LAMMPS

LAMMPS

lammps

Contents

            1. Description

            2. Requirement

            3. How to run job Lammps using HPC

            4. Contact Us

Description:

Version: lammps-28Jun14
LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator.
http://lammps.sandia.gov/

Requirement

Step 1: Please register to UTM HPC Account
Step 2: HPC admin (hpc@utm.my) will sent HPC account to your email.

How to run job Lammps using HPC

Submit a Lammps job

In the example below the input files in.chain will be run after submit a job using PBS script, pbs-script. For example, create a submission script file called pbs-script that contains the lines below. This example requests one node and eight processor cores.

#!/bin/sh
#PBS -N Lammps
#PBS -e defaults_error
#PBS -l nodes=1:ppn=8
#PBS -m abe
#PBS -M abe@utm.my

module load lammps/28Jun14

cd $HOME
mpirun -np 8 lmp_mpi < in.chain > run.1.out

Transfering Files & Submitting Job

Then, upload the submission script above and your input file to the HPC using WinSCP tool. To submit a job, please using a Putty tool and run the command below. How to access the HPC server, please go to our New User Guide page.
 
Submit the Lammps submission job to the queue:

Users can submit the job using two options as below. The pbs-script name below is the name of your submission script.

qsub -q bigmem pbs-script
OR
qsub -q hpc pbs-script
Please note that queue for hpc above have 8 core’s for each cluster/node so if you are using this cluster please make sure in the script, you state “ppn=8″ to make sure your job can run.

Contact Us

Any problem about HPC, please email to hpc@utm.my or our facebook (Lammps)