Working with supercomputers is a pain in the ass. It’s a miracle these things even boot up with so many devices to keep track of, but using software gets far more weird. Half the operations work on one version and another half works on the other, but only if the number of used processors is even… Maybe it’s ompi shenanigans, who knows. Weird.
Mine just randomly crashes like once every ten minutes. SSH session freezes and that's it. And nwchem doesn't give any readable error output if the input file is weird or incorrect, just kinda hangs. Then again, quantum chemistry packages are written mostly by scientists, not by programmers. Hence the weirdness.
@johnabs
Thanks a lot, I’ll definitely check these! The ssh fix should really help. Now slurm queue died, so I’m waiting for support to reply :D
It’s university computer so things are a bit sketchy, but at least I dont pay for it.
@academicalnerd To fix the first issues, you can try the following on your local machine. (BTW, if you’re not using TRAMP in Emacs to ssh, you’re doing it wrong IMO. It’s absolutely a killer feature for ssh to improve server interactivity and such)
As for nwchem, I checked out the user manual. Unfortunately, it has nothing on debugging simulations (typical of molecular dynamics and QM software). However, from my own work on a prior thesis, I can make 2 suggestions:
I hope that helps! Supercomputer access should always be fun, I hope this pushes it in that direction for you 😎 😂