ORCA

Introduction

ORCA is computational chemistry program focusing on quantum chemistry applications. The program is designed for modeling electronic structures as well as spectroscopic properties of molecules. The program contains a variety of different methods for performing such simulations.

Using ORCA on RCC Resources

The ORCA quantum chemistry tool is available on the HPC cluster for parallel processing of ORCA jobs.

Submitting a Job

Following is a sample ORCA input parameter file. Note the line ! PALx (x = 2 in the example) at the begining where x is the number of parallel processes you would like to run simultaneously.

MY-INPUT-FILE.inp:

## My first parallel ORCA calculation :-)
#
! PAL2
! HF SVP
* xyz 0 1
C 0.0 0.0 0.0
O 0.0 0.0 1.13
*

Create a SLURM submission script. For example:

#!/bin/bash
#SBATCH -t 00:10:00
#SBATCH -n 2
#SBATCH --ntasks-per-node=1
#SBATCH -p $PARTITION

module load orca
orca MY-INPUT-FILE.inp

Where $PARTITION is the name of a partition which you have access to. A list of the HPC partitions can be found here.  Note that ORCA may throw an error relating to an inability to access or execute an executable file.  If this occurs, then you will need to provide the entire absolute file path when calling ORCA.  See the Troubleshooting section below for details.

Testing ORCA scripts

You can also run ORCA scripts on the HPC login nodes for testing purposes only. Ensure that you remove the ! PALx line:

module load orca
orca MY-INPUT-FILE.inp

Troubleshooting

When you load the ORCA module (module load orca), the gnu-openmpi module is implicitely loaded. If you have already loaded other mpi modules, you will see this error:

gnu-openmpi(6):ERROR:150: Module 'gnu-openmpi' conflicts with the currently loaded module(s) 'intel-mvapich2'

Unload the conflicting module first, for example, by running module purge, before you load the ORCA module.

If ORCA throws an error relating to an inability to access or execute an executable file, the way to fix it, according to the ORCA Users Manual on the official website, is to type the entire path to ORCA on the HPC before the .inp file when calling the program.  For example:

$PATH MY-INPUT-FILE.inp 

Where $PATH is the complete absolute file path which can be found using which orca.  Once which orca, is run just copy and paste the absolute file path it returns into the command line or the line in your SLURM submission script that calls ORCA and type the input file's name as shown above.

Version
4.0.0.2