Gaussian is a set of programs for performing semi-empirical, density functional
theory and ab initio molecular orbital calculations.
NOTE: Users are required to sign a license agreement form before
being allowed to run this software. Please fill out the
support form
to check if you have permission to use this software.
Gaussian03 (g03) can be run using from 1 to 8 processors with shared memory. This program runs with 4 threads by default (that is, if no thread information is specified in your parameter file).
Example of a shell script to run g03 on the batch queue:
Script filename (sub.sh):
#!/bin/csh cd working_directory source /usr/local/g03/bsd/g03.login /usr/local/g03/bsd/g03l test.com test.log
NOTE: In /usr/local/g03/bsd/g03l the last character is the letter L in
lower case, not the number one. The second character in g03 and in g03l
is the number zero.
If your parameter file (test.com) does not specify the number of processors (or threads), you can submit the gaussian job to the batch queue (using for example up to 10h of wallclock time) with:
pcluster> ugsub T4-i1-t4-10h sub.sh
You can specify more than 4 threads (up to 8) in your parameter file (test.com) with the variable %NprocShared. For example, if you request 8 processors
with %NprocShared=8, then type the following to submit the job (for example, using up to 24h wallclock time):
pcluster> ugsub T8-i1-t8-24h sub.sh
Because currently pcluster does not have TCP-Linda, gaussian jobs cannot run across nodes;
therefore you can only use %NprocLinda=1 and a maximum of 8 processors. To run Gaussian03 using
a single processor, specify %Nprocshared=1 in your parameter file. Note that the header line
%nproc of older versions of Gaussian can still be used with the current version.
NOTE: Please send your job to a 4 processor queue if your parameter file does not specify the number of processors to be used (because the default number of threads is 4).
Type llqueue at your pcluster login prompt to get more
information about the batch queuenames.
PLEASE NOTE: g03 jobs generate temporary files called Gau-*
in your scratch area (that
is, in /scratch/yourusername). The job automatically deletes these files when it
completes successfully. However, if the job crashes (or if you remove a running g03 job), these
files are left in your scratch area. Please remove
leftover temporary files manually so they do not accumulate on the scratch area (these temporary
files can be huge and they fill up the scratch area very easily).
Script filename (sub.sh):