gres/gpu: _normalize_gres_conf: WARNING: The following autodetected GPUs are being ignored: gpu/nvml: _get_system_gpu_list_nvml: 4 GPU system device(s) detected error: _nvml_get_mem_freqs: Failed to get supported memory frequencies for the GPU : Not Supported Node reconfigured socket/core boundaries SocketsPerBoard=12:2(hw) CoresPerSocket=1:6(hw) With AutoDetect=nvml in nf and GresTypes=gpu in nf, this is what we see on a 4-GPU host after restarting slurmd: However, the version of Slurm in the standard debian repositories was apparently not compiled on a system with the necessary Nvidia library installed, so we recompiled Slurm 20.11 from the Debian source package with no modifications. We're running Debian 11, and the version of Slurm available for Debian 11 is 20.11.
We're trying to use Slurm's built-in Nvidia GPU detection mechanism to avoid having to specify GPUs explicitly in nf and nf.
0 kommentar(er)
