If it’s possible to see other GPUs within a job then that means that cgroups aren’t being used.
Look at the cgroup documentation of slurm (https://slurm.schedmd.com/cgroup.conf.html) With cgroups activated an `nvidia-smi` will only show the GPU allocated to the job. -greg From: slurm-users <slurm-users-boun...@lists.schedmd.com> on behalf of taleinterve...@sjtu.edu.cn <taleinterve...@sjtu.edu.cn> Date: Wednesday, 23 March 2022 at 5:50 pm To: slurm-users@lists.schedmd.com <slurm-users@lists.schedmd.com> Subject: [EXTERNAL] [slurm-users] how to locate the problem when slurm failed to restrict gpu usage of user jobs Hi, all: We found a problem that slurm job with argument such as --gres gpu:1 didn’t be restricted with gpu usage, user still can see all gpu card on allocated nodes. Our gpu node has 4 cards with their gres.conf to be: > cat /etc/slurm/gres.conf Name=gpu Type=NVlink_A100_40GB File=/dev/nvidia0 CPUs=0-15 Name=gpu Type=NVlink_A100_40GB File=/dev/nvidia1 CPUs=16-31 Name=gpu Type=NVlink_A100_40GB File=/dev/nvidia2 CPUs=32-47 Name=gpu Type=NVlink_A100_40GB File=/dev/nvidia3 CPUs=48-63 And for test, we submit simple job batch like: #!/bin/bash #SBATCH --job-name=test #SBATCH --partition=a100 #SBATCH --nodes=1 #SBATCH --ntasks=6 #SBATCH --gres=gpu:1 #SBATCH --reservation="gpu test" hostname nvidia-smi echo end Then in the out file the nvidia-smi showed all 4 gpu cards. But we expect to see only 1 allocated gpu card. Official document of slurm said it will set CUDA_VISIBLE_DEVICES env var to restrict the gpu card available to user. But we didn’t find such variable exists in job environment. We only confirmed it do exist in prolog script environment by adding debug command “echo $CUDA_VISIBLE_DEVICES” to slurm prolog script. So how do slurm co-operate with nvidia tools to make job user only see its allocated gpu card? What is the requirement on nvidia gpu drivers, CUDA toolkit or any other part to help slurm correctly restrict the gpu usage?
