Read more about the gqueue queue submission utility.

Cluster users who need to submit Gaussian jobs for execution on a cluster: no need to write GridEngine queue scripts to wrap your Gaussian input files. Let the gqueue utility do that for you! Here’s the sequence of steps; a sample session is shown below to give you a better idea of what to do:

1. Move into the directory containing your .com file
2. Determine the parallel environment (PE) type used for Gaussian calculations
3. Submit your .com file using gqueue and the PE type you chose

% cd [directory containing .com file]
% gqueue --pe-summary
Available parallel-environment types (first is default):

------------------------------------------------------------------
gaussian                            /etc/gqueue.d/scripts/gaussian
+---------------+-----+------------------------------------------+
 ID               CPU   Flags
+---------------+-----+------------------------------------------+
 gaussian_dual      2   round-up
------------------------------------------------------------------
   :

% gqueue -v --nproc=2 --pe-type=gaussian_dual H2O.com
Feb 02 10:50:41 gqueue: Job will use queue script: /home/frey/.gqueue/H2O.gqs_XXOznQN9
Feb 02 10:50:41 gqueue: Scanning /home/frey/H2O.com for processor count and memsize
Feb 02 10:50:41 gqueue: Selected parallel environment 'gaussian_dual' with nproc = 2
Feb 02 10:50:42 gqueue: Merging Gaussian input file H2O.com
Feb 02 10:50:42 gqueue: Working directory for calculation: /home/frey
Feb 02 10:50:42 gqueue: Log file for calculation: H2O.log
Your job 40967 ("H2O.gqs_XXOznQN9") has been submitted.

You can use the standard GridEngine utilities to monitor your job from here: qstat and qjobs. You can also check your cluster’s web-based monitoring tools.