Skip to main content

Containers

It works on your laptop. It doesn't work on the server. The server has a different Python version, different library paths, different OS dependencies. You spend hours debugging an environment problem, not an application problem.

Containers solve this. A container is not a virtual machine. It does not boot a full OS. It packages the application together with its runtime environment and uses Linux kernel features — namespaces and cgroups — to run it in isolation.

The container model​

A VM virtualizes hardware — each VM runs its own full operating system. Boot time is minutes, overhead is gigabytes.

A container virtualizes the OS — containers share the host kernel. They start in milliseconds, use megabytes of overhead. The kernel uses namespaces to give each container its own view of the filesystem, network, and processes. It uses cgroups to limit how much CPU and memory each container can consume.

What this section covers​

Docker — images, Dockerfile, running containers, lifecycle, volumes, and networking. Everything you need to package and run applications in containers.

Production containers — multi-stage builds, minimal base images, running as non-root, container networking internals, and the kernel primitives underneath.