mirror of
https://github.com/XShengTech/MEGREZ-Deploy.git
synced 2026-04-23 16:38:14 +08:00
[Docs] 📝 Refactor the README files of the main program and the controlled end, adjust the links and content structure
This commit is contained in:
Harry-zklcdc
parent
84f1083d10
commit
f1cb3a6f61
227
README.md
227
README.md
@ -2,229 +2,6 @@
|
||||
|
||||
## 目录
|
||||
|
||||
* [主程序](#主程序)
|
||||
* [被控端](#被控端)
|
||||
* [主程序](megrez/README.md)
|
||||
* [被控端](controler/README.md)
|
||||
|
||||
## 主程序
|
||||
|
||||
> [!NOTE]
|
||||
> 即后端服务
|
||||
|
||||
### 1. 安装依赖
|
||||
|
||||
#### 1.1. 安装
|
||||
|
||||
```bash
|
||||
sudo apt update && sudo apt install git
|
||||
```
|
||||
|
||||
#### 1.2. 安装 docker
|
||||
|
||||
```bash
|
||||
sudo curl -sSL get.docker.com | sh
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
sudo curl -fsSL https://get.docker.com | bash -s docker --mirror Aliyun
|
||||
```
|
||||
|
||||
### 2. 下载部署配置
|
||||
|
||||
```bash
|
||||
git clone https://github.com/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
git clone https://openi.pcl.ac.cn/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
cd MEGREZ-Deploy/megrez
|
||||
```
|
||||
|
||||
### 3. 启动主程序
|
||||
|
||||
```bash
|
||||
docker compose up -d
|
||||
```
|
||||
|
||||
|
||||
## 被控端
|
||||
|
||||
> [!NOTE]
|
||||
> 即被控制的机器
|
||||
|
||||
### 1. 安装依赖
|
||||
|
||||
#### 1.1. 安装依赖
|
||||
|
||||
```bash
|
||||
sudo apt update && sudo apt install git lxcfs
|
||||
```
|
||||
|
||||
#### 1.2. 安装 docker
|
||||
|
||||
```bash
|
||||
sudo curl -sSL get.docker.com | sh
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
sudo curl -fsSL https://get.docker.com | bash -s docker --mirror Aliyun
|
||||
```
|
||||
|
||||
#### 1.3 安装 NVIDIA 驱动
|
||||
|
||||
[NVIDIA Linux Driver](https://www.nvidia.com/en-us/drivers/unix/)
|
||||
|
||||
#### 1.4 安装 NVIDIA Container Toolkit
|
||||
|
||||
```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
|
||||
```
|
||||
|
||||
```bash
|
||||
sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
|
||||
```
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk runtime configure --runtime=docker
|
||||
sudo systemctl restart docker
|
||||
```
|
||||
|
||||
### 2. 配置环境
|
||||
|
||||
#### 2.1 配置 XFS 分区
|
||||
|
||||
```bash
|
||||
sudo mkfs.xfs /dev/sdb # 例如 /dev/sdb
|
||||
```
|
||||
|
||||
修改 `/etc/fstab` 添加 XFS 分区
|
||||
|
||||
> [!IMPORTANT]
|
||||
> 一定要添加 `prjquota` 参数
|
||||
|
||||
```bash
|
||||
/dev/sdb /path/to/docker/data xfs defaults,prjquota 0 0
|
||||
```
|
||||
|
||||
挂载 XFS 分区
|
||||
|
||||
```bash
|
||||
sudo mount -a
|
||||
```
|
||||
|
||||
#### 2.2 配置 Docker 存储路径至 XFS 文件系统分区
|
||||
|
||||
修改 `/etc/docker/daemon.json` 添加 `data-root` 字段
|
||||
|
||||
```json
|
||||
{
|
||||
"data-root": "/path/to/docker/data" // 例如 "/data/docker"
|
||||
}
|
||||
```
|
||||
|
||||
#### 2.3 配置 Docker 镜像加速
|
||||
|
||||
修改 `/etc/docker/daemon.json` 添加 `registry-mirrors` 字段
|
||||
|
||||
```json
|
||||
{
|
||||
"registry-mirrors": ["https://docker.1panelproxy.com"]
|
||||
}
|
||||
```
|
||||
|
||||
#### 2.4 重启 Docker
|
||||
|
||||
```bash
|
||||
sudo systemctl restart docker
|
||||
```
|
||||
|
||||
查看 Docker 状态
|
||||
|
||||
```bash
|
||||
docker info | grep 'Docker Root Dir'
|
||||
```
|
||||
|
||||
返回 `Docker Root Dir: /path/to/docker/data`
|
||||
|
||||
```bash
|
||||
docker info
|
||||
```
|
||||
|
||||
返回 `Registry Mirrors: https://docker.1panelproxy.com` 即可
|
||||
|
||||
|
||||
#### 2.5 配置 CDI 设备
|
||||
|
||||
启用 Docker 的 CDI 特性
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk runtime configure --runtime=docker --cdi-enabled
|
||||
systemctl restart docker
|
||||
```
|
||||
|
||||
生成 CDI 设备配置
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk cdi generate --output=/var/run/cdi/nvidia.yaml
|
||||
```
|
||||
|
||||
查看 CDI 设备配置
|
||||
|
||||
```bash
|
||||
nvidia-ctk cdi list
|
||||
```
|
||||
|
||||
有如下返回即可
|
||||
|
||||
```bash
|
||||
INFO[0000] Found 17 CDI devices
|
||||
nvidia.com/gpu=0
|
||||
nvidia.com/gpu=1
|
||||
nvidia.com/gpu=2
|
||||
nvidia.com/gpu=3
|
||||
nvidia.com/gpu=4
|
||||
nvidia.com/gpu=5
|
||||
nvidia.com/gpu=6
|
||||
nvidia.com/gpu=7
|
||||
nvidia.com/gpu=GPU-23bb08b6-****-****-****-************
|
||||
nvidia.com/gpu=GPU-5f996fb2-****-****-****-************
|
||||
nvidia.com/gpu=GPU-a55f05cc-****-****-****-************
|
||||
nvidia.com/gpu=GPU-b3d3f52f-****-****-****-************
|
||||
nvidia.com/gpu=GPU-c8d9b1fb-****-****-****-************
|
||||
nvidia.com/gpu=GPU-c94df367-****-****-****-************
|
||||
nvidia.com/gpu=GPU-cddc6468-****-****-****-************
|
||||
nvidia.com/gpu=GPU-ea606b9e-****-****-****-************
|
||||
nvidia.com/gpu=all
|
||||
```
|
||||
|
||||
### 3. 下载部署配置
|
||||
|
||||
```bash
|
||||
git clone https://github.com/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
git clone https://openi.pcl.ac.cn/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
cd MEGREZ-Deploy/controler
|
||||
```
|
||||
|
||||
修改 `docker-compose.yml` 文件中的 `gpu-docker-api` 的 `environment` 字段的 `APIKEY` 为任意字符串
|
||||
|
||||
```yaml
|
||||
environment:
|
||||
- APIKEY=CHANGETHIS # 修改为任意字符串
|
||||
```
|
||||
修改 `docker-compose.yml` 文件中的 `gpu-docker-api` 的 `volumes` 字段的 `PATH_TO_DOCKER_STORAGE` 为 Docker 存储路径
|
||||
|
||||
```yaml
|
||||
volumes:
|
||||
- /var/run/docker.sock:/var/run/docker.sock
|
||||
- /path/to/docker/data:/path/to/docker/data
|
||||
```
|
||||
|
||||
|
||||
### 4. 启动被控端
|
||||
|
||||
```bash
|
||||
docker compose up -d
|
||||
```
|
||||
205
controler/README.md
Normal file
205
controler/README.md
Normal file
@ -0,0 +1,205 @@
|
||||
# 被控端
|
||||
|
||||
> [!NOTE]
|
||||
> 即被控制的机器
|
||||
|
||||
## 1. 安装依赖
|
||||
|
||||
### 1.1. 安装依赖
|
||||
|
||||
```bash
|
||||
sudo apt update && sudo apt install git lxcfs
|
||||
```
|
||||
|
||||
### 1.2. 安装 docker
|
||||
|
||||
```bash
|
||||
sudo curl -sSL get.docker.com | sh
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
sudo curl -fsSL https://get.docker.com | bash -s docker --mirror Aliyun
|
||||
```
|
||||
|
||||
### 1.3 安装 NVIDIA 驱动
|
||||
|
||||
[NVIDIA Linux Driver](https://www.nvidia.com/en-us/drivers/unix/)
|
||||
|
||||
### 1.4 安装 NVIDIA Container Toolkit
|
||||
|
||||
```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
|
||||
```
|
||||
|
||||
```bash
|
||||
sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
|
||||
```
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk runtime configure --runtime=docker
|
||||
sudo systemctl restart docker
|
||||
```
|
||||
|
||||
## 2. 配置环境
|
||||
|
||||
### 2.1 配置 XFS 分区
|
||||
|
||||
```bash
|
||||
sudo mkfs.xfs /dev/sdb # 例如 /dev/sdb
|
||||
```
|
||||
|
||||
修改 `/etc/fstab` 添加 XFS 分区
|
||||
|
||||
> [!IMPORTANT]
|
||||
> 一定要添加 `prjquota` 参数
|
||||
|
||||
```bash
|
||||
/dev/sdb /path/to/docker/data xfs defaults,prjquota 0 0
|
||||
```
|
||||
|
||||
挂载 XFS 分区
|
||||
|
||||
```bash
|
||||
sudo mount -a
|
||||
```
|
||||
|
||||
### 2.2 配置 Docker 存储路径至 XFS 文件系统分区
|
||||
|
||||
修改 `/etc/docker/daemon.json` 添加 `data-root` 字段
|
||||
|
||||
```json
|
||||
{
|
||||
"data-root": "/path/to/docker/data" // 例如 "/data/docker"
|
||||
}
|
||||
```
|
||||
|
||||
### 2.3 配置 Docker 镜像加速
|
||||
|
||||
修改 `/etc/docker/daemon.json` 添加 `registry-mirrors` 字段
|
||||
|
||||
```json
|
||||
{
|
||||
"registry-mirrors": ["https://docker.1panelproxy.com"]
|
||||
}
|
||||
```
|
||||
|
||||
### 2.4 重启 Docker
|
||||
|
||||
```bash
|
||||
sudo systemctl restart docker
|
||||
```
|
||||
|
||||
查看 Docker 状态
|
||||
|
||||
```bash
|
||||
docker info | grep 'Docker Root Dir'
|
||||
```
|
||||
|
||||
返回 `Docker Root Dir: /path/to/docker/data`
|
||||
|
||||
```bash
|
||||
docker info
|
||||
```
|
||||
|
||||
返回 `Registry Mirrors: https://docker.1panelproxy.com` 即可
|
||||
|
||||
|
||||
### 2.5 配置 CDI 设备
|
||||
|
||||
启用 Docker 的 CDI 特性
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk runtime configure --runtime=docker --cdi.enabled
|
||||
systemctl restart docker
|
||||
```
|
||||
|
||||
生成 CDI 设备配置
|
||||
|
||||
```bash
|
||||
sudo nvidia-ctk cdi generate --output=/var/run/cdi/nvidia.yaml
|
||||
```
|
||||
|
||||
查看 CDI 设备配置
|
||||
|
||||
```bash
|
||||
nvidia-ctk cdi list
|
||||
```
|
||||
|
||||
有如下返回即可
|
||||
|
||||
```bash
|
||||
INFO[0000] Found 17 CDI devices
|
||||
nvidia.com/gpu=0
|
||||
nvidia.com/gpu=1
|
||||
nvidia.com/gpu=2
|
||||
nvidia.com/gpu=3
|
||||
nvidia.com/gpu=4
|
||||
nvidia.com/gpu=5
|
||||
nvidia.com/gpu=6
|
||||
nvidia.com/gpu=7
|
||||
nvidia.com/gpu=GPU-23bb08b6-****-****-****-************
|
||||
nvidia.com/gpu=GPU-5f996fb2-****-****-****-************
|
||||
nvidia.com/gpu=GPU-a55f05cc-****-****-****-************
|
||||
nvidia.com/gpu=GPU-b3d3f52f-****-****-****-************
|
||||
nvidia.com/gpu=GPU-c8d9b1fb-****-****-****-************
|
||||
nvidia.com/gpu=GPU-c94df367-****-****-****-************
|
||||
nvidia.com/gpu=GPU-cddc6468-****-****-****-************
|
||||
nvidia.com/gpu=GPU-ea606b9e-****-****-****-************
|
||||
nvidia.com/gpu=all
|
||||
```
|
||||
|
||||
### 2.6 启动 LXCFS
|
||||
|
||||
```bash
|
||||
sudo systemctl start lxcfs
|
||||
```
|
||||
|
||||
## 3. 下载部署配置
|
||||
|
||||
```bash
|
||||
git clone https://github.com/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
git clone https://openi.pcl.ac.cn/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
cd MEGREZ-Deploy/controler
|
||||
```
|
||||
|
||||
修改 `docker-compose.yml` 文件中的 `gpu-docker-api` 的 `environment` 字段的 `APIKEY` 为任意字符串
|
||||
|
||||
```yaml
|
||||
environment:
|
||||
- APIKEY=CHANGETHIS # 修改为任意字符串
|
||||
```
|
||||
修改 `docker-compose.yml` 文件中的 `gpu-docker-api` 的 `volumes` 字段的 `PATH_TO_DOCKER_STORAGE` 为 Docker 存储路径
|
||||
|
||||
```yaml
|
||||
volumes:
|
||||
- /var/run/docker.sock:/var/run/docker.sock
|
||||
- /path/to/docker/data:/path/to/docker/data
|
||||
```
|
||||
|
||||
|
||||
## 4. 启动被控端
|
||||
|
||||
```bash
|
||||
docker compose up -d
|
||||
```
|
||||
|
||||
## 5. 拉取镜像
|
||||
|
||||
```bash
|
||||
docker pull ubuntu:22.04 # 这个必须拉取，否则数据盘扩容/缩容功能无法使用
|
||||
|
||||
# 下面的是打包好的镜像，可以直接使用
|
||||
docker pull xshengtech/megrez:pytorch-2.3.1_cuda-12.1_python-3.11_ubuntu-22.04
|
||||
docker pull xshengtech/megrez:pytorch-2.4.0_cuda-12.1_python-3.12_ubuntu-22.04
|
||||
docker pull xshengtech/megrez:pytorch-1.8.1_cuda-11.1_python-3.8_ubuntu-20.04
|
||||
docker pull xshengtech/megrez:pytorch-1.10.2_cuda-11.1_python-3.9_ubuntu-20.04
|
||||
docker pull xshengtech/megrez:pytorch-2.0.1_cuda-11.8_python-3.10_ubuntu-22.04
|
||||
docker pull xshengtech/megrez:pytorch-2.1.2_cuda-12.1_python-3.10_ubuntu-22.04
|
||||
docker pull xshengtech/megrez:pytorch-1.13.1_cuda-11.7_python-3.10_ubuntu-22.04
|
||||
docker pull xshengtech/megrez:ubuntu-22.04
|
||||
```
|
||||
38
megrez/README.md
Normal file
38
megrez/README.md
Normal file
@ -0,0 +1,38 @@
|
||||
# 主程序
|
||||
|
||||
> [!NOTE]
|
||||
> 即后端服务
|
||||
|
||||
## 1. 安装依赖
|
||||
|
||||
### 1.1. 安装
|
||||
|
||||
```bash
|
||||
sudo apt update && sudo apt install git
|
||||
```
|
||||
|
||||
### 1.2. 安装 docker
|
||||
|
||||
```bash
|
||||
sudo curl -sSL get.docker.com | sh
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
sudo curl -fsSL https://get.docker.com | bash -s docker --mirror Aliyun
|
||||
```
|
||||
|
||||
## 2. 下载部署配置
|
||||
|
||||
```bash
|
||||
git clone https://github.com/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
# 国内用户可以使用以下命令
|
||||
git clone https://openi.pcl.ac.cn/XShengTech/MEGREZ-Deploy.git
|
||||
|
||||
cd MEGREZ-Deploy/megrez
|
||||
```
|
||||
|
||||
## 3. 启动主程序
|
||||
|
||||
```bash
|
||||
docker compose up -d
|
||||
```
|
||||
Loading…
Reference in New Issue
Block a user