LAMMPS

Warning

The ARCHER2 Service is not yet available. This documentation is in development.

LAMMPS, is a classical molecular dynamics code, (“Large-scale Atomic/Molecular Massively Parallel Simulator”). LAMMPS has potentials for solid-state materials (metals, semiconductors) and soft matter (biomolecules, polymers) and coarse-grained or mesoscopic systems. It can be used to model atoms or, more generically, as a parallel particle simulator at the atomic, mesoscopic, or continuum scale.

Using LAMMPS on ARCHER2

LAMMPS is freely available to all ARCHER2 users.

The centrally installed version of LAMMPS is compiled with all the standard packages included: ASPHERE, BODY, ..

We do not install any USER packages. If you are interested in a USER package, we would encourage you to try to compile your own version and we can help out if necessary (see below).

Running parallel LAMMPS jobs

LAMMPS can exploit multiple nodes on ARCHER2 and will generally be run in exclusive mode using more than one node.

For example, the following script will run a LAMMPS MD job using 4 nodes (128x4 cores) with MPI only.

Warning

The following script is provisional and needs to be verified

#!/bin/bash --login

#SBATCH --job-name=lammps_test
#SBATCH --nodes=4
#SBATCH --ntasks=512
#SBATCH --tasks-per-node=128
#SBATCH --cpus-per-task=1
#SBATCH --time=00:20:00

# Replace [budget code] below with your project code (e.g. t01)
#SBATCH --account=[budget code]

# Load the relevant LAMMPS module
module load lammps/2019-12

srun lmp_mpi -in in.test > out.test

Hints and Tips

Compiling LAMMPS

The large range of optional packages availble for LAMMPS, and opportunity for extensibility, may mean that it is convenient for users to compile their own copy. In practice, LAMMPS is relatively easy to compile, so we encourage users to have a go.

Compilation instructions for LAMMPS on ARCHER2 can be found on GitHub:

https://github.com/hpc-uk/build-instructions/tree/master/LAMMPS

If you get stuck, please contact the Service Desk.