Docker for Data Science
Containers have emerged in the Cloud and made it possible to utilize hardware infrastructure in the most flexible and efficient way. Containers make it possible to run whole application stacks from a single configuration file. The Institute of Data Science is already making use of this container infrastructure and we would like to share our knowledge.
First we will introduce Docker and how it is different from virtual machines with the help of examples. During the hands on sessions, you will learn how to use the “docker” command. We will also walk through troubleshooting by looking into log files and connecting into running containers to execute commands.
Finally the participants will create their own Docker images either from scratch or based on existing images. We will also look into building whole systems with docker-compose.
The target audience are UM staff who are dealing with IT infrastructure or any other Scientist with enthusiasm on Reproducible Science.
Programme
| Time | Topic |
|---|---|
| 10:00 - 10:15 | Welcome with coffee and tea, Introduction of the tutorial and organizers |
| 10:15 - 11:00 | Docker Environment: How it is different from virtual machines? |
| 11:00 - 12:00 | The “docker” command: Usage and troubleshooting |
| 12:00 - 13:00 | Lunch Break |
| 13:00 - 14:30 | Create your own Docker image |
| 14:30 - 15:00 | Overview of orchestration frameworks, applications and discussion |
Prerequisite knowledge or skills required
Basic Linux command line knowledge recommended (but not mandatory)
Maximum number of participants
15 - 20
Information
Relevant links
Go to the IDS website
Follow IDS on Facebook
Follow IDS on Twitter