Automated workflows

To obtain all mechanistic and kinetic information for your system, you can simply run a single script per each type of calculation: low-level and high-level.

First, load the module:

module load amk/2021

Then, the llcalcs.sh script can be employed to run the low-level workflow:

nohup llcalcs.sh molecule.dat ntasks niter runningtasks >llcalcs.log 2>&1 &

where:
molecule is the name of your molecule
ntasks is the number of tasks per iteration
niter is the number of iterations
runningtasks is the number of simultaneous parallel tasks

Finally, all high-level calculations can be accomplished using hlcalcs.sh:

nohup hlcalcs.sh molecule.dat runningtasks >hlcalcs.log 2>&1 &

The llcalcsh.sh and hlcalcs.sh scripts can also be run using a Slurm job scheduler and the following scripts are examples:

#!/bin/bash
ntasks=20
niter=2
filename=FA
module load amk/2021
sbatch --output=llcalcs.log --error=llcalcs.err -n $ntasks --mem-per-cpu=2G -t 00:10:00 llcalcs.sh ${filename}.dat $ntasks $niter

#!/bin/bash
ntasks=8
ncores=4
filename=FA
module load amk/2021
module load g16
sbatch --output=hlcalcs.log --error=hlcalcs.err -n $ntasks -c $ncores --mem-per-cpu=2G -t 00:30:00 hlcalcs.sh ${filename}.dat

Details about the workflows involved in llcals.sh and hlcalcs.sh can be looked up in this section.

If something goes wrong, you might want to have a look at this section describing how to stop the calculations.