Due to the COVID-19 pandemic and the difficulties to lecture the courses in the given circumstances, we make video recordings available. Even if you have taken some of the lectures already, it may be useful to have a second look at some parts.
Training for starters
These videos are recordings from our training for starters series. — Last update: March 2021.
Supercomputers for Starters
- Introduction (13m48s)
- Processors in supercomputers (50m)
- The memory hierarchy (12m22s)
- Storing data on supercomputers (32m10s)
- Middleware: Turning the hardware into a useable supercomputer (11m21s)
- What can we expect? (23m57s)
- Accelerators (32m30s)
- Some conclusions (13m26s)
HPC@UAntwerp introduction
- Introduction (17m28s)
- Getting a VSC account (12m16s)
- Connecting to the cluster (22m20s)
- Preparing for your job (29m56s)
- Job system basics (54m13s)
- Job types 1: Interactive jobs (23m26s)
- Job types 2: Parallel batch jobs (19m19s)
- Job types 3: Organising multiple jobs (24m15s)
- Monitoring (16m44s)
- Best practices and user support (13m45s)
Transitioning to Vaughan and Slurm
This video is meant for users of the CalcUA systems and other VSC supercomputers with previous experience on clusters with Torque/Moab. It introduces the new storage system, our new AMD Rome-based cluster Vaughan and the new scheduling software Slurm. — Last update: June 2021.
- 0:00:00 Start
- 0:00:18 Overview of the presentation
- 0:02:06 The storage system
- 0:15:47 The new compute cluster
- 0:29:25 Logging in on the new cluster
- 0:31:13 The organisation of the module system
- 0:44:07 Slurm: Intro and job scripts
- 1:26:52 Slurm: Commands
- 1:50:24 Slurm examples: Shared and distributed memory jobs
- 2:00:26 Slurm examples: Job arrays and parameter exploration with atools
- 2:08:50 Slurm examples: Workflows
- 2:15:50 Slurm: Interactive jobs
- 2:27:07 File transfer with Globus
- 2:33:01 Wrap-up
There is also a version on YouTube.