Name: nvidia-container-runtime
Owner: NVIDIA Corporation
Description: NVIDIA container runtime
Created: 2017-09-05 22:03:28.0
Updated: 2018-04-02 10:44:37.0
Pushed: 2018-03-29 18:54:22.0
Homepage: null
Size: 159314
Language: Go
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
A modified version of runc adding a custom pre-start hook to all containers.
If environment variable NVIDIA_VISIBLE_DEVICES
is set in the OCI spec, the hook will configure GPU access for the container by leveraging nvidia-container-cli
from project libnvidia-container.
tup a rootfs based on Ubuntu 16.04
(mktemp -d) && mkdir rootfs
-sS http://cdimage.ubuntu.com/ubuntu-base/releases/16.04/release/ubuntu-base-16.04-core-amd64.tar.gz | tar --exclude 'dev/*' -C rootfs -xz
eate an OCI runtime spec
ia-container-runtime spec
-i 's;"sh";"nvidia-smi";' config.json
-i 's;\("TERM=xterm"\);\1, "NVIDIA_VISIBLE_DEVICES=0";' config.json
n the container
nvidia-container-runtime run nvidia_smi
nvidia-container-runtime
package: apt-get install nvidia-container-runtime
nvidia-container-runtime
package: yum install nvidia-container-runtime
To register the nvidia
runtime, use the method below that is best suited to your environment.
You might need to merge the new argument with your existing configuration.
mkdir -p /etc/systemd/system/docker.service.d
tee /etc/systemd/system/docker.service.d/override.conf <<EOF
vice]
Start=
Start=/usr/bin/dockerd --host=fd:// --add-runtime=nvidia=/usr/bin/nvidia-container-runtime
systemctl daemon-reload
systemctl restart docker
tee /etc/docker/daemon.json <<EOF
"runtimes": {
"nvidia": {
"path": "/usr/bin/nvidia-container-runtime",
"runtimeArgs": []
}
}
pkill -SIGHUP dockerd
You can optionally reconfigure the default runtime by adding the following to /etc/docker/daemon.json
:
ault-runtime": "nvidia"
dockerd --add-runtime=nvidia=/usr/bin/nvidia-container-runtime [...]
Each environment variable maps to an command-line argument for nvidia-container-cli
from libnvidia-container.
These variables are already set in our official CUDA images.
NVIDIA_VISIBLE_DEVICES
This variable controls which GPUs will be made accessible inside the container.
0,1,2
, GPU-fef8089b
?: a comma-separated list of GPU UUID(s) or index(es).all
: all GPUs will be accessible, this is the default value in our container images.none
: no GPU will be accessible, but driver capabilities will be enabled.void
or empty or unset: nvidia-container-runtime
will have the same behavior as runc
.NVIDIA_DRIVER_CAPABILITIES
This option controls which driver libraries/binaries will be mounted inside the container.
compute,video
, graphics,utility
?: a comma-separated list of driver features the container needs.all
: enable all available driver capabilities.utility
.compute
: required for CUDA and OpenCL applications.compat32
: required for running 32-bit applications.graphics
: required for running OpenGL and Vulkan applications.utility
: required for using nvidia-smi
and NVML.video
: required for using the Video Codec SDK.display
: required for leveraging X11 display.NVIDIA_REQUIRE_*
A logical expression to define constraints on the configurations supported by the container.
cuda
: constraint on the CUDA driver version.driver
: constraint on the driver version.arch
: constraint on the compute architectures of the selected GPUs.Multiple constraints can be expressed in a single environment variable: space-separated constraints are ORed, comma-separated constraints are ANDed.
Multiple environment variables of the form NVIDIA_REQUIRE_*
are ANDed together.
NVIDIA_DISABLE_REQUIRE
Single switch to disable all the constraints of the form NVIDIA_REQUIRE_*
.
NVIDIA_REQUIRE_CUDA
The version of the CUDA toolkit used by the container. It is an instance of the generic NVIDIA_REQUIRE_*
case and it is set by official CUDA images.
If the version of the NVIDIA driver is insufficient to run this version of CUDA, the container will not be started.
cuda>=7.5
, cuda>=8.0
, cuda>=9.0
?: any valid CUDA version in the form major.minor
.CUDA_VERSION
Similar to NVIDIA_REQUIRE_CUDA
, for legacy CUDA images.
In addition, if NVIDIA_REQUIRE_CUDA
is not set, NVIDIA_VISIBLE_DEVICES
and NVIDIA_DRIVER_CAPABILITIES
will default to all
.
This project is released under the BSD 3-clause license.