LAMMPS = Large-scale Atomic/Molecular Massively Parallel Simulator

This is the documentation for the LAMMPS 2001 version, written in F90, which has been superseded by more current versions. See the LAMMPS WWW Site for more information.

LAMMPS is a classical molecular dynamics code designed for simulating molecular and atomic systems on parallel computers using spatial-decomposition techniques. It runs on any parallel platform that supports F90 and the MPI message-passing library or on single-processor workstations.

LAMMPS 2001 is copyrighted code that is distributed freely as open-source software under the GNU Public License (GPL). See the LICENSE file or for more details. Basically the GPL allows you as a user to use, modify, or distribute LAMMPS however you wish, so long as any software you distribute remains under the GPL.

Features of LAMMPS 2001 include:

For users of LAMMPS 99, this version is written in F90 to take advantage of dynamic memory allocation. This means the user does not have to fiddle with parameter settings and re-compile the code so often for different problems. This enhancment means there are new rules for the ordering of commands in a LAMMPS input script, as well as a few new commands to guide the memory allocator. Users should read the beginning sections of the input_commands file for an explanation.

More details about the code can be found here, in the HTML- or text-based documentation. The LAMMPS Web page is at , which includes benchmark timings and a list of papers written using LAMMPS results. They illustrate the kinds of scientific problems that can be modeled with LAMMPS. These two papers describe the parallel algorithms used in the code. Please cite these if you incorporate LAMMPS results in your work. And if you send me citations for your papers, I'll be pleased to add them to the LAMMPS WWW page.

S. J. Plimpton, R. Pollock, M. Stevens, "Particle-Mesh Ewald and rRESPA for Parallel Molecular Dynamics Simulations", in Proc of the Eighth SIAM Conference on Parallel Processing for Scientific Computing, Minneapolis, MN, March 1997.

S. J. Plimpton, "Fast Parallel Algorithms for Short-Range Molecular Dynamics", J Comp Phys, 117, 1-19 (1995).

LAMMPS was originally developed as part of a 5-way CRADA collaboration between 3 industrial partners (Cray Research, Bristol-Myers Squibb, and Dupont) and 2 DoE laboratories (Sandia National Laboratories and Lawrence Livermore National Laboratories).

The primary author of LAMMPS is Steve Plimpton, but others have written or worked on significant portions of the code:

Other CRADA partners involved in the design and testing of LAMMPS are

If you have questions about LAMMPS, please contact me:

Steve Plimpton
Sandia National Labs
Albuquerque, NM 87185

More Information about LAMMPS

  • Basics
  • how to make, run, and test LAMMPS with the example problems
  • Input Commands
  • a complete listing of input commands used by LAMMPS
  • Data Format
  • the data file format used by LAMMPS
  • Force Fields
  • the equations LAMMPS uses to compute force-fields
  • Units
  • the input/output and internal units for LAMMPS variables
  • History
  • a brief timeline of features added to LAMMPS
  • Deficiencies
  • features LAMMPS does not (yet) have