centos7.9 docker使用GPU 配置

centos7.9 docker使用GPU 配置

安装Docker

• 安装docker

# 安装Docker的必要依赖
yum install -y yum-utils device-mapper-persistent-data lvm2

# 添加Docker的YUM存储库
yum-config-manager --add-repo http://download.docker.com/linux/centos/docker-ce.repo
# 使用阿里云的存储库，如果需要
# yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo

# 安装Docker及其组件
yum install -y docker-ce docker-ce-cli

# 启动Docker并设置开机启动
systemctl start docker
systemctl enable docker

• 查看docker版本

docker --version

安装Docker-compose

• 安装Docker-compose

# 安装Docker Compose
sudo curl -L https://github.com/docker/compose/releases/download/1.21.2/docker-compose-$(uname -s)-$(uname -m) -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

• 查看docker-compose版本

docker-compose --version

安装GPU驱动

• 安装驱动环境

# 下载Anaconda安装脚本并执行安装
wget https://repo.continuum.io/archive/Anaconda3-2019.07-Linux-x86_64.sh
sudo sh Anaconda3-2019.07-Linux-x86_64.sh

# 查看系统中是否存在NVIDIA GPU
if lspci | grep -i nvidia &> /dev/null; then
    echo "NVIDIA GPU found"
else
    echo "No NVIDIA GPU found"
    exit 1
fi

# 安装内核开发工具
yum install -y kernel-devel

# 检查内核版本和源码版本
kernel_version=$(ls /boot | grep vmlinu)
source_version=$(rpm -qa | grep kernel-devel)
echo "Kernel Version: $kernel_version"
echo "Kernel Source Version: $source_version"

# 下载NVIDIA GPU驱动
wget http://cn.download.nvidia.com/tesla/450.51.06/NVIDIA-Linux-x86_64-450.51.06.run

# 阻止Nouveau并重新构建initramfs镜像
su root -c 'echo "blacklist nouveau" >> /lib/modprobe.d/dist-blacklist.conf'
su root -c 'echo "options nouveau modeset=0" >> /lib/modprobe.d/dist-blacklist.conf'
mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r).img.bak
dracut /boot/initramfs-$(uname -r).img $(uname -r)

# 修改运行级别为文本模式并重启
systemctl set-default multi-user.target
reboot

• 安装nvidia驱动

lsmod | grep nouveau

# 授予NVIDIA驱动安装程序可执行权限
chmod +x NVIDIA-Linux-x86_64-450.51.06.run

# 安装NVIDIA GPU驱动并自动确认安装选项 --kernel-source-path= 查看/usr/src/kernels/已存在的
./NVIDIA-Linux-x86_64-450.51.06.run --kernel-source-path=/usr/src/kernels/3.10.0-1160.95.1.el7.x86_64/ 

# 查看显卡信息
nvidia-smi

Docker 配置GPU

• Docker 配置GPU

# docker环境安装Nvidia-container-runtime 
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-container-runtime/$distribution/nvidia-container-runtime.repo | sudo tee /etc/yum.repos.d/nvidia-container-runtime.repo
yum install nvidia-container-runtime -y

# 修改daemon.json
echo '{
  "runtimes": {
    "nvidia": {
      "path": "nvidia-container-runtime",
      "runtimeArgs": []
    }
  },
  "default-runtime": "nvidia"
}' | sudo tee /etc/docker/daemon.json
# 重启docker并验证
systemctl restart docker
docker info

• docekr info 详情

# 此项表明成功配置GPU 
Runtimes: io.containerd.runc.v2 nvidia runc

docker-compose.yml 使用GPU

# docker-compose.yml 添加GPU相关环境信息
    environment:
      - NVIDIA_VISIBLE_DEVICES=all
      - NVIDIA_DRIVER_CAPABILITIES=compute,utility