nanyang-system-developers

Containers and Deployments

What is a Container?

A container is a lightweight, standalone package that includes everything needed to run a piece of software: the code, runtime, system tools, libraries, and settings. Containers make it easy to run applications reliably across different computers and environments.

Containers vs Virtual Environments

Virtual environments (like Python’s venv or Poetry) only isolate Python packages for a single project, but still use your system’s operating system and Python installation.
Containers go further: they package the entire operating system environment, which includes other software installed on your system, so your app runs the same way everywhere, regardless of the host system. This means you can run a container on your laptop, a server, or in the cloud, and it will behave the same.

Summary: Two virtual environments will have different versions of Python or packages, but two containers will have different installed software, even different operating systems. This makes containers more powerful for deploying applications.

Why Use Containers?

Consistency: Your app runs the same way on your laptop, your teammate’s laptop, or a cloud server.
Isolation: Each container is separated from others, so you can run multiple apps with different requirements on the same machine.
Portability: Containers can be easily shared and deployed anywhere that supports container technology.

Containers in the Cloud

Many cloud platforms use containers to host applications. For example:

Google Cloud Engine and Replit both use containers to run your code in isolated environments. This means your app gets its own environment, and you don’t have to worry about conflicting with other users’ code.

Docker

Docker is the most popular tool for creating and running containers. With Docker, you can:

Build a container image for your app (using a Dockerfile)
Run your app in a container on your laptop or server
Share your container image with others or deploy it to the cloud

For more, see the Docker documentation.

Dev Containers in GitHub Codespaces

GitHub Codespaces give you a Docker container in the cloud, called a dev container, to work in. A dev container is a special container with all the tools and dependencies your project needs. This means you can start coding immediately, without installing anything on your own computer, and without affecting your local setup.

You can configure the setup for each repository’s dev container by adding a configuration file. This configuration file appears in the repository as .devcontainer/devcontainer.json. This file specifies:

The base image to use for the container (e.g. mcr.microsoft.com/devcontainers/python:3.11 which comes with Python 3.11 pre-installed)
Any additional software or tools to install (e.g. poetry for Python dependency management)
Any settings or configurations needed for your project
Startup scripts or commands to run when the container starts, e.g. scripts to set up the environment or install dependencies

When you create a new codespace, GitHub uses this configuration file to set up the dev container automatically.

Summary

Containers are a powerful way to package and run applications, making development and deployment more reliable and consistent. They go beyond virtual environments by isolating the entire operating system environment, not just Python packages. Tools like Docker and platforms like GitHub Codespaces, Google Cloud Engine, and Replit make it easy to use containers in real projects.