This session teaches participants how to use Compute Canada’s queuing environment on the national systems, using the job scheduler Slurm. Learn how the scheduler works, how it allocates jobs, what are reasonable requests to minimize wait time, how to make the best use of the resources to be more efficient, how to get more throughput, how to get more jobs running at the same time, and how to troubleshoot and deal with crashes. This workshop is designed for either new HPC users familiar with Linux and Shell Scripting, but who have not had experience with using Slurm, or, for experienced users transitioning to Slurm or seeking to improve efficiency with the scheduler. In order to get the most from the session, participants are strongly encouraged to have a Compute Canada/ACENET account and to bring a laptop to do the exercises.
This session will be delivered online.
To get the most from ACENET basics, please register for a Compute Canada Account. To register contact your supervising professor, ask for their CCRI, then visit https://ccdb.computecanada.ca/account_application. If your professor is not registered with Compute Canada, please have them register, then follow up with you. In addition to a Compute Canada account, you will want a computer with Windows, MacOS X, or a Unix-based operating system (not a ChromeBook), and a stable internet connection. A registered account is not mandatory, just recommended to get the most out of our lessons. You can attend the first session to see how ACENET can assist in accelerating your computational research before you decide to obtain an account, if you wish.
Connections details for the session as well as further preparation information will be sent out prior to the session.