I recently started using Paralution. At first it all worked fine and I very easily could start doing some computations using the GPU. Unfortunately, that’s not the case anymore…
When I now run any Paralution program (my own or any of the examples), I get the following error:
modprobe: ERROR: could not insert ‘nvidia_uvm': Invalid argument
Querying for GPU devices failed – falling back to host backend
I use Paralution on the cluster of the company I do my thesis for and so I have no sudo-rights… The cluster administrator told me nothing changed on the cluster since last time, when Paralution was working for me. But I also did not change anything so why did this stop working? Any help for this?