orca

Summary

Source: Download page on orca forum - requires registration

Documentation: https://orcaforum.kofo.mpg.de/index.php

License: see below

Path: /software/orca

ORCA is an ab initio quantum chemistry program package that contains modern electronic structure methods including density functional theory, many-body perturbation, coupled cluster, multireference methods, and semi-empirical quantum chemistry methods. Its main field of application is larger molecules, transition metal complexes, and their spectroscopic properties.More information about Orca can be found on the orca forum.

Installations on Maxwell

Various versions are installed on the maxwell cluster. To set the environment use the module command

module load maxwell 
module load orca         # will select the most recent version
module load orca/6.0.1   # will select a specific version

xwhich orca              # will list available orca versions and corresponding module commands

When using orca you implicitly agree to the license and terms of use, for ORCA and a number of Third Party products included into ORCA.

Orca License

The EULA license for orca can be obtained from the orca forum. A current copy of the EULA license and Third Party License can be found under /software/orca/license/. Please read prior to using ORCA!

Particularly important:

OBLIGATIONS OF LICENSEE:

LICENSEE shall immediately notify SGK if LICENSEE or ACADEMIA LICENSEE is working for, is filing a patent as a result of LICENSEE’s use of SOFTWARE.

MPI and SGK are entitled to use the information in the patent application for marketing purposes and to publish it for this specific purpose.

If results obtained with SOFTWARE are published in the scientific literature, LICENSEE shall reference the SOFTWARE as:

“F. Neese: Software update: the ORCA program system, version 4.0 (WIREs Comput Mol Sci 2018, 8:e1327. doi: 10.1002/wcms.1327)”. Using specific methods included in SOFTWARE may require citing additional articles, as described in the manual. LICENSEE agrees to honor the request to cite additional papers as appropriate.

Orca sample batch script

a simple sample batch script is outlined below. orca produces a lot of temporary files, so better don't use your home-directory as the working directory. Use DUST (if available) or your groups GPFS space (if available). It's in many case possible to work entirely in memory, but that's a more complex setup.

Lets assume your batch-script is called sbatch-orca.sh

#!/bin/bash
#SBATCH --constraint='[9374F|75F3]'
#SBATCH --partition=upex
#SBATCH --time=0-08:00:00
#SBATCH --nodes=1
#SBATCH --job-name=orca-%j
#SBATCH --output=orca-%j.out
unset LD_PRELOAD

source /etc/profile.d/modules.sh
module load maxwell orca

# input file (without .inp)
input=whatever

# use all physical cores
N=$(( $(nproc) / 2 ))

# modify input-file accordingly 
perl -p -e "s|%PAL NPROCS .* END|%PAL NPROCS $N END|" $input.inp > $input.$SLURM_JOB_ID.inp

# run orca
`which orca` $input.$SLURM_JOB_ID.inp "-N $N"

... and submit your batch-job with

sbatch sbatch-orca.sh

You will of course need to replace "whatever" with the name of your input-file and the partition and runtime of your job (8 hours in the sample).

SLURM will replace %j with the SLURM JobId, which is also the value of $SLURM_JOB_ID. You can later on display the job-script used with ...

sacct -j <job-id> -B