docker支持GPU安装和配置

本实践将在 Ubuntu 22.04.3LTS 系统上进行安装测试
参考资料 : nvidia-container-toolkit
目前安装版本

nvidia-container-toolkit -version
NVIDIA Container Runtime Hook version 1.14.3
commit: 53b24618a542025b108239fe602e66e912b7d6e2

安装操作脚本

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list \
  && \
    sudo apt-get update

sudo apt-get install -y nvidia-container-toolkit

sudo nvidia-ctk runtime configure --runtime=docker

cat /etc/docker/daemon.json 
{
    "default-runtime": "nvidia",

    "bip": "192.168.88.1/24",
    "registry-mirrors": [
        "https://docker.nju.edu.cn"
    ],
    "runtimes": {
        "nvidia": {
            "args": [],
            "path": "nvidia-container-runtime"
        }
    }
}

sudo systemctl restart docker

sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi

vim /usr/local/bin/nvidia-docker

nvidia-docker 脚本

新版本nvidia-docker命令行已经移除
nvidia-docker参考 : nvidia-docker
nvidia-docker

#! /bin/bash
# Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved.

NV_DOCKER=${NV_DOCKER:-"docker"}

DOCKER_ARGS=()
NV_DOCKER_ARGS=()
while [ $# -gt 0 ]; do
    arg=$1
    shift
    DOCKER_ARGS+=("$arg")
    case $arg in
        run|create)
            NV_DOCKER_ARGS+=("--runtime=nvidia")
            if [ ! -z "${NV_GPU}" ]; then
                NV_DOCKER_ARGS+=(-e NVIDIA_VISIBLE_DEVICES="${NV_GPU// /,}")
            fi
            break
        ;;
        version)
            printf "NVIDIA Docker: @VERSION@\n"
            break
        ;;
        --)
            break
        ;;
    esac
done

if [ ! -z $NV_DEBUG ]; then
    set -x
fi

exec $NV_DOCKER "${DOCKER_ARGS[@]}" "${NV_DOCKER_ARGS[@]}" "$@"