I have a Linux (Mint) PC with a Nvidia GPU, which i want to make into a CUDA build server. I want to write my CUDA code on a different (if possible Windows) machine and build it on the server. So to summarize I want to:
- Write my code on a client
- Connect to my server
- Transfer the source code to my server
- Build it there
- Run it on server
- Copy it back to the client machine
My question is how do I all of this in a hassle free way and if possible in a single step (by only running a single command)?