How can I execute a CUDA code on an NVIDIA card via network using a VPN ?
For information, I’m using Ubuntu 16.04 as an operating system.
Thank you.
You probably want to ask that question here:
[url]https://devtalk.nvidia.com/default/board/58/cuda-setup-and-installation/[/url]
A short answer is that the system actually running the GPU commands must have X11 running (this is the interface to the GPU driver which CUDA uses) and the application bound to that server via the “DISPLAY” environment variable (normally “export DISPLAY=:0” if local to the computer you are logged in to).
A server can be real (displaying to a monitor) or it can be virtual (displaying to a buffer which is treated as if it were a monitor). So if you use a remote login to another computer and want that other computer to use its GPU, then “DISPLAY” must exist there and must be running. One solution is to attach a monitor there and run a GUI display to it…an alternative is to run a virtual desktop there and never attach a monitor (remote desktop sharing software does this…it makes it possible to see remotely by viewing a copy). There are several remote desktop sharing solutions available.
EDIT: I just saw your post at [url]Executes code on an NVIDIA Jetson TX2 via network using VPN - Jetson TX2 - NVIDIA Developer Forums a good place for the question since it is a TX2.