[
https://issues.apache.org/jira/browse/MESOS-5401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15436204#comment-15436204
]
Sunzhe commented on MESOS-5401:
-------------------------------
Thank you for the reply.
The built-in containerizer is not the Docker container, right? I run through
Marathon like below:
{code}
...
"cmd" : "nvidia-smi"
"cpus" : 1,
"gpus" : 1,
"mem" : 1024,
"instances" : 1,
"container": {
"type" : "DOCKER",
"docker" : {
"image" : "nvidia/cuda"
}
}
...
{code}
it is error, "general: GPU resources only work with the Mesos containerizer"
> Add ability to inject a Volume of Nvidia libraries/binaries into a
> docker-image container in mesos containerizer.
> -----------------------------------------------------------------------------------------------------------------
>
> Key: MESOS-5401
> URL: https://issues.apache.org/jira/browse/MESOS-5401
> Project: Mesos
> Issue Type: Improvement
> Reporter: Kevin Klues
> Assignee: Kevin Klues
> Labels: gpu, mesosphere
> Fix For: 1.0.0
>
>
> In order to support Nvidia GPUs with docker containers in Mesos, we need to
> be able to consolidate all Nvidia libraries into a common volume and inject
> that volume into the container.
> This tracks the support in the mesos containerizer. The docker containerizer
> support will be tracked separately.
> More info on why this is necessary here:
> https://github.com/NVIDIA/nvidia-docker/
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
