...
| (number holds cores per node) | Location | Max. walltime | Nodes | Max nodes per job | Max jobs per user | Max memory per node | Shared | NPL per node hour | Remark |
|---|---|---|---|---|---|---|---|---|---|
| standard96 | Lise | 12:00:00 | 952 | 256 | (var) | 362 GB | ✘ | 14 | default partition |
| standard96:test | Lise | 1:00:00 | 32 dedicated +128 on demand | 16 | 1 | 362 GB | ✘ | 14 | test nodes with higher priority but lower walltime |
| large96 | Lise | 12:00:00 | 28 | 4 | (var) | 747 GB | ✘ | 21 | fat nodes |
| large96:test | Lise | 1:00:00 | 2 dedicated +2 on demand | 2 | 1 | 747 GB | ✘ | 21 | fat test nodes with higher priority but lower walltime |
| large96:shared | Lise | 48:00:00 | 2 dedicated | 1 | (var) | 747 GB | ✓ | 21 | fat nodes for data pre- and postprocessing |
| huge96 | Lise | 24:00:00 | 2 | 1 | (var) | 1522 GB | ✓ | 28 | very fat nodes for data pre- and postprocessing |
| medium40 | Emmy | 12:00:00 | 368 | 128 | unlimited | 362 GB | ✘ | 6 | default partition |
| medium40:test | Emmy | 1:00:00 | 16 dedicated +48 on demand | 8 | unlimited | 362 GB | ✘ | 6 | test nodes with higher priority but lower walltime |
| large40 | Emmy | 12:00:00 | 11 | 4 | unlimited | 747 GB | ✘ | 12 | fat nodes |
| large40:test | Emmy | 1:00:00 | 3 | 2 | unlimited | 747 GB | ✘ | 12 | fat test nodes with higher priority but lower walltime |
| large40:shared | Emmy | 24:00:00 | 2 | 1 | unlimited | 747 GB | ✓ | 12 | for data pre- and postprocessing |
| gpu | Emmy | 12:00:00 | 1 | 1 | unlimited | ✘ | equipped with 4 x NVIDIA Tesla V100 32GB |
...
The default partitions are suitable for most calculations. The :test partitions are, as the name suggests, intended for shorter and smaller test runs. These have a higher priotity and a few dedicated nodes, but are limited in time and number of nodes. The :shared
Shared nodes are mainly suitable for postprocessing. Nearly all post-processing. A job running on a shared node is only accounted for its core fraction (cores of job / all cores per node). All non shared nodes are exclusive to one job, except for the nodes in these :shared partitionswhich implies that full NPL are paid.
Parameters
| Parameter | SBATCH flag | Comment |
|---|---|---|
| # nodes | -N #<#> | |
| # tasks | -n #<#> | |
| # tasks per node | #SBATCH --tasks-per-node #<#> | Different defaults between mpirun and srun |
| partition | -p <name> | standard96 /medium40(Lise), medium40 (Emmy) |
# CPUs cores per task | -c #<#> | Default 1, interesting for OpenMP/Hybrid jobs |
| TimelimitWall time limit | -t hh:mm:ss | |
| --mail-type=ALL | See sbatch manpage for different types | |
| Project/Account | -A <project> | Specify project for NPL accounting |
Job Scripts
A job script can be any script that contains special instruction for Slurm. Most commonly used forms are shell scripts, such as bash or plain sh. But other scripting languages (e.g. Python, Perl, R) are also possible.
...