View source for Choosing a node (constraints)

SLURM picks a node for your job based on the partition you choose and the constraints you set. If you don't tell SLURM what you need, it uses defaults; for anything that exceeds a default, you have to be explicit.

== Defaults ==

If you submit a job without any directives, SLURM assumes:

* '''Partition:''' main
* '''Quality of Service:''' std
* '''CPUs:''' 1
* '''Wall time:''' 1 hour
* '''Maximum wall time:''' 3 weeks
* '''Memory:''' 100 MB per node

For anything beyond these limits, set the relevant <code>#SBATCH</code> directive in your sbatch script.

== Selecting GPU hardware ==

The cluster has nodes with NVIDIA GPUs (partition <code>gpu</code>) and nodes with AMD GPUs (partition <code>gpu_amd</code>). To run on a GPU node you need to do two things in your sbatch script: select the right partition, and request the GPU as a generic resource.

=== NVIDIA GPUs ===

Six nodes with NVIDIA GPUs are available in the <code>gpu</code> partition. Request one with:

<syntaxhighlight lang="bash">
#SBATCH --partition=gpu
#SBATCH --gres=gpu:1
</syntaxhighlight>

Replace <code>1</code> with the number of GPUs you need. The <code>--gres</code> line is mandatory — without it, your job will either fail or run on the CPU instead of the GPU.

If you don't ask for a particular GPU model, SLURM picks whatever is free. The scheduler is configured to prefer A100s first, then A6000s, then V100s. The price per GPU-hour is the same regardless of model.

To restrict the job to a specific model, add a <code>--constraint</code> directive. To see the constraint names available for GPU nodes, run:

<syntaxhighlight lang="bash">
scontrol show -o node | grep -o -e "NodeName=\w*" -e "ActiveFeatures=[[:alnum:][:punct:]]*" | paste - - | column -t | grep gpu
</syntaxhighlight>

Then constrain the job to one of the listed models:

<syntaxhighlight lang="bash">
# Restrict to A100 GPUs
#SBATCH --constraint='nvidia&A100'
</syntaxhighlight>

A rough rule of thumb: the A100/80 GB cards are about twice as fast as the A6000/48 GB or V100/16 GB. Whether you can take advantage of that depends on whether your workload can actually use the extra memory and saturate the GPU.

=== AMD GPUs ===

Two nodes with AMD GPUs are available in the <code>gpu_amd</code> partition. The mechanics are identical to NVIDIA — request the partition and a GPU as a generic resource:

<syntaxhighlight lang="bash">
#SBATCH --partition=gpu_amd
#SBATCH --gres=gpu:1
</syntaxhighlight>

== See also ==

* [[Partitions / Queues]]
* [[Using Slurm | Scheduler overview]]
* [[Tariffs | Costs associated with resource usage]]