud Run, Railway) are simpler and cheaper.
What is the difference between Docker and Kubernetes?
Docker builds and runs containers on a single machine. Kubernetes orchestrates containers across a cluster of machines. You use Docker to create container images. You use Kubernetes to run those images at scale with automated management. They are complementary, not competing tools.
Related terms
The most widely used platform for building, running, and sharing containerized applications.
Lightweight, portable packages that bundle application code with all its dependencies.
An architecture where an application is built as a collection of small, independent services.
On-demand computing infrastructure (servers, storage, networking) provided by a third party.
Adding more servers to handle increased load, instead of upgrading existing servers.
Want the complete playbook?
Picks and Shovels is the definitive guide to developer marketing. Amazon #1 bestseller with practical strategies from 30 years of marketing to developers.