I am trying to profile a GPU I have set up on a remote Ubuntu system using NVIDIA Visual Profiler. I am following the guide here (https://docs.nvidia.com/cuda/profiler-users-guide/index.html#remote-profiling) but it doesn’t mention anything about how to set up a connection using keys, nor does it look like I have the ability to.
Any help is appreciated.
Bonus question: I have CUDA 10.0 running on both systems (my host is a Windows), but the two systems have different revisions (10.0.326 on Ubuntu and 10.0.130 on Windows). Is that going to be OK?