Checkpoints for running Converge
 

Hello all,

I am using an HPC at the university I'm attending where the maximum allowable wall time for calculation is 24 hrs. My simulations, indeed, need more than that, so the supporting team of the HPC facility informed me that I can use "checkpoints". This way I can run for 24 hours and pick up the slack from where it stopped with the next job submission.

I went through the Converge manual and Getting-started manual, but I couldn't find anything similar. So, may I ask you, please, is this option available? And if so, I would be truly appreciative if you could guide me on how to do it.

That should be standard restart files.


You can specify crank angle or time based restart files (every 1h for example).


May be a bit difficult to automate this.