Homec4science
Diffusion lammps (efficient_neuronet) Browse

lammps/doc/src/99/basics.html
832f6a9f1125efficient_neuronet

Last Change
Raw File

basics.html
View Options

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META NAME="Generator" CONTENT="Cosmo Create 1.0.3">
</HEAD>
<BODY>
<H2>
Basics of Using LAMMPS</H2>
<P>
<A HREF="README.html">Return</A> to top-level of LAMMPS documentation.</P>
<UL>
<LI>
<A HREF="#_cch3_931273040">Distribution</A>
<LI>
<A HREF="#_cch3_930327142">Making LAMMPS</A>
<LI>
<A HREF="#_cch3_930327155">Running LAMMPS</A>
<LI>
<A HREF="#_cch3_930759879">Examples</A>
<LI>
<A HREF="#_cch3_931282515">Other Tools</A>
</UL>
<HR>
<H3>
<A NAME="_cch3_931273040">Distribution</A></H3>
<P>
When you unzip/untar the LAMMPS distribution you should have 5
directories: </P>
<UL>
<LI>
src = source files for LAMMPS
<LI>
doc = HTML documentation
<LI>
examples = sample problems with inputs and outputs
<LI>
msi2lmp = tool for converting files from DISCOVER to LAMMPS format
(this requires that you have DISCOVER force field files)
<LI>
tools = serial program for replicating data files
</UL>
<HR>
<H3>
<A NAME="_cch3_930327142">Making LAMMPS</A></H3>
<P>
The src directory contains the F77 and C source files for LAMMPS as
well as several sample Makefiles for different machines. To make LAMMPS
for a specfic machine, you simply type</P>
<P>
make machine</P>
<P>
from within the src directoy. E.g. &quot;make sgi&quot; or &quot;make
t3e&quot;. This should create an executable named lmp_sgi or lmp_t3e.</P>
<P>
In the src directory, there is one top-level Makefile and several
low-level machine-specific files named Makefile.xxx where xxx = the
machine name. If a low-level Makefile exists for your platform, you do
not need to edit the top-level Makefile. However you should check the
system-specific section of the low-level Makefile to make sure the
various paths are correct for your environment. If a low-level Makefile
does not exist for your platform, you will need to add a suitable
target to the top-level Makefile. You will also need to create a new
low-level Makefile using one of the existing ones as a template. If you
wish to make LAMMPS for a single-processor workstation that doesn't
have an installed MPI library, you can specify the serial target which
uses a directory of MPI stubs to link against - e.g. &quot;make
serial&quot;. You will need to make the stub library (see STUBS
directory) on your workstation before doing this.</P>
<P>
Note that the two-level Makefile system allows you to make LAMMPS for
multiple platforms. Each target creates its own object directory for
separate storage of its *.o files.</P>
<P>
There are a couple compiler switches of interest which can be specified
in the low-level Makefiles. If you use a F77FLAGS switch of -DSYNC then
synchronization calls will be made before the timing routines in
integrate.f. This may slow down the code slightly, but will make the
reported timings at the end of a run more accurate. The CCFLAGS setting
in the low-level Makefiles requires a FFT setting, for example
-DFFT_SGI or -DFFT_T3E. This is for inclusion of the appropriate
machine-specific native 1-d FFT libraries on various platforms.
Currently, the supported machines and switches (used in fft_3d.c) are
FFT_SGI, FFT_DEC, FFT_INTEL, FFT_T3E, and FFT_FFTW. The latter is a
publicly available portable FFT library, <A HREF="http://www.fftw.org">FFTW</A>,
which you can install on any machine. If none of these options is
suitable for your machine, please contact me, and we'll discuss how to
add the capability to call your machine's native FFT library.</P>
<HR>
<H3>
<A NAME="_cch3_930327155">Running LAMMPS</A></H3>
<P>
LAMMPS is run by redirecting a file of input commands into it.</P>
<P>
lmp_sgi &lt; in.lj</P>
<P>
lmp_t3e &lt; in.lj</P>
<P>
The input file contains commands that specify the parameters for the
simulation as well as read other necessary files such as a data file
that describes the initial atom positions, molecular topology, and
force-field parameters. The <A HREF="input_commands.html">input_commands</A>
page describes all the possible commands that can be used. The <A
HREF="data_format.html">data_format</A> page describes the format of
the data file. </P>
<P>
LAMMPS can be run on any number of processors, including a single
processor. In principle you should get identical answers on any number
of processors and on any machine. In practice, numerical round-off can
cause slight differences and eventual divergence of dynamical
trajectories. </P>
<P>
When LAMMPS runs, if you get an error message to the screen about
&quot;boosting&quot; something, it means one (or more) data arrays are
not allocated large enough. Some of these errors are detected at setup,
others like neighbor list overflow may not occur until the middle of a
run. When the latter happens the program will either gracefully stop
(if all processors incurred the same error) or hang (with an error
message). Unfortunately in the current version of LAMMPS which uses
static memory allocation, changing the array size(s) requires you to
edit the appropriate line(s) in the param.h file and recompile the code.</P>
<P>
I've tried to be careful about detecting memory-overflow errors in
LAMMPS. If the code ever crashes or hangs without spitting out an error
message first, it's probably a bug, so let me know about it. Of course
this applies to problems due to algorithmic or parallelism issues, not
to physics mistkaes, like specifying too big a timestep or putting 2
atoms on top of each other! One exception is that different MPI
implementations handle buffering of messages differently. If the code
hangs without an error message, it may be that you need to specify an
MPI setting or two (usually via an environment variable) to enable
buffering or boost the sizes of messages that can be buffered. </P>
<HR>
<H3>
<A NAME="_cch3_930759879">Examples</A></H3>
<P>
There are several sample problems in the examples directory. All of
them use an input file (in.*) of commands and a data file (data.*) of
initial atomic coordinates and produce one or more output files. The
*.xxx.P files are outputs on P processors on a particular machine which
you can compare your answers to.</P>
<P>
(1) lj</P>
<P>
Simple atomic simulations of Lennard-Jones atoms of 1 or 3 species with
various ensembles -- NVE, NVT, NPT.</P>
<P>
(2) charge</P>
<P>
A few timestep simulation of a box of charged atoms for testing the
Coulombic options -- cutoff, Ewald, particle-mesh Ewald (PPPM).</P>
<P>
(3) class2</P>
<P>
A simple test run of phenyalanine using DISCOVER cff95 class II force
fields.</P>
<P>
(4) min</P>
<P>
An energy minimization of a transcription protein.</P>
<P>
(5) lc</P>
<P>
Small (250 atom) and large (6750 atom) simulations of liquid crystal
molecules with various Coulombic options and periodicity settings. The
large-system date file was created by using the &quot;replicate&quot;
tool on the small-system data file.</P>
<P>
(6) flow</P>
<P>
2-d flow of Lennard-Jones atoms in a channel using various contraint
options.</P>
<P>
(7) polymer</P>
<P>
Simulations of bead-spring polymer models with one chain type and two
chain types (different size monomers). The two-chain system also has
freely diffusing monomers. This illustrates use of the setup_chain
program in the tools directory and also how to use soft potentials to
untangle the initial configurations.</P>
<HR>
<H3>
<A NAME="_cch3_931282515">Other Tools</A></H3>
<P>
The msi2lmp directory has source code for a tool that converts MSI
Discover files to LAMMPS input data files. This tool requires you to
have the Discover force-field description files in order to convert
those parameters to LAMMPS parameters. See the README file in the
msi2lmp directory for additional information.</P>
<P>
The tools directory has a C file called replicate.c which is useful for
generating new LAMMPS data files from existing ones - e.g. scaling the
atom coordinates, replicating the system to make a larger one, etc. See
the comments at the top of replicate.c for instructions on how to use
it.</P>
<P>
The tools directory has a F77 program called setup_lj (compile and link
with print.c) which can be used to generate a 3-d box of Lennard Jones
atoms (one or more atom types) like those used in examples/lj.</P>
<P>
The tools directory also has a F77 program called setup_chain.f
(compile and link with print.c) which can be used to generate random
initial polymer configurations for bead-spring models like those used
in examples/polymer. It uses an input polymer definition file (see
examples/polymer for two sample def files) that specfies how many
chains of what length, a random number seed, etc.</P>
</BODY>
</HTML>
c4science · Help