It’s time to look at GPUs inside Docker container..
Docker is the leading container platform which provides both hardware and software encapsulation by allowing multiple containers to run on the same system at the same time each with their own set of resources (CPU, memory, etc) and their own dedicated set of dependencies (library version, environment variables, etc.). Docker can now be used to containerize GPU-accelerated applications. In case you’re new to GPU-accelerated computing, it is basically the use of graphics processing unit to accelerates high performance computing workloads and applications. This means you can easily containerize and isolate accelerated application without any modifications and deploy it on any supported GPU-enabled infrastructure.
Docker does not natively support NVIDIA GPUs within containers. Though there are available workaround like fully installing the NVIDIA drivers inside the container and map in the character devices corresponding to the NVIDIA GPUs (e.g.
/dev/nvidia0) on launch but still it is not recommended.
Here comes nvidia-docker plugin for a rescue…
The nvidia-docker is an open source project hosted on GITHUB and it provides driver-agnostic CUDA images & docker command line wrapper that mounts the user mode components of the driver and the GPUs (character devices) into the container at launch. With this enablement, the NVIDIA Docker plugin enables deployment of GPU-accelerated applications across any Linux GPU server with NVIDIA Docker support. What does this mean? – Using Docker, we can develop and prototype GPU applications on a workstation, and then ship and run those applications anywhere that supports GPU containers. Earlier this year, the nvidia-docker 1.0.1 release announced the support for Docker 17.03 Community & Enterprise Edition both.
Some of the key notable benefits includes –
- Legacy accelerated compute apps can be containerized and deployed on newer systems, on premise, or in the cloud.
- Ease of Deployment
- Isolation of Resource
- Bare Metal Performance
- Facilitate Collaboration
- Run access heterogeneous CUDA toolkit environments (sharing the host driver)
- Specific GPU resources can be allocated to container for better isolation and performance.
- You can easily share, collaborate, and test applications across different environments.
- Portable and reproducible builds
Let’s talk about libnvidia-container a bit..
libnvidia is NVIDIA container runtime library. The repository provides a library and a simple CLI utility to automatically configure GNU/Linux containers leveraging NVIDIA hardware.The implementation relies on kernel primitives and is designed to be agnostic of the container runtime. Basic features includes –
- Integrates with the container internals
- Agnostic of the container runtime
- Drop-in GPU support for runtime developers
- Better stability, follows driver releases
- Brings features seamlessly (Graphics, Display, Exclusive mode, VM, etc.)
~ source: NVIDIA
Under this blog post, I will show you how to get started with nvidia-docker to interact with NVIDIA GPU system and then look at few of interesting applications which can be build for GPU-accelerated data center. Let us get started –
Docker Version: 17.06
OS: Ubuntu 16.04 LTS
Environment : Manager Server Instance with GPU
GPU: GeForce GTX 1080 Graphics card
- Verify that GPU card is equipped in your hardware:
- Install nvidia-docker & nvidia-docker-plugin under Ubuntu 16.04 using wget as shown below:
Initializing nvidia-docker service:
Whenever nvidia-docker is installed, it creates a Docker volume and mounts the devices into a docker container automatically.
Did you know?
It is possible to avoid replying on nvidia-wrapper to launch GPU containers using ONLY docker and that can be done by using the REST API directly as shown below:
NVIDIA’s System Management Interface
If you want to know the status of your NVIDIA GPU, then nvidia-smi is the handy command which can be run using nvidia-cuda container. This is generally useful when you’re having trouble getting your NVIDIA GPUs to run GPGPU code.
Listing all NVIDIA Devices:
Listing all available data on the particular GPU:
Listing details for each GPU:
Listing the available clock speeds:
Building & Testing NVIDIA-Docker Images
If you look at samples/ folder under the nvidia-docker repository , there are couple of images that can be used to quickly test
nvidia-docker on your machine. Unfortunately, the samples are not available on the Docker Hub, hence you will need to build the images locally. I have built few of them which I am going to showcase:
Running the DeviceQuery container
You can leverage ajeetraina/nvidia-devicequery container directly as shown below:
Listing the current GPU clock speed, default clock speed & maximum possible clock speed:
Retrieving the System Topology:
The topology refers to how the PCI-Express devices (GPUs, InfiniBand HCAs, storage controllers, etc.) connect to each other and to the system’s CPUs. This can be retrieved as follow:
A Quick Look at NVIDIA Deep Learning..
The NVIDIA Deep Learning GPU Training System, a.k.a DIGITS is a webapp for training deep learning models. It puts the power of deep learning into the hands of engineers & data scientists. It can be used to rapidly train the highly accurate deep neural network (DNNs) for image classification, segmentation and object detection tasks.The currently supported frameworks are: Caffe, Torch, and Tensorflow.
DIGITS simplifies common deep learning tasks such as managing data, designing and training neural networks on multi-GPU systems, monitoring performance in real time with advanced visualizations, and selecting the best performing model from the results browser for deployment. DIGITS is completely interactive so that data scientists can focus on designing and training networks rather than programming and debugging.
To test-drive DIGITS, you can get it up and running in a single Docker container:
In the above command, NV_GPU is a method of assigning GPU resources to a container which is critical for leveraging DOCKER in a Multi GPU System. This passes GPU ID 0 from the host system to the container as resources. Note that if you passed GPU ID 2,3 for example, the container would still see the GPUs as ID 0,1 inside the container, with the PCI ID of 2,3 from the host system. As I have a single GPU card, I just passed it as NV_GPU=0.
You can open up web browser and verify if its running on the below address:
The below is the snippet from my w3m text browser:
How about Docker Compose? Is it supported?
Yes, of course.
Let us see how Docker compose works for nvidia-docker.
- First we need to figure out the nvidia driver version
As shown above, the nvidia driver version displays 375.66.
- Create a docker volume that uses the nvidia-docker plugin.
Verify it with the below command:
Now let us look at docker-compose YAML file shown below:
If you have ever worked with docker-compose, you can easily understand what each line specifies. I specified /dev/nvidia0 as I had a single GPU card, capture the correct volume driver name which we specified in the last step.
Just initiate the docker-compose as shown below:
This will start a container which ran nvidia-smi and then exited immediately. To keep it running, one can add tty: true inside Docker-compose file.
Let us see another interesting topic…TensorFlow
I have a sample TensorFlow based docker-compose file as shown below:
Verify if the container is up and running –
In the next blog post, I will showcase how to push the GPU metrics to prometheus & cAdvisor.
Did you find this blog helpful? Feel free to share your experience. Get in touch @ajeetsraina.
If you are looking out for contribution/discussion, join me at Docker Community Slack Channel