Gaussian 03

by Kathrin Helen Hopmann — last modified Jul 09, 2010 11:04 AM

Information on use of Gaussian 03 software on stallo

This page describes how to submit serial and parallel Gaussian calculations on stallo.

For help with inputfiles, see: Gaussian 03 Inputfiles

For information on Gaussview for g03, see: Gaussview for g03

GAUSSIAN 03 ON STALLO

To run Gaussian 03 on stallo, the module has to be loaded:

$ module load g03

If you download a run script (see below), this will be taken care of by the script.

Two run scripts are available (serial and parallel jobs respectively). Download one of the scripts and use it as described.

Run script for serial jobs: g03serial.pbs

Manually change account name, email, job name (without extension), and wall time in run script before use. Job submission:

$ qsub g03serial.pbs

Run script for parallel jobs: g03lparallel.job

Manually change account name and email in run script before use.

Job submission (example with file waterdimer.inp, 3 processors per node, 1 node, 20 h):

$ ./g03lparallel.sh waterdimer 3 1 20

NOTE:

%NProcshared=P (P = number of processors per node, maximum 8 on stallo)

%NProcLinda=N (N = number of nodes, keyword only required if you run on >1 node)

If you plan to run on >1 node (using %NProcLinda), make a new file in your $HOME directory:

.tsnet.config

containing only the line:

Tsnet.Node.lindarsharg: ssh

Note: if you run >1 node (using %NProcLinda), you have to use >1 processor per node
RECOMMENDATIONS:
If you are running geometry optimizations using DFT methods, we recommends using hybrid functionals. For pure functionals (no HF exchange), unless you have tested it (and see a benefit), it is recommended to run Gaussian on 1 node only, with maximum 3 processors on Stallo.
If your molecule is >60 atoms, Gaussian uses FMM by default (which is not parallelized), so do not run parallel in this case, unless you turn off FMM (NoFMM keyword).

Output (.out) and checkpointfiles (.chk) are transported to $HOME upon job completion.
Jobs are run in /global/work/$USER. The temporary directory is $PBS_JOBID or $JOB (depending on run script). Temporary directories are removed upon job completion. If jobs abort prematurely, temporary directories need to be removed manually.
To submit a .chk file along with an .inp file and/or retrieve a .chk upon job completion, make sure the .chk has the same name as the input file (e.g. filenames waterdimer.inp and waterdimer.chk, with keyword added to .inp: %chk=waterdimer).
RESTART OF JOBS: Retrieve the .chk file from the temporary directory and add the restart command to the input (opt=restart or scf=restart, depending on job). Make sure .chk and .inp have the same name. Submit as usual.
Gaussian version 03 is no longer supported by online manual on the gaussian.com web-site, but it is possible to download the manual.