Mila documentation

This documentation contains most of the information required to run scientific and data-intensive computing at Mila and the available resources for its members. It is targeted towards Mila researchers and students.

Support

To reach the Mila system administrator support, send an email to: helpdesk@mila.quebec

Contribution

For any error in the documentation, missing or unclear sections, please open an issue or make a pull request on the github page.

Introduction

• What is a Cluster ?
  • Login Node
  • File system
• Ressources Available at Mila

Slurm

• Quick Start
• Basic Usage
  • Submitting jobs
  • Managing jobs
  • Cluster status
  • Useful Commands
  • Fairshare/Priority
• Advanced Usage
  • Handling preemption
  • Packing jobs

Clusters

• Mila Cluster
• Compute Canada Clusters

Software

• Modules
  • The module command
  • Available Software
  • Default package location
• Python
  • Pip/Virtualenv
  • Conda
• Tools
  • JupyterHub

Containers

• Docker/Shifter
  • Containers
  • Using in Slurm
  • Command line
  • Volumes
  • GPU
  • Remember
  • Example
• Singularity
  • Why should you use containers ?
  • Building containers
  • Using containers on clusters