Has anyone used NX eMMC 16Gb for building a cluster in jetson mate? Not enough space for me... :<

Hi,
I have bought the jetson mate and try to build a cluster. But when I initialize the cluster and it starts download, NX is out of space. I couldn’t even restart the NX…
https://wiki.seeedstudio.com/Jetson-Mate/
I wonder what is the use of eMMC 16G as it is too small. :< How can I know what SDK I can remove for my application?
Is there any way I can expend the memory? Adding SSD is not an option as Kubernetes installation has to be done in each NX/nano.
Thx

You can try to expand it with external storage.

https://docs.nvidia.com/jetson/l4t/index.html#page/Tegra%20Linux%20Driver%20Package%20Development%20Guide/flashing.html#wwpID0E0QN0HA

I assume here is the procedure to make use of my NX (16G eMMc)…
I remove my NX module (with SD card slot) from the development board, and install the NX (with 16G eMMc) module to the development board. Then, I add a 250Gb SSD to the development board. Is it right?
May I know what I shall do next?
Here is the procedure I can think of

  1. Use Nvidia SDK manager to install 4.61 Jetson OS (without the SDK)
  2. Follow this video, Native Boot from SSD on Jetson Xavier (AGX and NX) - No SD Card Needed! - YouTube

Thanks

If that jetsonhacks video is more easier to understand, then just try it.