Supercomputing Institute Technical IBM

How to run GAMESS
on the IBM SP and the SGI Origin


  1. Introduction
  2. The following sections demonstrate how to run GAMESS in both interactive and batch modes on the IBM SP and the SGI Origin.

  3. Interactive running of GAMESS
  4. Batch running of GAMESS
  5. A user defined scratch directory GAMESS_SCR can be defined with the command

    where directory is the particular scratch directory where you would like GAMESS to put its temporary files. This command can be added to your .cshrc file.

    1. IBM SP
      • load the gamess module

          module load gamess

      • launch the executeable

          qgms [options] input_file

        where input_file is the GAMESS input file and the options for the qgms script are as follows: (default values in brackets)

        -h             prints the options menu
        -n [winter]    specifies the type of node (options: night, winter)
        -p [1]         specifies the total number of processors
        -r [yes]       specifies if the job should be run (options: no, yes)
        -s [scracth1]  specifies the scratch space (options: scratch1, scratch2)
        -t [5:00:00]   specifies the amount of wall clock time (format hh:mm:ss)
        

        The qgms script creates the LoadLeveler file input_file.ll, creates a temporary directory for the GAMESS temporary files, and submits the job to the queue. If the -r flag has been set to "no" then the job will not be submitted; however the LoadLeveler script and the temporary directory will still be created. The temporary directory is removed when the calculation has finished. The LoadLeveler script can be submitted to the queue using the llsubmit command.

        Note: When requesting the number of parallel processors you must request an even number of processors! This is because GAMESS uses one processor for computing and the another for data management. Therefore, when specifying the number of processors with the -p option make sure to specify the total number of processors that you want to use. For example, 2, 4, 6, 8, etc.

      • Examples:

        1. Print out the options menu for qgms
            qgms -h

      • Run the GAMESS job, exam01 using the following description: run on four winterhawk+ processors with a maximum wall clock time of 5 hours and place the temporary files in /scratch1.
          qgms -p 4 exam01

        The output file, exam01.out is created in the current working directory with the input file. The LoadLeveler script, h2o.ll is also placed in the same directory as the input file.

      • Run the GAMESS calculation, exam01, on the nighthawk nodes with maximum wall clock time of 12 hours and place the temporary files in /scratch2.
          qgms -n night -s scratch2 -t 12:00:00 exam01

      • Create the LoadLeveler script, but do not submit the job to the queue.
          qgms -r no exam01

        The job can be run by typing

          llsubmit exam01.ll

        This command is echoed in the exam01.ll file.

    2. SGI Origin
      • load the gamess module

          module load gamess

      • launch the executeable

          qgms [options] input_file

        where input_file is the GAMESS input file and the options for the qgms script are as follows: (default values in brackets)

        -h                 prints the options menu
        -m [300mb]         specifies the amount of memory (format: ###mb)
        -n [12k]           specifies the type of node (options: 10k, 12k)
        -p [1]             specifies the total number of processors
        -r [yes]           specifies if the job should be run (options: no, yes)
        -s [scracth.o3.3]  specifies the scratch space (options: 31, 32, 33)
        -t [5:00:00]       specifies the amount of wall clock time (format hh:mm:ss)
        

        The qgms script creates the PBS file input_file.pbs, creates a temporary directory for the GAMESS temporary files, and submits the job to the queue. If the -r flag has been set to "no" then the job will not be submitted; however the PBS script and the temporary directory will still be created. The temporary directory is removed when the calculation has finished. The PBS script can be submitted to the queue using the qsub command.

        Note: When requesting the number of parallel processors you must request an even number of processors! This is because GAMESS uses one processor for computing and the another for data management. Therefore, when specifying the number of processors with the -p option make sure to specify the total number of processors that you want to use. For example, 2, 4, 6, 8, etc.

      • Examples:

        1. Print out the options menu for qgms
            qgms -h

      • Run the GAMESS job, h2o using the follwoing description: run on four R12000 processors with a maximum wall clock time of 5 hours and place the temporary files in /scratch.o3.3.
          qgms -p 4 h2o

        The output file, h2o.out is created in the current working directory with the input file. The PBS script, h2o.pbs is also placed in the same directory as the input file.

      • Run the GAMESS calculation, h2o, on the R10000 nodes with maximum wall clock time of 12 hours and place the temporary files in /scratch.o3.2.
          qgms -n 10k -s 32 -t 12:00:00 h2o

      • Create the PBS script, but do not submit the job to the queue.
          qgms -r no h2o

        The job can be run by typing

          qsub h2o.pbs

        This command is echoed in the h2o.pbs file.

  • Problems?
  • Please direct questions or problems to help@msi.umn.edu


    The University of Minnesota is an equal opportunity educator and employer.

    This information is available in alternative formats upon request by individuals with disabilities. Please send email to alt-format@msi.umn.edu or call 612-624-0528.

    HOME | QUESTIONS | FEEDBACK
    Events | Links | People | Publications | Support | Welcome
     


    URL: http://www.msi.umn.edu /software/gamess/content.html
    This page last modified on Wednesday, 23-Apr-2008 11:14:31 CDT  
    Please direct questions or problems to help@msi.umn.edu  
    Website related questions or problems should be directed to webmaster@msi.umn.edu
    The Supercomputing Institute does not collect personal information on visitors to our website. For the University of Minnesota policy, see www.privacy.umn.edu.