Hi, we are a startup developing device remote management solution for NV Jetson systems.
The goal here is to allow users to operate devices deployed to the field remotely with our ready-to-use cloud platform and web portal. We are also able to setup some alerts, for example when GPU usage is too high. Would be great to get some feedbacks and suggestions from y’all.
We firstly install our agent on Jetson Nano and connect to the cloud and web portal. You can see there are some basic device information available once the device is connected.
The next step is to develop the plugin and include additional hardware information collected (with NV tool) and update all the information to the web portal.
(1) Build websocket connection with our agent for plugin to connect to our cloud
(2) register plugin for additional hardware information and alert
(3) register alert plugin with alert triggering threshold (eg. cputemp > 95)
As you can see in the video, the “memory and processor usage for jetson nano” section is added to the web portal. And users are able to define alert settings and how to get notified when device is abnormal.
Will this be useful functions to end customers when they are managing multiple jetson devices on the field? Give us some suggestions, thank you!