Biowulf at the NIH
RSS Feed
TURBOMOLE on Biowulf
Turbomole Logo

TURBOMOLE is a quantum chemical program package, initially developed in the group of Prof. Dr. Reinhart Ahlrichs at the University of Karlsruhe and at the Forschungszentrum Karlsruhe. Until 2007 the main development of the program was conducted by students and postdoctoral researches in the group of Ahlrichs, who usually changed their field of work after leaving the group. Therefore, the code was well localized at and, consequently, owned by the University of Karlsruhe.

Presently TURBOMOLE is one of the fastest and most stable codes available for standard quantum chemical applications. Unlike many other programs, the main focus in the development of TURBOMOLE has not been to implement all new methods and functionals, but to provide a fast and stable code which is able to treat molecules of industrial relevance at reasonable time and memory requirements.

How to Use

Load the turbomole module:

module load turbomole

To see all versions of TURBOMOLE, type

module avail turbomole

See here for more information on using environment modules on Helix Systems.

Generate a TURBOMOLE coord file from a .xyz input using the x2t command,

x2t xyzinfile > coord

then generate a control file using the define command.

Once the coord and control are made, run the appropriate executable.

Running on Biowulf

Here is a sample qsub script turbomole_pbs.sh for running a TURBOMOLE job on Biowulf:

#!/bin/sh #PBS -N turbomole #PBS -e turbomole.err #PBS -o turbomole.out module load turbomole ######## ENTER YOUR JOB HERE ################################## cd $PBS_O_WORKDIR jobex -ri > jobex.out ###############################################################

This would be submitted like this:

qsub -l nodes=x -v PARNODES=y,OMP_NUM_THREADS=z turbomole_pbs.sh
x = the number of nodes
y = the total number of cores available on x nodes
z = the number of cores available on each node

OpenMP Parallel Jobs

The programs dscf and ricc2 are parallelized to run on shared-memory multiprocessor machines. They are named dscf_smp and ricc2_smp. To use these programs, the environment variable OMP_NUM_THREADS needs to be set to the number of processors.

On a quad-core, 8 processor node, this can be done with

export OMP_NUM_THREADS=8

prior to execution.

Documentation

User manuals