Press "Enter" to skip to content

15分钟连接Jetson Nano与K8S,轻松搭建机器学习集群

本站内容均来自兴趣收集,如不慎侵害的您的相关权益,请留言告知,我们将尽快删除.谢谢.

在本文中我将展示如何将Jetson Nano开发板连接到Kubernetes集群以作为一个GPU节点。我将介绍使用GPU运行容器所需的NVIDIA docker设置,以及将Jetson连接到Kubernetes集群。在成功将节点连接到集群后,我还将展示如何在Jetson Nano上使用GPU运行简单的TensorFlow 2训练会话。

 

K3s还是K8s?

 

K3s是一个轻量级Kubernetes发行版,其大小不超过100MB。在我看来,它是单板计算机的理想选择,因为它所需的资源明显减少。你可以查看我们的往期文章,了解更多关于K3s的教程和生态。在K3s生态中,有一款不得不提的开源工具K3sup,这是由Alex Ellis开发的,用于简化K3s集群安装。你可以访问Github了解这款工具:

 

https://github.com/alexellis/k3sup

 

我们需要准备什幺?

 

一个K3s集群——只需要一个正确配置的主节点即可

 

NVIDIA Jetson Nano开发板,并安装好开发者套件

 

如果你想了解如何在开发板上安装开发者套件,你可以查看以下文档:

 

https://developer.nvidia.com/embedded/learn/get-started-jetson-nano-devkit#write

 

K3sup

 

15分钟的时间

 

计划步骤

 

 

设置NVIDIA docker

 

添加Jetson Nano到K3s集群

 

运行一个简单的MNIST例子来展示Kubernetes pod内GPU的使用

 

 

设置NVIDIA docker

 

在我们配置Docker以使用nvidia-docker作为默认的运行时之前,我需要先解释一下为什幺要这样做。默认情况下,当用户在Jetson Nano上运行容器时,运行方式与其他硬件设备相同,你不能从容器中访问GPU,至少在没有黑客攻击的情况下不能。如果你想自己测试一下,你可以运行以下命令,应该会看到类似的结果:

 

[email protected]:~# echo "python3 -c 'import tensorflow'" | docker run -i icetekio/jetson-nano-tensorflow /bin/bash
2020-05-14 00:10:23.370761: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'libcudart.so.10.2'; dlerror: libcudart.so.10.2: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-10.2/targets/aarch64-linux/lib:
2020-05-14 00:10:23.370859: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2020-05-14 00:10:25.946896: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-10.2/targets/aarch64-linux/lib:
2020-05-14 00:10:25.947219: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-10.2/targets/aarch64-linux/lib:
2020-05-14 00:10:25.947273: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
/usr/lib/python3/dist-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters

 

如果你现在尝试运行相同的命令,但在docker命令中添 –runtime=nvidia
参数,你应该看到类似以下的内容:

 

[email protected]:~# echo "python3 -c 'import tensorflow'" | docker run --runtime=nvidia -i icetekio/jetson-nano-tensorflow /bin/bash
2020-05-14 00:12:16.767624: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.2
2020-05-14 00:12:19.386354: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libnvinfer.so.7
2020-05-14 00:12:19.388700: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libnvinfer_plugin.so.7
/usr/lib/python3/dist-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters

 

nvidia-docker已经配置完成,但是默认情况下并没有启用。要启用docker运行nvidia-docker运行时作为默认值,需要将 “default-runtime”:”nvidia”
添加到/etc/docker/daemon.json配置文件中,如下所示:

 

{
    "runtimes": {
        "nvidia": {
            "path": "nvidia-container-runtime",
            "runtimeArgs": []
        }
    },
    "default-runtime": "nvidia"
}

 

现在你可以跳过docker run命令中 –runtime=nvidia
参数,GPU将被默认初始化。这样K3s就会用nvidia-docker运行时来使用Docker,让Pod不需要任何特殊配置就能使用GPU。

 

将Jetson作为K8S节点连接

 

使用K3sup将Jetson作为Kubernetes节点连接只需要1个命令,然而要想成功连接Jetson和master节点,我们需要能够在没有密码的情况下同时连接到Jetson和master节点,并且在没有密码的情况下做sudo,或者以root用户的身份连接。

 

如果你需要生成SSH 密钥并复制它们,你需要运行以下命令:

 

ssh-keygen -t rsa -b 4096 -f ~/.ssh/rpi -P ""
ssh-copy-id -i .ssh/rpi [email protected]

 

默认情况下,Ubuntu安装要求用户在使用sudo命令时输入密码,因此,更简单的方法是用root账户来使用K3sup。要使这个方法有效,需要将你的 ~/.ssh/authorized_keys
复制到 /root/.ssh/
目录下。

 

在连接Jetson之前,我们查看一下想要连接到的集群:

 

[email protected]:~$ kubectl get node -o wide
NAME      STATUS   ROLES    AGE   VERSION        INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
nexus     Ready    master   32d   v1.17.2+k3s1   192.168.0.12   <none>        Ubuntu 18.04.4 LTS   4.15.0-96-generic   containerd://1.3.3-k3s1
rpi3-32   Ready    <none>   32d   v1.17.2+k3s1   192.168.0.30   <none>        Ubuntu 18.04.4 LTS   5.3.0-1022-raspi2   containerd://1.3.3-k3s1
rpi3-64   Ready    <none>   32d   v1.17.2+k3s1   192.168.0.32   <none>        Ubuntu 18.04.4 LTS   5.3.0-1022-raspi2   containerd://1.3.3-k3s1

 

你可能会注意到,master节点是一台IP为 192.168.0.12
的 nexus
主机,它正在运行containerd。默认状态下,k3s会将containerd作为运行时,但这是可以修改的。由于我们设置了nvidia-docker与docker一起运行,我们需要修改containerd。无需担心,将containerd修改为Docker我们仅需传递一个额外的参数到k3sup命令即可。所以,运行以下命令即可连接Jetson到集群:

 

k3sup join --ssh-key ~/.ssh/rpi  --server-ip 192.168.0.12  --ip 192.168.0.40   --k3s-extra-args '--docker'

 

IP 192.168.0.40
是我的Jetson Nano。正如你所看到的,我们传递了 –k3s-extra-args’–docker’
标志,在安装k3s agent 时,将 –docker
标志传递给它。多亏如此,我们使用的是nvidia-docker设置的docker,而不是containerd。

 

要检查节点是否正确连接,我们可以运行 kubectl get node -o wide

 

[email protected]:~$ kubectl get node -o wide
NAME      STATUS   ROLES    AGE   VERSION        INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
nexus     Ready    master   32d   v1.17.2+k3s1   192.168.0.12   <none>        Ubuntu 18.04.4 LTS   4.15.0-96-generic   containerd://1.3.3-k3s1
rpi3-32   Ready    <none>   32d   v1.17.2+k3s1   192.168.0.30   <none>        Ubuntu 18.04.4 LTS   5.3.0-1022-raspi2   containerd://1.3.3-k3s1
rpi3-64   Ready    <none>   32d   v1.17.2+k3s1   192.168.0.32   <none>        Ubuntu 18.04.4 LTS   5.3.0-1022-raspi2   containerd://1.3.3-k3s1
jetson    Ready    <none>   11s   v1.17.2+k3s1   192.168.0.40   <none>        Ubuntu 18.04.4 LTS   4.9.140-tegra       docker://19.3.6

 

简易验证

 

我们现在可以使用相同的docker镜像和命令来运行pod,以检查是否会有与本文开头在Jetson Nano上运行docker相同的结果。要做到这一点,我们可以应用这个pod规范:

 

apiVersion: v1
kind: Pod
metadata:
  name: gpu-test
spec:
  nodeSelector:
    kubernetes.io/hostname: jetson
  containers:
  - image: icetekio/jetson-nano-tensorflow
    name: gpu-test
    command:
    - "/bin/bash"
    - "-c"
    - "echo 'import tensorflow' | python3"
  restartPolicy: Never

 

等待docker镜像拉取,然后通过运行以下命令查看日志:

 

[email protected]:~$ kubectl logs gpu-test 
2020-05-14 10:01:51.341661: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.2
2020-05-14 10:01:53.996300: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libnvinfer.so.7
2020-05-14 10:01:53.998563: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libnvinfer_plugin.so.7
/usr/lib/python3/dist-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters

 

如你所见,我们的日志信息与之前在Jetson上运行Docker相似。

 

运行MNIST训练

 

我们有一个支持GPU的运行节点,所以现在我们可以测试出机器学习的 “Hello world”,并使用MNIST数据集运行TensorFlow 2模型示例。

 

要运行一个简单的训练会话,以证明GPU的使用情况,应用下面的manifest:

 

apiVersion: v1
kind: Pod
metadata:
  name: mnist-training
spec:
  nodeSelector:
    kubernetes.io/hostname: jetson
  initContainers:
    - name: git-clone
      image: iceci/utils
      command:
        - "git"
        - "clone"
        - "<https://github.com/IceCI/example-mnist-training.git>"
        - "/workspace"
      volumeMounts:
        - mountPath: /workspace
          name: workspace
  containers:
    - image: icetekio/jetson-nano-tensorflow
      name: mnist
      command:
        - "python3"
        - "/workspace/mnist.py"
      volumeMounts:
        - mountPath: /workspace
          name: workspace
  restartPolicy: Never
  volumes:
    - name: workspace
      emptyDir: {}

 

从下面的日志中可以看到,GPU正在运行:

 

...
2020-05-14 11:30:02.846289: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2020-05-14 11:30:02.846434: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.2
....

 

如果你在节点上,你可以通过运行tegrastats命令来测试CPU和GPU的使用情况:

 

[email protected]:~$ tegrastats --interval 5000
RAM 2462/3964MB (lfb 2x4MB) SWAP 362/1982MB (cached 6MB) CPU [52%@1479,41%@1479,43%@1479,34%@1479] EMC_FREQ 0% GR3D_FREQ 9% [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] POM_5V_IN 3410/3410 POM_5V_GPU 451/451 POM_5V_CPU 1355/1355
RAM 2462/3964MB (lfb 2x4MB) SWAP 362/1982MB (cached 6MB) CPU [53%@1479,42%@1479,45%@1479,35%@1479] EMC_FREQ 0% GR3D_FREQ 9% [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] POM_5V_IN 3410/3410 POM_5V_GPU 451/451 POM_5V_CPU 1353/1354
RAM 2461/3964MB (lfb 2x4MB) SWAP 362/1982MB (cached 6MB) CPU [52%@1479,38%@1479,43%@1479,33%@1479] EMC_FREQ 0% GR3D_FREQ 10% [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] POM_5V_IN 3410/3410 POM_5V_GPU 493/465 POM_5V_CPU 1314/1340

 

总  结

 

如你所见,将Jetson Nano连接到Kubernetes集群是一个非常简单的过程。只需几分钟,你就能利用Kubernetes来运行机器学习工作负载——同时也能利用NVIDIA袖珍GPU的强大功能。你将能够在Kubernetes上运行任何为Jetson Nano设计的GPU容器,这可以简化你的开发和测试。

 

作者:

 

Jakub Czapliński,Icetek编辑

 

原文链接:

 

https://medium.com/icetek/how-to-connect-jetson-nano-to-kubernetes-using-k3s-and-k3sup-c715cf2bf212

Be First to Comment

发表评论

电子邮件地址不会被公开。 必填项已用*标注