Computational Cluster Programs

How to Run GROMACS on ATS-Hosted Clusters

How to Run GROMACS in Batch

To run the molecular dynamics part (mdrun) of the GRMACS suite of programs in batch, you are required to have previously generated an input file containing information about the topology, the structure and the parameters of your system. Such input file, which generally has a .tpr, a .tpb or a .tpa extension, is generated via the grompp part of GROMACS. This input file is the control file to be provided to the ATS queue scripts (e.g., gromacs.q etc., see later) when submitting a serial or parallel mdrun job.

To run pre- and post-processing parts of GROMACS (such as pdb2gmx, grompp, etc.) you need to source the following script:

/u/local/apps/gromacs/4.0.5/serial/bin/GMXRC

(or:

/u/local/apps/gromacs/4.0.5/parallel/bin/GMXRC

for the parallel installation).

Please note that the serial GROMACS codes in /u/local/apps/gromacs/4.0.5/serial/bin have a _d suffix, while the parallel GROMACS codes in /u/local/apps/gromacs/4.0.5/parallel/bin have a _mpi suffix.

Please submitt your the pre- and post-processing GROMACS jobs either invoking an interactive node via qrsh (see http://www.ats.ucla.edu/clusters/hoffman2/computing/sge_qrsh.htm) or using job.q to generate an SGE command file to run such jobs in batch.

Please note that you will need to add the following lines in such command file:

source /u/local/apps/gromacs/4.0.5/serial/bin/GMXRC

source /u/local/compilers/intel/11.1/current/bin/ifortvars.csh intel64

To Run GROMACS from the UCLA Grid Portal

GROMACS is not available from the UCLA Grid Portal.

To Run GROMACS from the Cluster Head Node using the Queue Scripts

The easiest way to run GROMACS in batch from the head node is to use the queue scripts. See Running a Batch Job on an ATS-Hosted Cluster for a discussion of the queue scripts and how they are used.

The following queue scripts are available for GROMACS:

gromacs.q

Asks you whether you want to submit a parallel or serial job and submits the run accordingly.

gromacs.q.parallel

Submits parallel jobs.

gromacs.q.serial

Submits serial jobs.

To Run GROMACS from the Cluster Head Node Using SGE Comands

See Running a Batch Job on an ATS-Hosted Cluster for guidelines to follow to create the required SGE command file. Alternatively, you could create an SGE command file with one of the queue scripts listed above. After saving the command file, you can modify it if necessary. See Commonly-Used SGE Commands for a list of the most commonly used SGE commands.

How to Run GROMACS Interactively

GROMACS is not available interactively.