Introduction
EUMETSAT infrastructure contains RX A6000 NVIDIA GPU cards. To employ the GPU, one need to provision one of the following flavors:
| Flavor name | vCPU | RAM | vGPU Type | vGPU RAM | SSD storage (GB) |
|---|
| vm.a6000.1 | 2 | 14 GB | RTXA6000-6C | 6 GB | 40 |
| vm.a6000.2 | 4 | 28 GB | RTXA6000-12C | 12 GB | 80 |
| vm.a6000.4 | 8 | 56 GB | RTXA6000-24C | 24 GB | 160 |
| vm.a6000.8 | 16 | 112 GB | RTXA6000-48C | 48 GB | 320 |
Provision
To use the GPUs:
- Provision new Centos or Ubuntu instance.
- Select layout ending with
eumetsat-gpu and one of the plans listed above. Beside that, configure your instance as preferred and continue deployment process.
- Once VM is deployed, you can verify GPUs for example using
nvidia-smi program from command line (see below for confirming library installations and drivers).
Usage
Essential commands
You can see GPU information using nvidia-smi
# Login to your instance and run below command
$ nvidia-smi
# Check if the input you received shows the NVIDIA-SMI, Driver and CUDA versions. You can also see the GPU hardware (e.g., RTXA6000-6C) and the GPU memory
Mon Feb 5 13:01:43 2024
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.223.02 Driver Version: 470.223.02 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| 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 NVIDIA RTXA6000-6C On | 00000000:00:05.0 Off | 0 |
| N/A N/A P8 N/A / N/A | 512MiB / 5976MiB | 0% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+ |
# NVIDIA tools are available in /usr/local/cuda-11.8/bin/. You can add them to PATH following:
$ export PATH=$PATH:/usr/local/cuda-11.8/bin/ |
Installing Libraries
You can install a variety of libraries using different methods. Below, we have a basic tutorial showing you how you can install libraries such as TensorFlow, Keras and PyTorch with conda package manger. CUDA version is currently 11.4 which need to be the same with drivers and thus can't be changed. Tensorflow library compatibility is available at: https://www.tensorflow.org/install/source#gpu.
Using conda
# install miniforge (or any conda manager)
$ wget https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh
# make it executable
$ chmod +x Miniforge3-Linux-x86_64.sh
# run and install the executable
$ ./Miniforge3-Linux-x86_64.sh |
# create a conda environment called ML with Python 3.8
$ conda create -n ML python=3.8
# activate the environment
$ conda activate ML
# install packages, note that installing tensorflow-gpu and keras also installs many number of extra libraries such as CUDA toolkit, cuDNN (CUDA Deep Neural Network library), Numpy, Scipy, Pillow
(ML) $ conda install tensorflow-gpu keras
# (OPTIONAL) cudatoolkit is installed automatically while installing keras and tensorflow-gpu, but if you need a specific (or latest) version run below command.
(ML) $ conda install -c anaconda cudatoolkit
# (OPTIONAL) installing pytorch GPU, pytorch might need cuda 11.8
(ML) $ conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia |
Installation Confirmations
Here we run a few initial commands for different libraries & drivers for confirming the library integrations with the GPU.
(ML) $ python3 --version
Python 3.8.18 |
(ML) $ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Wed_Sep_21_10:33:58_PDT_2022
Cuda compilation tools, release 11.8, V11.8.89
Build cuda_11.8.r11.8/compiler.31833905_0 |
(ML) $ conda list | grep tensorflow
tensorflow 2.13.1 cuda118py38h409af0c_1 conda-forge
tensorflow-base 2.13.1 cuda118py38h52ca5c6_1 conda-forge
tensorflow-estimator 2.13.1 cuda118py38ha2f8a09_1 conda-forge
tensorflow-gpu 2.13.1 cuda118py38h0240f8b_1 conda-forge |
(ML) $ conda list | grep keras
keras 2.13.1 pyhd8ed1ab_0 conda-forge |
import tensorflow as tf
tf.test.is_built_with_cuda()
True
tf.config.list_physical_devices('GPU')
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
print(tf.__version__)
2.13.1
|
import torch
print(torch.__version__)
2.2.0
print(torch.cuda.is_available())
True
print(torch.version.cuda)
11.8
if torch.cuda.is_available():
# Create a tensor and move it to GPU
x = torch.tensor([1.0, 2.0]).cuda()
print(x) # Print the tensor to verify it's on the GPU
else:
print("CUDA is not available. Check your PyTorch installation.")
tensor([1., 2.], device='cuda:0')
|
Using Docker
If you want to use GPUs in docker, you need to take few extra steps after creating the VM.
$ sudo apt install -y docker.io
$ sudo usermod -aG docker $USER |
$ sudo yum-config-manager \
--add-repo \
https://download.docker.com/linux/centos/docker-ce.repo
$ sudo yum install docker-ce docker-ce-cli containerd.io
$ sudo systemctl --now enable docker
$ sudo usermod -aG docker $USER |
To provide support for docker to use the GPU, you need to install the NVIDIA Container Toolkit. You can follow instructions on NVIDIA's website or basically do:
$ distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
$ curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
$ curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
$ sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
$ sudo systemctl restart docker |
$ distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \
&& curl -s -L https://nvidia.github.io/libnvidia-container/$distribution/libnvidia-container.repo | sudo tee /etc/yum.repos.d/nvidia-container-toolkit.repo
$ sudo yum clean expire-cache && sudo yum install -y nvidia-docker2
$ sudo systemctl restart docker |
Test the install with:
$ docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi
Wed Feb 28 13:20:24 2024
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.223.02 Driver Version: 470.223.02 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| 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 NVIDIA RTXA6000-6C On | 00000000:00:05.0 Off | 0 |
| N/A N/A P8 N/A / N/A | 512MiB / 5976MiB | 0% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+
|
And run something useful..
$ sudo docker run --gpus all --env NVIDIA_DISABLE_REQUIRE=1 -it --rm -v $(realpath ~/notebooks):/tf/notebooks -p 8888:8888 tensorflow/tensorflow:latest-gpu-jupyter |