Running Gaussian 03 Jobs on Clark
Before you can run Gaussian 03 jobs on clark, you must be authorized as a member of the gaussian group. To determine if you are authorized, enter the id command, and verify that gaussian is listed as one of the groups to which your account belongs. If you are not a member of the gaussian group, please contact the system administrator.Gaussian 03 jobs must be run as batch jobs under PBS, submitted via the qsub command. For general information about how to submit PBS batch jobs on clark, see Submitting Jobs via PBS. There is no special queue for Gaussian 03 jobs. They should be submitted to the default queue.
The PBS job script file for a Gaussian 03 job should look something like this:
#PBS -l cput=24:00:00,ncpus=2,mem=2gbSpecify values for cput (CPU time required), ncpus (number of CPUs required), and mem (amount of memory required) that are appropriate for your job. The parameters on the #PBS statement can also be specified on the qsub command, but it is convenient to put them in the job file so that they are not forgotten. See the man pages for pbs and qsub for more details about submitting PBS jobs. The run_g03 command sets up the Gaussian 03 environment, and then invokes g03 with the specified parameters.
run_g03 test1.com
The ncpus resource specification is very important for job scheduling on clark. Job scheduling is based upon the load level of the system and the number of CPUs that are in use. The PBS scheduler cannot "look inside" of a job to determine how many CPUs it will actually use, so the scheduler must be told how many CPUs a job will use via the ncpus resource. Please be sure that the value of ncpus is the same as the value of %nproc in the Gaussian 03.com file. If the value of ncpus is too high, then your job may wait in the queue, even though CPU resources are available. If the value of ncpus is too low, then your job may be cancelled if it actually uses more CPUs than you specified. If ncpus is not specified, PBS will assume that 1 CPU is needed. (Note: clark has 64 CPUs. However, since clark is shared by many users, jobs that request more than 8 CPUs may have to wait longer in the job queue, depending upon system load.).
Use the qstat command to see which jobs are queued/executing on clark:
qstat -aThe directory to use for temporary work files on clark is /scratch. The run_g03 command uses that directory by default. If you explicitly specify the pathnames for work files in your .com files, please modify them accordingly.
If you run the Gaussian 03 utility newzmat, you will need to set up the Gaussian 03 environment for your interactive session. Place the following lines in the .profile file in your home directory:
g03root=/usr/localFor more information about accessing clark in general, see Accessing Clark.
GAUSS_SCRDIR=/scratch
export g03root GAUSS_SCRDIR
. $g03root/g03/bsd/g03.profile