What Is Docker? Containers Explained Simply

Docker is a tool for packaging software into containers: self-contained bundles that include the application, its runtime, libraries and configuration. A container built on a developer's laptop runs identically on a colleague's machine, a test server and a cloud host, because everything it depends on travels with it.

Before containers, deploying an app meant installing the right language version, system packages and settings on every machine by hand, and something always differed. Docker replaces that with a recipe.

Images and containers

An image is the recipe's result: a read-only snapshot of a filesystem with the app inside. A container is a running instance of an image. You can start ten containers from one image, the way you can run ten copies of a program.

Images are built from a Dockerfile, a short text file of instructions:

FROM node:22-alpine
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build
CMD ["node", "server.js"]

Each line adds a layer: start from a base image with Node.js, copy the code, install dependencies, build, and declare the command to run. Layers are cached, so rebuilding after a small change is fast.

Why teams use Docker

• Consistency. The same image runs in development, testing and production.
• Isolation. Two apps needing different Python versions run side by side without conflict.
• Speed. Containers start in milliseconds and share the host's operating system kernel, unlike virtual machines.
• Portability. Every major cloud can run containers directly, and orchestrators such as Kubernetes schedule thousands of them.
• Reproducible onboarding. A new developer runs one command instead of following a ten-page setup document.

Basic commands

docker build -t myapp .          # build an image from the Dockerfile in this folder
docker run -p 3000:3000 myapp    # start a container, mapping port 3000
docker ps                        # list running containers
docker logs <container>          # read a container's output
docker compose up -d             # start a multi-container stack in the background

Compose files are written in YAML. If you have configuration in JSON, the JSON to YAML converter produces correctly quoted YAML for them.

Related concepts

• Registries such as Docker Hub and GitHub Container Registry store and share images, like package registries for containers.
• Kubernetes runs containers across many machines, handling scaling, restarts and networking.
• Volumes persist data outside a container so it survives restarts; containers themselves are meant to be disposable.
• Environment variables configure containers at start time (-e DATABASE_URL=...), keeping secrets out of images.