Data parallelism on the clusters [online, CIP202]

Learn how to submit a large number of small jobs to the scheduler
About this Event

Managing a large number of short calculations can become a headache when you try to submit hundreds or thousands of jobs to the scheduler, which imposes limitations.
In this workshop, we present various ways to group short calculations in a smaller number of jobs that can be more conveniently submitted to the job scheduler.

Registration

Academic participant: $10 (This rate applies to anyone studying, teaching or working at a university or CEGEP)
Non-academic participant: $200 (This rate applies to anyone who does not fit the academic profile)

➔ If you are not sure which rate is right for you, please contact us by e-mail at [email protected]

Prerequisites

Being able to connect to a remote server with SSH
Having completed CIP201, or being able to submit multi-core jobs to the SLURM scheduler and monitor them

Course plan

Introduction to data parallelism
Job arrays
Parameter sweeps with GNU Parallel
Other tools - META-Farm and GLOST

Instructor

Olivier Fisette, analyst in advanced research computing at Calcul Québec.

Language

English

Technical prerequisites

We will use the Zoom platform. Because this event is a practical workshop, it is very useful having a secondary screen where you would get the instructor window on one screen and your own window on your main screen.

We will use the Jupyter Lab interface. Make sure you have a modern Web browser like Google Chrome, Firefox, Edge or Safari.

Notes

A certificate of participation will be send to each participant who attends at least 60% of the workshop.
The workshop is not recorded.
The workshop could be canceled if the number of registrations is too low.

Contact

For any question, please write to [email protected]