Загрузка графического процессора при 0 — Nvidia Tesla T4, CUDA 11.5, Ubuntu 20.04

#pytorch #gpu

#пыторч #графический процессор

Вопрос:

графический процессор не улавливается потоком демона pytorch. Вот вывод команд Nvidia:

 ==============NVSMI LOG==============  Timestamp : Mon Dec 6 03:21:37 2021 Driver Version : 495.29.05 CUDA Version : 11.5  Attached GPUs : 1 GPU 00000000:00:1E.0  Product Name : Tesla T4   nvidia-smi  Mon Dec 6 03:22:09 2021  -----------------------------------------------------------------------------  | NVIDIA-SMI 495.29.05 Driver Version: 495.29.05 CUDA Version: 11.5 | |------------------------------- ---------------------- ----------------------  | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=============================== ====================== ======================| | 0 Tesla T4 On | 00000000:00:1E.0 Off | 0 | | N/A 38C P0 31W / 70W | 2432MiB / 15109MiB | 0% Default | | | | N/A |  ------------------------------- ---------------------- ----------------------    -----------------------------------------------------------------------------  | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 2213 C python 2429MiB |  -----------------------------------------------------------------------------    sudo systemctl status nvidia-persistenced.service ● nvidia-persistenced.service - NVIDIA Persistence Daemon  Loaded: loaded (/lib/systemd/system/nvidia-persistenced.service; enabled; vendor preset: enabled)  Active: active (running) since Sun 2021-12-05 16:06:44 UTC; 11h ago  Main PID: 551 (nvidia-persiste)  Tasks: 1 (limit: 18834)  Memory: 864.0K  CGroup: /system.slice/nvidia-persistenced.service  └─551 /usr/bin/nvidia-persistenced --verbose  Dec 05 16:06:43 ip-172-31-11-249 nvidia-persistenced[551]: Verbose syslog connection opened Dec 05 16:06:43 ip-172-31-11-249 systemd[1]: Starting NVIDIA Persistence Daemon... Dec 05 16:06:43 ip-172-31-11-249 nvidia-persistenced[551]: Started (551) Dec 05 16:06:43 ip-172-31-11-249 nvidia-persistenced[551]: device 0000:00:1e.0 - registered Dec 05 16:06:44 ip-172-31-11-249 nvidia-persistenced[551]: device 0000:00:1e.0 - persistence mode enabled. Dec 05 16:06:44 ip-172-31-11-249 nvidia-persistenced[551]: device 0000:00:1e.0 - NUMA memory onlined. Dec 05 16:06:44 ip-172-31-11-249 nvidia-persistenced[551]: Local RPC services initialized Dec 05 16:06:44 ip-172-31-11-249 systemd[1]: Started NVIDIA Persistence Daemon.   

Комментарии:

1. Пожалуйста, предоставьте достаточно кода, чтобы другие могли лучше понять или воспроизвести проблему.