Instructor notes
Last updated on 2026-10-28 | Edit this page
Timing
The hpc-intro episodes keep their upstream times. The last two episodes are from the SURF fundamentals deck.
| Episode | Teaching | Exercises |
|---|---|---|
| Why use a Cluster? | 15 | 5 |
| Connecting to the remote cluster | 35 | 0 |
| Exploring the cluster | 25 | 10 |
| Scheduling jobs | 45 | 30 |
| Environment variables | 10 | 5 |
| Software modules | 30 | 15 |
| Transferring files | 15 | 15 |
| Running a parallel job | 30 | 60 |
| Job arrays | 15 | 5 |
| Using resources effectively | 10 | 20 |
| Your responsibilities as an HPC user | 15 | 5 |
| Accounting on Snellius | 20 | 10 |
| Python and OpenMP on a node | 10 | 25 |
If the room is short on time, keep the 32 SBU example and one
pi = 3.141592654 line. The genoa cost can go. On a two-core
node, submit sbatch --cpus-per-task=2 pi.sh and do not edit
the script.
Source
Accounting numbers are the system-details table in
course-module-fundamentals-of-hpc/slides/fundamentals-of-hpc.org.
The Python and OpenMP scripts are
course-module-fundamentals-of-hpc/hands-on/.
Branch: fix/slides-fundamentals-source
Commit: 2ebb95294418d5831143e393e93b294af39464b1
The course-cluster note is the slide “This course runs on a course
cluster built like Snellius”. The course cluster is a SURF Research
Cloud workspace. The course tab requests --partition=rome.
The published scripts keep --partition=rome.
Upstream
The rest of the lesson is hpc-intro at
7081080674aceeb0a4bc57b9348da22d876af802, CC-BY 4.0,
Carpentries authors in AUTHORS. See
UPSTREAM.md.