lammps-rpm-rspm

Description

LAMMPS implementation of the replica-permutation method (RPM) and replica sub-permutation method (RSPM). These methods are improved alternatives to the replica-exchange method.

The below is a list of codes that are put in src directory:

1. temper_permutation.h, temper_permutation.cpp: RPM in the NVT ensemble.
2. temper_subpermutation.h, temper_subpermutation.cpp: RSPM in the NVT ensemble.
3. temper_permutation_npt.h, temper_permutation_npt.cpp: RPM in the NPT ensemble.
4. temper_subpermutation_npt.h, temper_subpermutation_npt.cpp: RSPM in the NPT ensemble.

Build

First of all, clone a repository of LAMMPS and change the branch as follows:

# Clone repository
$ git clone https://github.com/lammps/lammps.git
$ cd lammps
$ git checkout stable_3Mar2020

After changing the branch, put the codes of the RPM and RSPM (i.e., temper_*permutation*.h and temper_*permutation*.cpp) in the lammps/src/USER-MISC/. Then, install required packages and compile codes.

$ cd lammps/src

# Installing packages
$ make yes-kspace
$ make yes-molecule
$ make yes-rigid
$ make yes-replica
$ make yes-user-molfile
$ make yes-user-misc
$ make yes-opt
$ make yes-user-omp
$ make yes-user-intel

# Compile
$ make serial -j 10
$ make mpi -j 10

After compile, you will find two executable binary files; lmp_serial and lmp_mpi.

LAMMPS version

Our source codes were tested with LAMMPS stable version 3Mar20. We should remark that there were significant changes in the release of stable version 29Oct20 (In details, please see https://github.com/lammps/lammps/releases/tag/stable_29Oct2020). Because we have not been tested our codes with this version yet, we recommend that you use the stable 3Mar20 version of LAMMPS in order to perform RPM and RSPM.

Using codes

RPM and RSPM in the NVT ensemble

The RPM and RSPM in the NVT ensemble are provided as commands temper/permutation and temper/subpermutation, respectively. The syntax is as follows:

temper/permutation N M temp fix-ID seed1 seed2 index

temper/subpermutation N M temp fix-ID seed1 seed2 index

• N: the total number of simulation time steps.
• M: a time step interval of the temperature permutation (sub-permutation).
• fix-ID: a fix command ID of a thermostat.
• seed1: random number seed to decide on which set of subsets are used in the permutation attempts.
• seed2: random number seed for the Suwa-Todo algorithm.
• index (optional): specifies the temperature label of the replica.

Note that these commands have the same syntax as the temper command (https://docs.lammps.org/temper.html).

RPM and RSPM in the NPT ensemble

The RPM and RSPM in the NPT ensemble are provided as commands temper/permutation/npt and temper/subpermutation/npt, respectively. The syntax is as follows:

temper/permutation/npt N M temp fix-ID seed1 seed2 pressure index

temper/subpermutation/npt N M temp fix-ID seed1 seed2 pressure index

• pressure: an pressure value of the replica.

Note that these commands have the same syntax as the temper/npt command (https://docs.lammps.org/temper_npt.html).

Running simulations

Sample input files are put in inputs directory.

# An example simulation using temper/permutation command
$ cd inputs/poly-ethlene-chain/rpmd
mpirun -np 8 ${LMP} -partition 8x1 -in input-seed1.in > run-seed1.log

# An exmaple simulation using temper/subpermutation command
$ cd inputs/poly-ethlene-chain/rspmd
mpirun -np 8 ${LMP} -partition 8x1 -in input-seed1.in > run-seed1.log

# An example simulation using temper/permutation/npt
$ cd inpust/ala-dipeptide/rpmd
mpirun -np 80 ${LMP} -partition 10x8 -in md-rpmd-seed1.in > md-rpmd-seed1.log

# An example simulation using temper/subpermutation/npt
$ cd inputs/ala-dipeptide/rspmd
mpirun -np 80 ${LMP} -partition 10x8 -in md-rspmd-seed1.in > md-rspmd-seed1.lo

References

1. S. G. Itoh and H. Okumura, JCTC (2013)
2. M. Yamauchi and H. Okumura, JCP (2017)
3. M. Yamauchi and H. Okumura, JCC (2019)
4. M. Yamauchi, G. La Penna, S. G. Itoh, and H. Okumura, Comput. Phys. Commun. (2022)