银河麒麟V10 Server系统中离线安装nvidia docker方法

银河麒麟V10 Server使用的是Centos7/8相同的数据包管理，因此可以使用yum和dnf获取安装包。分享一下所需的安装包：

链接：pan.baidu.com/s/1QboemtMj…
提取码：w73j

一. Docker安装

1. 卸载自带的docker

sudo dnf remove docker

2. cd到下载的docker rpm包目录

cd <docker19.03安装>

3. 依次安装

sudo rpm -Uvh container-selinux-2.107-1.el7_6.noarch.rpm --nodeps --force
sudo rpm -Uvh containerd.io-1.3.7-3.1.el7.x86_64.rpm --nodeps --force
sudo rpm -Uvh docker-ce-cli-19.03.13-3.el7.x86_64.rpm --nodeps --force
sudo rpm -Uvh docker-ce-19.03.13-3.el7.x86_64.rpm --nodeps --force

4. 启动docker服务

sudo systemctl enable docker
sudo systemctl start docker
sudo systemctl daemon-reload
sudo systemctl restart docker

5*. 【可选】测试 Docker 是否工作正常

sudo docker run --rm hello-world

二. NVIDIA Container Toolkit安装

1. cd到下载的nvidia docker rpm包目录

cd <nvidia docker安装>

2. 依次安装

sudo rpm -Uvh libnvidia-container-tools-1.10.0-1.x86_64.rpm --nodeps --force
sudo rpm -Uvh nvidia-container-toolkit-1.10.0-1.x86_64.rpm --nodeps --force
sudo rpm -Uvh libnvidia-container1-1.10.0-1.x86_64.rpm --nodeps --force
sudo rpm -Uvh nvidia-docker2-2.11.0-1.noarch.rpm --nodeps --force

3. 重启 Docker

sudo systemctl restart docker

4*. 【可选】运行基本 CUDA 容器测试是否正常运行

sudo docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi

如果看到类似以下内容，说明NVIDIA Container Toolkit安装完成，正常运行。

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 450.51.06    Driver Version: 450.51.06    CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| 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   34C    P8     9W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+