Introduction to Docker

🟡 LEVEL 2 | Prerequisites: Linux basics | Time: 4-6 hours total

Learning Path

This module teaches Docker through a structured progression designed for master students in mathematics and scientific computing:

📚 Concepts

What are Containers?
Problems Docker Solves
Docker Architecture

🛠️ Setup

Installation Guide
Essential Commands

📝 Building Images

Dockerfile Fundamentals
Hands-on: First Steps
Advanced Usage

🚀 Automation & Orchestration

Docker Compose
GitHub Actions + Docker

Quick Start

If you’re in a hurry, here’s the minimum path:

1. Install Docker       → 15 min
2. Basic commands       → 30 min
3. Dockerfile basics    → 45 min
4. GitHub Actions       → 30 min
                        --------
   Total:                 2 hours

Module Contents

Concepts (Why Containers?)

What are Containers?: Understand containers vs. VMs, images vs. containers, OCI standards.
Problems Docker Solves: The "matrix from hell", reproducibility challenges, and how Docker addresses them.
Docker Architecture: Client-server model, Docker daemon, registries, and how containers work.

Hands-On Skills

Installation Guide: Install Docker Desktop (Windows/macOS) or Docker Engine (Linux).
Essential Commands: docker run, docker build, docker exec, volumes, ports, and more.
Dockerfile Fundamentals: Create custom images for your mathematical computing projects.

Advanced Topics

Docker Compose: Orchestrate multi-container applications (e.g., Jupyter + database).
GitHub Actions + Docker: Automate image building and publishing with CI/CD.

Why Docker for Mathematics Students?

Challenge Docker Solution

Challenge	Docker Solution
"Works on my machine"	Package exact versions of Python, NumPy, MATLAB in a container
Complex dependencies	Dockerfile captures all setup steps reproducibly
Sharing research	`docker pull yourimage` gives collaborators your exact environment
HPC deployment	Build locally, run on clusters (via Apptainer/Singularity)
Long-term preservation	Archive container images with published papers

"Works on my machine"

Package exact versions of Python, NumPy, MATLAB in a container

Complex dependencies

Dockerfile captures all setup steps reproducibly

Sharing research

docker pull yourimage gives collaborators your exact environment

HPC deployment

Build locally, run on clusters (via Apptainer/Singularity)

Long-term preservation

Archive container images with published papers

Prerequisites Check

Before starting, ensure you have:

Basic Linux command-line skills (cd, ls, mkdir)
A computer with 8GB+ RAM and 20GB+ free disk space
Administrator/sudo access for installation
A GitHub account (for the Actions tutorial)

Ready? Start with What are Containers? →