Differences

This shows you the differences between two versions of the page.

--- habrok:advanced_job_management:running_jobs_on_gpus [2024/03/26 12:52] – [Available GPU types] admin
+++ habrok:advanced_job_management:running_jobs_on_gpus [2025/09/29 09:25] (current) – [Running interactive jobs] Formatting pedro
@@ Line 18: / Line 18: @@
 ==== Available GPU types ====
-^ Node ^ GPU type ^  GPUs per node ^  Memory per GPU ^  CPUs per node ^  Memory per node ^ Slurm name ^ Notes ^
+^ Node ^ GPU type ^  GPUs per node ^  Memory per GPU ^  CPUs per node ^  Memory per node ^ Slurm name ^
-| A100 | Nvidia A100  |  4 |  40 GB |  64 |  512 GB | a100 | A100 |
+| A100 | Nvidia A100  |  4 |  40 GB |  64 |  512 GB | a100 |
-| V100 | Nvidia V100  |  1 |  32 GB |  8|  128 GB | v100 | |
+| V100 | Nvidia V100  |  2 |  32 GB |  36 |  128 GB | v100 |
+| L40S | Nvidia L40S  |  2 |  48 GB |  56 |  512 GB | l40s |
 ==== Example ====
@@ Line 41: / Line 42: @@
 <code>
-gpu1.hpc.rug.nl
+gpu1.hb.hpc.rug.nl
-gpu2.hpc.rug.nl
+gpu2.hb.hpc.rug.nl
 </code>
-These machines have an NVIDIA V100 GPU each, which can be shared by multiple users. The tool ''nvidia-smi'' will show if the GPU is in use.
+These machines have an NVIDIA L40S GPU each, which can be shared by multiple users. The tool ''nvidia-smi'' will show if the GPU is in use.
 ** Please keep in mind that this is a shared machine, so allow everyone to make use of these GPUs and do not perform long runs here. Long runs should be submitted as jobs to scheduler. **
@@ Line 55: / Line 56: @@
 </code>
-There is currently an issue with using ''srun --gpus-per-node'', but there is a workaround by using '' --gres'' instead:
+There is currently an issue with using ''srun %%--%%gpus-per-node'', but there is a workaround by using '' %%--%%gres'' instead:
 <code>
 srun --gres=gpu:1 --time=01:00:00 --pty /bin/bash