Deploying AI Applications with NVIDIA EGX on NVIDIA Jetson Xavier NX Microservers

Technical Blogs & Events Technical Blog
1

Originally published at: https://developer.nvidia.com/blog/deploying-ai-apps-with-egx-on-jetson-xavier-nx-microservers/

Modern expectations for agile capabilities and constant innovation—with zero downtime—calls for a change in how software for embedded and edge devices are developed and deployed. Adopting cloud-native paradigms like microservices, containerization, and container orchestration at the edge is the way forward but complexity of deployment, management, and security concerns gets in the way of scaling.…

2

All things good up until running command:
helm install --name-template deepstream deepstream-helmchart/

error:
Error: failed to download “deepstream-helmchart/” (hint: running helm repo update may help)

~/deepstream-helmchart/templates$ helm repo update
Hang tight while we grab the latest from your chart repositories…
…Successfully got an update from the “stable” chart repository
Update Complete. ⎈ Happy Helming!⎈

error persisted.

3

Missing an import step for newbie:

helm create deepstream-helmchart

4

Tac,

You don’t need to run helm create as we provided instructions how to create a helm directory and how to add the files to helm directory.

image
image1168×468 51.1 KB

Thanks,
Anurag G

5

I ran into a small issue following the doc. The doc has you name the Chart file chart.yaml. For me on a cluster of Jetson Nanos, I got the following error. I then renamed the file to Chart.yaml (capital C is key) and the deployment worked as planned.

kejones@jetson-nano-1:~$ helm install --name-template deepstream deepstream-helmchart/
Error: validation: chart.metadata is required
kejones@jetson-nano-1:~$ mv deepstream-helmchart/chart.yaml deepstream-helmchart/Chart.yaml
kejones@jetson-nano-1:~$ helm install --name-template deepstream deepstream-helmchart/
NAME: deepstream
LAST DEPLOYED: Wed Sep 30 18:26:51 2020
NAMESPACE: gpu-test
STATUS: deployed
REVISION: 1
TEST SUITE: None
6

Thank you for the feedback, updated the doc with Chart.yaml. it should be good now.

7

Small nit – the instructions say to install calico pod-network addon but the example commands deploy flannel.

8

Thank You, Jacob will update according to that.

9

Thanks for the detailed instructions, I was able to make it to the end. Can you add some explanation on what the sample output (below, from blog) means. Also is there any image/video output (since supposed to be IVA example)? I didn’t see any.

$ kubectl logs deepstream-9f8b6b68d-rc5lq
**PERF: 101.42 (100.29) 101.42 (100.29) 101.42 (100.29) 101.42 (100.29)

10

Hi Sudesh,

Thanks for the following this blog. We will update the instructions about the video output for IVA.