在容器环境中使用物理机的Nvidia GPU，宿主机自身需要安装好显卡驱动，还要结合使用官方提供的Toolkit，创建新的容器运行时，并在启动容器时使用对应的运行时环境。以下是在Ubuntu 22.04系统中，docker容器使用物理机显卡的简单配置过程.
Bash
nvidia-smi -L参考：https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html
- 配置APT仓库
Bash
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- 安装软件包
Bash
sudo apt-get install -y nvidia-container-toolkit- docker容器运行时配置
Bash
sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker- 测试
Bash
docker run -it --rm --gpus=all ubuntu:22.04 nvidia-smi -L附：docker配置文件(/etc/docker/daemon.json)参考
执行nvidia-ctk runtime configure –runtime=docker时会往配置中文件中写入相应的runtime配置
JSON
{
"default-ulimits": {
"memlock": {
"Hard": -1,
"Name": "memlock",
"Soft": -1
}
},
"exec-opts": [
"native.cgroupdriver=cgroupfs"
],
"runtimes": {
"nvidia": {
"path": "nvidia-container-runtime",
"runtimeArgs": []
}
},
"bridge": "none",
"iptables": false,
"live-restore": true
}
发表回复