Recently I started to use Docker and Github together to build and deploy analysis projects with a single command line. To do this, I setup a github repository with my Dockerfile (build instructions) and upload data and scripts needed for my analysis.
Whenever I want to perform the analysis, I simply run a single command on my Windows workstation. It builds my container, runs it, passes parameters, and exits when the analysis is done.
This allows me to re-use my containers and easily create new analysis experiments simply by creating a new repository.
In this livestream I will walk through my use cases and how I use docker step-by-step.