Learn Docker
& Containers

Before installing anything, get the mental model right. A Virtual Machine (VM) simulates an entire computer — its own operating system, its own kernel, its own hardware. It's powerful but heavy. A container shares the host computer's operating system but keeps everything else isolated — making it lightweight, fast to start, and cheap to run. You can spin up 50 containers on a laptop that could only handle 3 VMs. Install Docker Desktop (free), run docker run hello-world, and watch the magic happen.

An image is the blueprint — a read-only snapshot of an app and all its dependencies. A container is a running instance of that image. One image can launch dozens of identical containers simultaneously. Images are stored and shared on Docker Hub — a public library with hundreds of thousands of ready-made images for databases, web servers, programming languages, and more. Pull an nginx image, run it, and open your browser to see a live web server running inside a container in seconds.

A Dockerfile is a plain text recipe that tells Docker how to build your image — start from a base image, copy your code in, install dependencies, and define what command runs when the container starts. It's usually just 10–15 lines long. Write a Dockerfile for a simple Python or Node.js app, build it with docker build, and run it with docker run. When your app starts inside a container you built yourself, this stage's most important lesson lands: you now control the entire environment your app runs in.

Real applications are never just one thing. A typical web app has a frontend, a backend API, and a database — three separate services that need to run together and talk to each other. Docker Compose lets you define all of them in a single YAML file and start the whole stack with one command: docker compose up. This is how development environments are built at companies of every size. Write a Compose file that runs a web app alongside a PostgreSQL or MySQL database container.

By default, when a container stops, all data inside it disappears. That's a problem for databases and file uploads. Volumes solve this by connecting a folder inside the container to a folder on your real machine — so data persists even when the container is gone. Similarly, containers need to talk to each other over a network. Docker creates its own internal networks and lets you define which containers can see each other. Both concepts are essential for running any stateful application in production.

Once you've built an image, you need somewhere to store it so servers and pipelines can pull it down. That place is called a container registry. Docker Hub is the public one — free for public images. AWS has ECR (Elastic Container Registry), Azure has ACR, and Google Cloud has Artifact Registry — all private registries that plug directly into cloud deployments. Tag your image, push it to a registry, then pull it down on a cloud VM and run it. This is the exact flow your CI/CD pipeline will automate later.

Docker runs containers on one machine. Kubernetes (often called "K8s") runs and manages containers across hundreds of machines — automatically restarting crashed containers, scaling up when traffic spikes, and rolling out updates with zero downtime. You don't need to master Kubernetes at this stage — just understand what problem it solves. Learn the core concepts: Pods (a running container), Deployments (manages replicas), and Services (exposes a pod to the network). Use Minikube to run a local Kubernetes cluster for free.