Simpatico - Simulation Package for Polymeric and Molecular Liquids

Copyright 2010 - 2020 The Regents of the University of Minnesota Distributed under the terms of the GNU General Public License.

Simpatico is a C++ package for Molecular Dynamics (MD), Monte Carlo (MC), and hybrid MC simulations of classical mechanical models of polymeric and molecular liquids. It has thus far been used primarily for simulating course-grained models of polymer liquids.

The simpatico package contains:

• A program for parallel MD simulations, named ddSim.

• Programs for Monte Carlo (MC) and molecular dynamics (MD) simulations on a single processor, named mcSim and mdSim.

The ddSim parallel MD program uses a spatial domain decomposition strategy to allow efficient simulation of very large systems. The mcSim and mdSim simulation are convenient for simulations on a personal computer. The single-processor simulations also provide convenient ways to submit embarassingly parallel simulations of many similar systems with one system per CPU core, which can be used to improve statistics in MC sampling or to explore a parameter space.

All three simulation programs (ddSim, mdSim and mcSim) provide very flexible, extensible facilities for on-the-fly output and analysis of selected physical variables during a simulation. The single-processor mcSim and mdSim programs can also be used for postprocess analysis of trajectories, by applying the same analysis algorithms to a sequence of configuration snapshots read from trajectory file.

Simpatico is distributed only in source code form, and so must be compiled from source.

The simpatico source code is maintained in the github repository

https://github.com/dmorse/simpatico.

It may be obtained by using a git version control system client to clone the repository. To do so, enter the command:

git clone --recursive https://github.com/dmorse/simpatico.git

Note the use of the --recursive option to the git clone command: This is necessary to clone some git submodules that are maintained in separate repositories.

A recent copy of the web manual for simpatico is available online, at

http://dmorse.github.com/simpatico/doc/html/index.html.

This manual provides both user and developer documentation in an integrated form.

The web manual for simpatico is prepared using the Doxygen documentation utility (www.doxygen.org), and can be regenerated by users. Running doxygen creates html web pages that are deposited in the directory named doc/html/ within a users copy of the simpatico working tree. These web pages are created from both comments extracted from the C++ source code and from a set of text files with the file extension *.dox. Files containing text for the main sections of the web manual are located in the directory doc/manual/, and are readable in any text editor.

After cloning the source code, you can use doxygen to generate a local copy of the html documentation within the simpatico/doc/html directory of the source code tree. This requires that doxygen be installed on your computer and that the executable be in a directory in your PATH. The basic instructions (after doxygen is installed) are:

• cd to the simpatico/ root directory

• Enter "make html"

This should create many html files in the simpatico/doc/html directory. To begin reading the documentation, point a browser at the file simpatico/doc/html/index.html, which is the main page of the manual.

The simpatico source code is ansi standard C++, and must be compiled from source. The single-processor versions of the mcSim and mdSim MC and MD simulation program do not depend on any external libraries. Multi-processor programs (ddSim and multi-processor versions of mcSim and mdSim) require an MPI library. The build system requires the gnu version of the unix make utility (gmake) and a python interpreter.

Complete instructions for compiling various versions of simpatico are given in Sec. 2 of the online manual. Short instructions for compiling the default single-processor versions of mcSim and mdSim (for the impatient) are given below:

• Add the simpatico/bin directory to your linux command search PATH environment variable.

• Add the simpatico/scripts/python directory to your PYTHONPATH environment variable.

• cd to the simpatico/ root directory

• Enter "./setup" from the root directory to run a setup script (you only need to do this once, before compiling the first time)

• Enter "make mcMd" from the same directory

The resulting executables, named "mcSim" and "mdSim", will be installed in the simpatico bin/ directory. Please see the html documentation mentioned above for further instructions for compiling multi-processor programs that depend on MPI.

Instructions for enabling or disabling a variety of optional compile-time features are also given in the html documentation.