Gaussian is a widely used non-empirical molecular orbital calculation program. On a large-scale computer system, Gaussian16.C.01 can be used by batch request. Please note that the use of Gaussian is on a yearly application basis. If you wish to use Gaussian, please refer to the following page.

Application for Gaussian16

Basic usage

    This section explains how to use SQUID to run Gaussian. If you want to know how to use it in more detail, please refer to Gaussian official web page and the User Reference in the official web page.
    The execution of Gaussian is only allowed to process by batch request. Please connect to the front-end node, create the input data and job script necessary for the calculation, and submit it as a batch job. The input data and job script can be created on the local terminal or by connecting to the front-end node and using a text editor such as vi or emacs. For more information on how to use the editor, please refer to Editor usage page.

    Creating an input file

      To begin, create an input file. Using the editor, fill in the following example (single point energy calc on water) and save it with an appropriate name. Here we have saved it as "single_point_energy_calc_on_water".

      Line 1: Link0 Command Scratch file location and name (no blank line at the end) This example specifies the location and name of the checkpoint file.
      Line 2 This example has a line break in it, but it's not really necessary
      Line 3 Specify the type of calculation you want to do, the model chemistry and other options (an empty line marks the end)
      Line 5 A brief description of the calculation (an empty line marks the end)
      Line 7 Specify the molecular system to be considered (an empty line marks the end)
      In this example, 0 charge (neutral molecule), spin multiplicity 1 (singlet)
      Line 8-10 Specify the position of each atom in the molecule
      Line 11-12 Please be sure to include a new line at the end of the file. (If there is no line feed, an error will occur.)

      For a detailed description of the input files, please refer to the About Gaussian 16 Input.


    Creating a Job Script

      Next, create a job script. Use an editor and save it as "".

      Line 2 Specify the execution queue
      Line 3 Specify a group (if you are not sure, run the groups command. ocean" or "hpci" next to the group name).
      Line 5-8 Sets the environment variables required to start Gaussian.
      module load Running environmental modules to set up Gaussian environment variables
      newgrp gaussian This is the group setting to access the Gaussian directory
      export GAUSS_SCRDIR=/sqfs/work/[group name]/[User ID] This is the specification of the scratch directory of Gaussian
      Line 9 It is moved to the current directory at the time of job submission.
      PBS_O_WORKDIR This is an environment variable for our Scheduler, NQSV, which sets the current directory when qsub command is executed.
      Line 10 Gaussian execution command
      Redirect (<) the input file you have created, and redirect (>) the result to "result" (since the standard output/standard error output files are restricted, it is recommended to use Redirection (>) to a file is recommended.)
      Line 11 Be sure to insert a blank line.

      Please refer to the How to write a job script page for a detailed explanation of the contents of the job script. In this section, we will explain the items required for Gaussian use.

    Job Script Submission

      After the input file and job script are completed, submit the job.

      % qsub (enter)

      If the execution is successful, the calculation results will be saved in the result file "result".


    reference information