Usage Guidelines AI Computing Centre

User Tools

Site Tools

Trace:
getting_started

Differences

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

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
getting_started [2025/03/27 13:09] – [Single GPU (1hr)] fboeltergetting_started [2025/08/06 10:48] (current) – [3.1. Queues] lkohlmeyer
Line 16: Line 16:
  
 <code bash> <code bash>
-enroot create -n pyxis_torch_2412 /cm/shared/workspace/nvidia_pytorch_files/nvidia-pytorch-24.12-py312.sqsh+enroot create -n pyxis_torch_2412 /sc/home/<username>/nvidia-pytorch-24.12-py312.sqsh
 </code> </code>
  
Line 24: Line 24:
 </code> </code>
 ===== 2. Environment Setup ===== ===== 2. Environment Setup =====
-<code bash> 
-module load slurm 
-</code> 
- 
 <code bash> <code bash>
 # Allocate one GPU to be able to start the Enroot container # Allocate one GPU to be able to start the Enroot container
-srun --nodes=1 --ntasks=1 --gpus=1 --time=01:00:00 --export=ALL --pty bash+srun --nodes=1 --ntasks=1 --gpus=1 --time=01:00:00 --partition=aisc --account=aisc --export=ALL --pty bash
  
 # Start Enroot Container mounting the current working directory # Start Enroot Container mounting the current working directory
Line 51: Line 47:
 ==== Single GPU (1hr) ==== ==== Single GPU (1hr) ====
 <code bash> <code bash>
-module load slurm +srun --nodes=1 --ntasks=1 --gpus=1 --time=01:00:00 --partition=aisc-interactive --account=aisc --export=ALL --pty bash
-srun --nodes=1 --ntasks=1 --gpus=1 --time=01:00:00 --export=ALL --pty bash+
 </code> </code>
  
  
-You can also see the available and Used GPUs by running:+You can also see the available GPUs by running:
 <code bash> <code bash>
 gpualloc gpualloc
 </code> </code>
 +
 +===== 3.1. Queues =====
 +**Important:**
 +
 +
 +**Available SLURM partitions (queues):**
 +
 +At the moment AISC users are allowed to use partition called
 +  * aisc-interactive: for interactive jobs only, limited to 8h max (highest job priority)
 +  * aisc: for batch jobs only, limited to 5 days max (medium job priority)
 +  * aisc-longrun: batch jobs only, limited to 14 days (lowest job priority)
 +
 +
 ===== 4. SLURM Batch Multi-Node Job Template ===== ===== 4. SLURM Batch Multi-Node Job Template =====
 Use this when you need to scale beyond a single node. SLURM handles node allocation and job scheduling. Use this when you need to scale beyond a single node. SLURM handles node allocation and job scheduling.
getting_started.1743080959.txt.gz · Last modified: 2025/03/27 13:09 by fboelter