How to run hello ai world container on a kubernetes cluster

freewalker96 · March 14, 2021, 3:10am

I have followed the steps of hello AI world and ran the pre-built docker container in 1 Jetson nano. It is working on a single Jetson nano but I wanted to run the docker container on a Kubernetes cluster of 3 Jetson nano. I am not sure how to do it by writing an appropriate yaml file. Please advise me on how to proceed.

This is the link that I used:

github.com

dusty-nv/jetson-inference/blob/master/docs/aux-docker.md

<img src="https://github.com/dusty-nv/jetson-inference/raw/master/docs/images/deep-vision-header.jpg" width="100%">
<p align="right"><sup><a href="jetpack-setup-2.md">Back</a> | <a href="building-repo-2.md">Next</a> | </sup><a href="../README.md#hello-ai-world"><sup>Contents</sup></a>
<br/>
<sup>System Setup</sup></p>  

# Running the Docker Container

Pre-built Docker container images for this project are hosted on [DockerHub](https://hub.docker.com/r/dustynv/jetson-inference/tags).  Alternatively, you can [Build the Project from Source](building-repo-2.md).   

Below are the currently available container tags:

| Container Tag                                                                           | L4T version |          JetPack version         |
|-----------------------------------------------------------------------------------------|:-----------:|:--------------------------------:|
| [`dustynv/jetson-inference:r32.7.1`](https://hub.docker.com/r/dustynv/jetson-inference/tags) | L4T R32.7.1 | JetPack 4.6.1 |
| [`dustynv/jetson-inference:r32.6.1`](https://hub.docker.com/r/dustynv/jetson-inference/tags) | L4T R32.6.1 | JetPack 4.6 |
| [`dustynv/jetson-inference:r32.5.0`](https://hub.docker.com/r/dustynv/jetson-inference/tags) | L4T R32.5.0 | JetPack 4.5 |
| [`dustynv/jetson-inference:r32.4.4`](https://hub.docker.com/r/dustynv/jetson-inference/tags) | L4T R32.4.4 | JetPack 4.4.1 |
| [`dustynv/jetson-inference:r32.4.3`](https://hub.docker.com/r/dustynv/jetson-inference/tags) | L4T R32.4.3 | JetPack 4.4 |

This file has been truncated. show original

dusty_nv · March 14, 2021, 6:32pm

Hi @freewalker96, I haven’t personally Kubernetes or the container on it before, but perhaps someone from the community can share their experiences.

I don’t think the jetson-inference container should be very different from running another container with Kubernetes. You will want to use --runtime nvidia, along with mounting the camera devices and these data dirs. You can see how these are setup in the docker/run.sh script.

freewalker96 · March 16, 2021, 2:55am

Ok thank you @dusty_nv , I shall follow the steps as you said.