ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

git-svn-id: svn://svn.icms.temple.edu/lammps-ro/trunk@3053 f3b2605a-c512-4ea7-a41b-209d697bcdaa

Browse Source
This commit is contained in:
sjplimp
2009-08-12 16:45:55 +00:00
parent 61c84b9b46
commit 323fb937c3
16 changed files with 415 additions and 122 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
doc/Manual.html
View File

@ -97,7 +97,9 @@ listed above.
<BR>
2.7 <A HREF = "Section_start.html#2_7">Screen output</A>
<BR>
2.8 <A HREF = "Section_start.html#2_8">Tips for users of previous versions</A>
2.8 <A HREF = "Section_start.html#2_8">Running on GPUs</A>
<BR>
2.9 <A HREF = "2_9">Tips for users of previous versions</A>
<BR></UL>
<LI><A HREF = "Section_commands.html">Commands</A>

3
doc/Manual.txt
View File

@ -80,7 +80,8 @@ listed above.
2.5 "Running LAMMPS"_2_5 :b
2.6 "Command-line options"_2_6 :b
2.7 "Screen output"_2_7 :b
2.8 "Tips for users of previous versions"_2_8 :ule,b
2.8 "Running on GPUs"_2_8 :b
2.9 "Tips for users of previous versions"_2_9 :ule,b
"Commands"_Section_commands.html :l
3.1 "LAMMPS input script"_3_1 :ulb,b
3.2 "Parsing rules"_3_2 :b

19
doc/Section_commands.html
View File

@ -368,15 +368,16 @@ full description:
<TR ALIGN="center"><TD ><A HREF = "pair_colloid.html">colloid</A></TD><TD ><A HREF = "pair_coul.html">coul/cut</A></TD><TD ><A HREF = "pair_coul.html">coul/debye</A></TD><TD ><A HREF = "pair_coul.html">coul/long</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_dipole.html">dipole/cut</A></TD><TD ><A HREF = "pair_dpd.html">dpd</A></TD><TD ><A HREF = "pair_eam.html">eam</A></TD><TD ><A HREF = "pair_eam.html">eam/opt</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_eam.html">eam/alloy</A></TD><TD ><A HREF = "pair_eam.html">eam/alloy/opt</A></TD><TD ><A HREF = "pair_eam.html">eam/fs</A></TD><TD ><A HREF = "pair_eam.html">eam/fs/opt</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_gayberne.html">gayberne</A></TD><TD ><A HREF = "pair_gran.html">gran/hertz/history</A></TD><TD ><A HREF = "pair_gran.html">gran/hooke</A></TD><TD ><A HREF = "pair_gran.html">gran/hooke/history</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/charmm</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/charmm/implicit</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/long</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/long/opt</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_class2.html">lj/class2</A></TD><TD ><A HREF = "pair_class2.html">lj/class2/coul/cut</A></TD><TD ><A HREF = "pair_class2.html">lj/class2/coul/long</A></TD><TD ><A HREF = "pair_lj.html">lj/cut</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lj.html">lj/cut/opt</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/cut</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/debye</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/long</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lj.html">lj/cut/coul/long/tip4p</A></TD><TD ><A HREF = "pair_lj_expand.html">lj/expand</A></TD><TD ><A HREF = "pair_gromacs.html">lj/gromacs</A></TD><TD ><A HREF = "pair_gromacs.html">lj/gromacs/coul/gromacs</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lj_smooth.html">lj/smooth</A></TD><TD ><A HREF = "pair_lj96_cut.html">lj96/cut</A></TD><TD ><A HREF = "pair_lubricate.html">lubricate</A></TD><TD ><A HREF = "pair_meam.html">meam</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_morse.html">morse</A></TD><TD ><A HREF = "pair_morse.html">morse/opt</A></TD><TD ><A HREF = "pair_peri_pmb.html">peri/pmb</A></TD><TD ><A HREF = "pair_reax.html">reax</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_resquared.html">resquared</A></TD><TD ><A HREF = "pair_soft.html">soft</A></TD><TD ><A HREF = "pair_sw.html">sw</A></TD><TD ><A HREF = "pair_table.html">table</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_tersoff.html">tersoff</A></TD><TD ><A HREF = "pair_tersoff_zbl.html">tersoff/zbl</A></TD><TD ><A HREF = "pair_yukawa.html">yukawa</A>
<TR ALIGN="center"><TD ><A HREF = "pair_gayberne.html">gayberne</A></TD><TD ><A HREF = "pair_gayberne.html">gayberne/gpu</A></TD><TD ><A HREF = "pair_gran.html">gran/hertz/history</A></TD><TD ><A HREF = "pair_gran.html">gran/hooke</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_gran.html">gran/hooke/history</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/charmm</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/charmm/implicit</A></TD><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/long</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_charmm.html">lj/charmm/coul/long/opt</A></TD><TD ><A HREF = "pair_class2.html">lj/class2</A></TD><TD ><A HREF = "pair_class2.html">lj/class2/coul/cut</A></TD><TD ><A HREF = "pair_class2.html">lj/class2/coul/long</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lj.html">lj/cut</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/gpu</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/opt</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/cut</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lj.html">lj/cut/coul/debye</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/long</A></TD><TD ><A HREF = "pair_lj.html">lj/cut/coul/long/tip4p</A></TD><TD ><A HREF = "pair_lj_expand.html">lj/expand</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_gromacs.html">lj/gromacs</A></TD><TD ><A HREF = "pair_gromacs.html">lj/gromacs/coul/gromacs</A></TD><TD ><A HREF = "pair_lj_smooth.html">lj/smooth</A></TD><TD ><A HREF = "pair_lj96_cut.html">lj96/cut</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_lubricate.html">lubricate</A></TD><TD ><A HREF = "pair_meam.html">meam</A></TD><TD ><A HREF = "pair_morse.html">morse</A></TD><TD ><A HREF = "pair_morse.html">morse/opt</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_peri_pmb.html">peri/pmb</A></TD><TD ><A HREF = "pair_reax.html">reax</A></TD><TD ><A HREF = "pair_resquared.html">resquared</A></TD><TD ><A HREF = "pair_soft.html">soft</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_sw.html">sw</A></TD><TD ><A HREF = "pair_table.html">table</A></TD><TD ><A HREF = "pair_tersoff.html">tersoff</A></TD><TD ><A HREF = "pair_tersoff_zbl.html">tersoff/zbl</A></TD></TR>
<TR ALIGN="center"><TD ><A HREF = "pair_yukawa.html">yukawa</A>
</TD></TR></TABLE></DIV>
<P>These are pair styles contributed by users, which can be used if

2
doc/Section_commands.txt
View File

@ -513,6 +513,7 @@ full description:
"eam/fs"_pair_eam.html,
"eam/fs/opt"_pair_eam.html,
"gayberne"_pair_gayberne.html,
"gayberne/gpu"_pair_gayberne.html,
"gran/hertz/history"_pair_gran.html,
"gran/hooke"_pair_gran.html,
"gran/hooke/history"_pair_gran.html,
@ -524,6 +525,7 @@ full description:
"lj/class2/coul/cut"_pair_class2.html,
"lj/class2/coul/long"_pair_class2.html,
"lj/cut"_pair_lj.html,
"lj/cut/gpu"_pair_lj.html,
"lj/cut/opt"_pair_lj.html,
"lj/cut/coul/cut"_pair_lj.html,
"lj/cut/coul/debye"_pair_lj.html,

144
doc/Section_start.html
View File

@ -21,7 +21,8 @@ new and experienced users.
2.5 <A HREF = "#2_5">Running LAMMPS</A><BR>
2.6 <A HREF = "#2_6">Command-line options</A><BR>
2.7 <A HREF = "#2_7">Screen output</A><BR>
2.8 <A HREF = "#2_8">Tips for users of previous versions</A> <BR>
2.8 <A HREF = "#2_8">Running on GPUs</A><BR>
2.9 <A HREF = "#2_9">Tips for users of previous versions</A> <BR>
<HR>
@ -291,6 +292,7 @@ typing "make package".
<TR><TD >colloid </TD><TD > colloidal particle force fields</TD></TR>
<TR><TD >dipole </TD><TD > point dipole particles and force fields</TD></TR>
<TR><TD >dpd </TD><TD > dissipative particle dynamics (DPD) force field</TD></TR>
<TR><TD >gpu </TD><TD > GPU-enabled force field styles</TD></TR>
<TR><TD >granular </TD><TD > force fields and boundary conditions for granular systems</TD></TR>
<TR><TD >kspace </TD><TD > long-range Ewald and particle-mesh (PPPM) solvers</TD></TR>
<TR><TD >manybody </TD><TD > metal, 3-body, bond-order potentials</TD></TR>
@ -322,13 +324,21 @@ documentation.
</P>
<HR>
<P>Any or all packages can be included or excluded when LAMMPS is built.
The one exception is that to use the standard "opt" package, you must
also be using the "molecule" and "manybody" packages. You may wish to
exclude certain packages if you will never run certain kinds of
simulations. This will keep you from having to build auxiliary
libraries (see below) and will produce a smaller executable which may
run a bit faster.
<P>Any or all packages can be included or excluded independently when
LAMMPS is built.
</P>
<P>The two exceptions to this are the "gpu" and "opt" packages. Some of
the files in these packages require other packages to also be
included. If this is not the case, then those subsidiary files in
"gpu" and "opt" will not be installed either. To install all the
files in package "gpu", the "asphere" package must also be installed.
To install all the files in package "opt", the "kspace" and "manybody"
packages must also be installed.
</P>
<P>You may wish to exclude certain packages if you will never run certain
kinds of simulations. This will keep you from having to build
auxiliary libraries (see below) and will produce a smaller executable
which may run a bit faster.
</P>
<P>By default, LAMMPS includes only the "kspace", "manybody", and
"molecule" packages. As described below, some standard packages
@ -361,7 +371,15 @@ various options.
<P>A few packages require that additional libraries be compiled first,
which LAMMPS will link to when it builds. These are stored under
under the lammps/lib directory of the distribution. Currently, these
include libraries for the "meam", "poems", and "reax" packages.
include libraries for the "gpu", "meam", "poems", and "reax" packages.
You should look at the README files in the lib directories
(e.g. lib/reax/README) for instructions on how to build each library.
</P>
<P>The "gpu" library in lib/gpu contains code to enable portions of
LAMMPS to run on a GPU chip associated with your CPU. Currently, only
NVIDIA GPUs are supported. Building this library requires NVIDIA Cuda
tools to be installed on your system. See the <A HREF = "#2_8">Running on GPUs</A>
section below for more info about installing and using Cuda.
</P>
<P>The "meam" library in lib/meam computes the modified embedded atom
method potential, which is a generalization of EAM potentials that can
@ -393,19 +411,18 @@ Different definitions of this macro can be obtained by adding a
machine-specific macro definition to the CCFLAGS variable in your Makefile
e.g. -D_IBM. See pair_reax_fortran.h for more info.
</P>
<P>You should look at the README files in the lib directories
(e.g. lib/reax/README) for instructions on how to build each library.
In each cases, the library is built by typing something like
<P>As its README file describes, for each library, the library is built
by typing something like
</P>
<PRE>make -f Makefile.foo
</PRE>
<P>in the appropriate directory, e.g. in lib/reax.
</P>
<P>You should use the Makefile that is a match for your system. In the
case of libraries that consist of Fotran code, this requires your
system have a Fortran compiler, the settings for which will be in the
Makefile. If one of the provided Makefiles is not appropriate for
your system you can edit or add one as needed.
<P>You must use a Makefile that is a match for your system. If one of
the provided Makefiles is not appropriate for your system you will
need to edit or add one. For example, in the case of libraries of
Fortran code, your system must have a Fortran compiler, the settings
for which will be in the Makefile.
</P>
<P>Once the desired library or libraries are built, you can build LAMMPS.
From the lammps/src directory type this, for example, to build LAMMPS
@ -414,16 +431,22 @@ with ReaxFF:
<PRE>make yes-reax
make g++
</PRE>
<P>Note that building the library (reax in this case) is not sufficient
to use it from LAMMPS. You must also include the package that
uses/wraps the library before you build LAMMPS itself.
<P>Note that simply building the library (reax in this case), which must
be done first, is not sufficient to use it from LAMMPS. You must also
include the package that uses and wraps the library before you build
LAMMPS itself.
</P>
<P>Also note that to use a package and library that is Fortran based
(like meam or reax), the lo-level Makefile in lammps/src/MAKE, which
is Makefile.g++ in this example, must typically have settings for
LINKFORT and FORTLIB that are specific to the Fortran compiler
installed on your system. This is so that the C++ compiler can
perform a cross-language link using the appropriate Fortran libraries.
<P>To use the gpu package and library, the lo-level Makefile in
lammps/src/MAKE, such as Makefile.g++, must typically have settings
for LINKGPU and GPULIB that are specific to the NVIDIA software
installed on your system.
</P>
<P>To use the meam or reax packages and libraries that are Fortran based,
the lo-level Makefile in lammps/src/MAKE, such as Makefile.g++, must
typically have settings for LINKFORT and FORTLIB that are specific to
the Fortran compiler installed on your system. This is so that the
C++ compiler can perform a cross-language link using the appropriate
Fortran libraries.
</P>
<HR>
@ -723,41 +746,74 @@ communication, roughly 75% in the example above.
</P>
<HR>
<H4><A NAME = "2_8"></A>2.8 Tips for users of previous LAMMPS versions
<H4><A NAME = "2_8"></A>2.8 Running on GPUs
</H4>
<P>LAMMPS 2003 is a complete C++ rewrite of LAMMPS 2001, which was
written in F90. Features of earlier versions of LAMMPS are listed in
<A HREF = "Section_history.html">this section</A>. The F90 and F77 versions (2001
and 99) are also freely distributed as open-source codes; check the
<A HREF = "http://lammps.sandia.gov">LAMMPS WWW Site</A> for distribution information if you prefer those
versions. The 99 and 2001 versions are no longer under active
development; they do not have all the features of LAMMPS 2003.
<P> 1. I assume you have nvidia card ( cat /proc/driver/nvidia/cards/0 )
2. Go to http://www.nvidia.com/object/cuda_get.html
3. Install latest driver and toolkit (SDK is not necessary)
4. Run make in lib/gpu
5. ./nvc_get_devices will list supported devices and properties
</P>
<P>The pair styles supporting GPU acceleration have optional arguments to
determine how GPUs are selected. Each GPU is selected based on its ID
number reported by the graphics driver. A list of supported GPUs can be
obtained using the 'nvc_get_devices' executable that is built with the
library. To change how GPUs are selected, a GPU selection mode and a GPU
selection parameter are supplied. If the mode is "one_node", the GPU ID
is equal to the process rank plus the GPU parameter. For example,
</P>
<P>pair_style lj/cut/gpu 2.5 one_node 0
</P>
<P>will select GPUs 0, 1, and 2 for a 3 process run. If the mode is
"one_gpu", every process will select the same GPU id identified by the
GPU parameter. This option can be used to select a single GPU on
multiple nodes and should be run with exactly 1 process per node. If the
mode is "multi_gpu" the GPU parameter supplied should be the number of
GPUs per node. The GPU selected by a particular process will then be
equal to the rank%gpus_per_node. This final option can be used to run on
multiple nodes that contain multiple GPUs per node. It assumes that
processes will be filled in order by slots such that the ranks on any
given node are always sequential.
</P>
<P>double/single precision
atom limits
</P>
<HR>
<H4><A NAME = "2_9"></A>2.9 Tips for users of previous LAMMPS versions
</H4>
<P>The current C++ began with a complete rewrite of LAMMPS 2001, which
was written in F90. Features of earlier versions of LAMMPS are listed
in <A HREF = "Section_history.html">this section</A>. The F90 and F77 versions
(2001 and 99) are also freely distributed as open-source codes; check
the <A HREF = "http://lammps.sandia.gov">LAMMPS WWW Site</A> for distribution information if you prefer
those versions. The 99 and 2001 versions are no longer under active
development; they do not have all the features of C++ LAMMPS.
</P>
<P>If you are a previous user of LAMMPS 2001, these are the most
significant changes you will notice in LAMMPS 2003:
significant changes you will notice in C++ LAMMPS:
</P>
<P>(1) The names and arguments of many input script commands have
changed. All commands are now a single word (e.g. read_data instead
of read data).
</P>
<P>(2) All the functionality of LAMMPS 2001 is included in LAMMPS 2003,
<P>(2) All the functionality of LAMMPS 2001 is included in C++ LAMMPS,
but you may need to specify the relevant commands in different ways.
</P>
<P>(3) The format of the data file can be streamlined for some problems.
See the <A HREF = "read_data.html">read_data</A> command for details. The data file
section "Nonbond Coeff" has been renamed to "Pair Coeff" in LAMMPS
2003.
section "Nonbond Coeff" has been renamed to "Pair Coeff" in C++ LAMMPS.
</P>
<P>(4) Binary restart files written by LAMMPS 2001 cannot be read by
LAMMPS 2003 with a <A HREF = "read_restart.html">read_restart</A> command. This is
<P>(4) Binary restart files written by LAMMPS 2001 cannot be read by C++
LAMMPS with a <A HREF = "read_restart.html">read_restart</A> command. This is
because they were output by F90 which writes in a different binary
format than C or C++ writes or reads. Use the <I>restart2data</I> tool
provided with LAMMPS 2001 to convert the 2001 restart file to a text
data file. Then edit the data file as necessary before using the
LAMMPS 2003 <A HREF = "read_data.html">read_data</A> command to read it in.
data file. Then edit the data file as necessary before using the C++
LAMMPS <A HREF = "read_data.html">read_data</A> command to read it in.
</P>
<P>(5) There are numerous small numerical changes in LAMMPS 2003 that
mean you will not get identical answers when comparing to a 2001 run.
<P>(5) There are numerous small numerical changes in C++ LAMMPS that mean
you will not get identical answers when comparing to a 2001 run.
However, your initial thermodynamic energy and MD trajectory should be
close if you have setup the problem for both codes the same.
</P>

144
doc/Section_start.txt
View File

@ -18,7 +18,8 @@ new and experienced users.
2.5 "Running LAMMPS"_#2_5
2.6 "Command-line options"_#2_6
2.7 "Screen output"_#2_7
2.8 "Tips for users of previous versions"_#2_8 :all(b)
2.8 "Running on GPUs"_#2_8
2.9 "Tips for users of previous versions"_#2_9 :all(b)
:line
@ -285,6 +286,7 @@ class2 : class 2 force fields
colloid : colloidal particle force fields
dipole : point dipole particles and force fields
dpd : dissipative particle dynamics (DPD) force field
gpu : GPU-enabled force field styles
granular : force fields and boundary conditions for granular systems
kspace : long-range Ewald and particle-mesh (PPPM) solvers
manybody : metal, 3-body, bond-order potentials
@ -315,13 +317,21 @@ documentation.
:line
Any or all packages can be included or excluded when LAMMPS is built.
The one exception is that to use the standard "opt" package, you must
also be using the "molecule" and "manybody" packages. You may wish to
exclude certain packages if you will never run certain kinds of
simulations. This will keep you from having to build auxiliary
libraries (see below) and will produce a smaller executable which may
run a bit faster.
Any or all packages can be included or excluded independently when
LAMMPS is built.
The two exceptions to this are the "gpu" and "opt" packages. Some of
the files in these packages require other packages to also be
included. If this is not the case, then those subsidiary files in
"gpu" and "opt" will not be installed either. To install all the
files in package "gpu", the "asphere" package must also be installed.
To install all the files in package "opt", the "kspace" and "manybody"
packages must also be installed.
You may wish to exclude certain packages if you will never run certain
kinds of simulations. This will keep you from having to build
auxiliary libraries (see below) and will produce a smaller executable
which may run a bit faster.
By default, LAMMPS includes only the "kspace", "manybody", and
"molecule" packages. As described below, some standard packages
@ -354,7 +364,15 @@ various options.
A few packages require that additional libraries be compiled first,
which LAMMPS will link to when it builds. These are stored under
under the lammps/lib directory of the distribution. Currently, these
include libraries for the "meam", "poems", and "reax" packages.
include libraries for the "gpu", "meam", "poems", and "reax" packages.
You should look at the README files in the lib directories
(e.g. lib/reax/README) for instructions on how to build each library.
The "gpu" library in lib/gpu contains code to enable portions of
LAMMPS to run on a GPU chip associated with your CPU. Currently, only
NVIDIA GPUs are supported. Building this library requires NVIDIA Cuda
tools to be installed on your system. See the "Running on GPUs"_#2_8
section below for more info about installing and using Cuda.
The "meam" library in lib/meam computes the modified embedded atom
method potential, which is a generalization of EAM potentials that can
@ -386,19 +404,18 @@ Different definitions of this macro can be obtained by adding a
machine-specific macro definition to the CCFLAGS variable in your Makefile
e.g. -D_IBM. See pair_reax_fortran.h for more info.
You should look at the README files in the lib directories
(e.g. lib/reax/README) for instructions on how to build each library.
In each cases, the library is built by typing something like
As its README file describes, for each library, the library is built
by typing something like
make -f Makefile.foo :pre
in the appropriate directory, e.g. in lib/reax.
You should use the Makefile that is a match for your system. In the
case of libraries that consist of Fotran code, this requires your
system have a Fortran compiler, the settings for which will be in the
Makefile. If one of the provided Makefiles is not appropriate for
your system you can edit or add one as needed.
You must use a Makefile that is a match for your system. If one of
the provided Makefiles is not appropriate for your system you will
need to edit or add one. For example, in the case of libraries of
Fortran code, your system must have a Fortran compiler, the settings
for which will be in the Makefile.
Once the desired library or libraries are built, you can build LAMMPS.
From the lammps/src directory type this, for example, to build LAMMPS
@ -407,16 +424,22 @@ with ReaxFF:
make yes-reax
make g++ :pre
Note that building the library (reax in this case) is not sufficient
to use it from LAMMPS. You must also include the package that
uses/wraps the library before you build LAMMPS itself.
Note that simply building the library (reax in this case), which must
be done first, is not sufficient to use it from LAMMPS. You must also
include the package that uses and wraps the library before you build
LAMMPS itself.
Also note that to use a package and library that is Fortran based
(like meam or reax), the lo-level Makefile in lammps/src/MAKE, which
is Makefile.g++ in this example, must typically have settings for
LINKFORT and FORTLIB that are specific to the Fortran compiler
installed on your system. This is so that the C++ compiler can
perform a cross-language link using the appropriate Fortran libraries.
To use the gpu package and library, the lo-level Makefile in
lammps/src/MAKE, such as Makefile.g++, must typically have settings
for LINKGPU and GPULIB that are specific to the NVIDIA software
installed on your system.
To use the meam or reax packages and libraries that are Fortran based,
the lo-level Makefile in lammps/src/MAKE, such as Makefile.g++, must
typically have settings for LINKFORT and FORTLIB that are specific to
the Fortran compiler installed on your system. This is so that the
C++ compiler can perform a cross-language link using the appropriate
Fortran libraries.
:line
@ -716,40 +739,73 @@ communication, roughly 75% in the example above.
:line
2.8 Tips for users of previous LAMMPS versions :h4,link(2_8)
2.8 Running on GPUs :h4,link(2_8)
LAMMPS 2003 is a complete C++ rewrite of LAMMPS 2001, which was
written in F90. Features of earlier versions of LAMMPS are listed in
"this section"_Section_history.html. The F90 and F77 versions (2001
and 99) are also freely distributed as open-source codes; check the
"LAMMPS WWW Site"_lws for distribution information if you prefer those
versions. The 99 and 2001 versions are no longer under active
development; they do not have all the features of LAMMPS 2003.
1. I assume you have nvidia card ( cat /proc/driver/nvidia/cards/0 )
2. Go to http://www.nvidia.com/object/cuda_get.html
3. Install latest driver and toolkit (SDK is not necessary)
4. Run make in lib/gpu
5. ./nvc_get_devices will list supported devices and properties
The pair styles supporting GPU acceleration have optional arguments to
determine how GPUs are selected. Each GPU is selected based on its ID
number reported by the graphics driver. A list of supported GPUs can be
obtained using the 'nvc_get_devices' executable that is built with the
library. To change how GPUs are selected, a GPU selection mode and a GPU
selection parameter are supplied. If the mode is "one_node", the GPU ID
is equal to the process rank plus the GPU parameter. For example,
pair_style lj/cut/gpu 2.5 one_node 0
will select GPUs 0, 1, and 2 for a 3 process run. If the mode is
"one_gpu", every process will select the same GPU id identified by the
GPU parameter. This option can be used to select a single GPU on
multiple nodes and should be run with exactly 1 process per node. If the
mode is "multi_gpu" the GPU parameter supplied should be the number of
GPUs per node. The GPU selected by a particular process will then be
equal to the rank%gpus_per_node. This final option can be used to run on
multiple nodes that contain multiple GPUs per node. It assumes that
processes will be filled in order by slots such that the ranks on any
given node are always sequential.
double/single precision
atom limits
:line
2.9 Tips for users of previous LAMMPS versions :h4,link(2_9)
The current C++ began with a complete rewrite of LAMMPS 2001, which
was written in F90. Features of earlier versions of LAMMPS are listed
in "this section"_Section_history.html. The F90 and F77 versions
(2001 and 99) are also freely distributed as open-source codes; check
the "LAMMPS WWW Site"_lws for distribution information if you prefer
those versions. The 99 and 2001 versions are no longer under active
development; they do not have all the features of C++ LAMMPS.
If you are a previous user of LAMMPS 2001, these are the most
significant changes you will notice in LAMMPS 2003:
significant changes you will notice in C++ LAMMPS:
(1) The names and arguments of many input script commands have
changed. All commands are now a single word (e.g. read_data instead
of read data).
(2) All the functionality of LAMMPS 2001 is included in LAMMPS 2003,
(2) All the functionality of LAMMPS 2001 is included in C++ LAMMPS,
but you may need to specify the relevant commands in different ways.
(3) The format of the data file can be streamlined for some problems.
See the "read_data"_read_data.html command for details. The data file
section "Nonbond Coeff" has been renamed to "Pair Coeff" in LAMMPS
2003.
section "Nonbond Coeff" has been renamed to "Pair Coeff" in C++ LAMMPS.
(4) Binary restart files written by LAMMPS 2001 cannot be read by
LAMMPS 2003 with a "read_restart"_read_restart.html command. This is
(4) Binary restart files written by LAMMPS 2001 cannot be read by C++
LAMMPS with a "read_restart"_read_restart.html command. This is
because they were output by F90 which writes in a different binary
format than C or C++ writes or reads. Use the {restart2data} tool
provided with LAMMPS 2001 to convert the 2001 restart file to a text
data file. Then edit the data file as necessary before using the
LAMMPS 2003 "read_data"_read_data.html command to read it in.
data file. Then edit the data file as necessary before using the C++
LAMMPS "read_data"_read_data.html command to read it in.
(5) There are numerous small numerical changes in LAMMPS 2003 that
mean you will not get identical answers when comparing to a 2001 run.
(5) There are numerous small numerical changes in C++ LAMMPS that mean
you will not get identical answers when comparing to a 2001 run.
However, your initial thermodynamic energy and MD trajectory should be
close if you have setup the problem for both codes the same.

2
doc/dump_modify.html
View File

@ -75,8 +75,6 @@ current coordinate. Note that for dump style <I>custom</I> these various
values can be printed in the dump file by using the appropriate atom
attributes in the dump command itself.
</P>
<P>The <I>flush</I> option determines whether a flush operation in invoked
after a dump snapshot is written to the dump file. A flush insures
the output in that file is current (no buffering by the OS), even if

2
doc/dump_modify.txt
View File

@ -68,8 +68,6 @@ current coordinate. Note that for dump style {custom} these various
values can be printed in the dump file by using the appropriate atom
attributes in the dump command itself.
:link(pizza,http://www.cs.sandia.gov/~sjplimp/pizza.html)
The {flush} option determines whether a flush operation in invoked
after a dump snapshot is written to the dump file. A flush insures
the output in that file is current (no buffering by the OS), even if

2
doc/pair_coeff.html
View File

@ -106,6 +106,7 @@ the pair_style command, and coefficients specified by the associated
<LI><A HREF = "pair_eam.html">pair_style eam/fs</A> - Finnis-Sinclair EAM
<LI><A HREF = "pair_eam.html">pair_style eam/fs/opt</A> - optimized version of Finnis-Sinclair EAM
<LI><A HREF = "pair_gayberne.html">pair_style gayberne</A> - Gay-Berne ellipsoidal potential
<LI><A HREF = "pair_gayberne.html">pair_style gayberne/gpu</A> - GPU-enabled Gay-Berne ellipsoidal potential
<LI><A HREF = "pair_gran.html">pair_style gran/hertz/history</A> - granular potential with Hertzian interactions
<LI><A HREF = "pair_gran.html">pair_style gran/hooke</A> - granular potential with history effects
<LI><A HREF = "pair_gran.html">pair_style gran/hooke/history</A> - granular potential without history effects
@ -117,6 +118,7 @@ the pair_style command, and coefficients specified by the associated
<LI><A HREF = "pair_class2.html">pair_style lj/class2/coul/cut</A> - COMPASS with cutoff Coulomb
<LI><A HREF = "pair_class2.html">pair_style lj/class2/coul/long</A> - COMPASS with long-range Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut</A> - cutoff Lennard-Jones potential with no Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut/gpu</A> - GPU-enabled version of cutoff LJ
<LI><A HREF = "pair_lj.html">pair_style lj/cut/opt</A> - optimized version of cutoff LJ
<LI><A HREF = "pair_lj.html">pair_style lj/cut/coul/cut</A> - LJ with cutoff Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut/coul/debye</A> - LJ with Debye screening added to Coulomb

2
doc/pair_coeff.txt
View File

@ -102,6 +102,7 @@ the pair_style command, and coefficients specified by the associated
"pair_style eam/fs"_pair_eam.html - Finnis-Sinclair EAM
"pair_style eam/fs/opt"_pair_eam.html - optimized version of Finnis-Sinclair EAM
"pair_style gayberne"_pair_gayberne.html - Gay-Berne ellipsoidal potential
"pair_style gayberne/gpu"_pair_gayberne.html - GPU-enabled Gay-Berne ellipsoidal potential
"pair_style gran/hertz/history"_pair_gran.html - granular potential with Hertzian interactions
"pair_style gran/hooke"_pair_gran.html - granular potential with history effects
"pair_style gran/hooke/history"_pair_gran.html - granular potential without history effects
@ -113,6 +114,7 @@ the pair_style command, and coefficients specified by the associated
"pair_style lj/class2/coul/cut"_pair_class2.html - COMPASS with cutoff Coulomb
"pair_style lj/class2/coul/long"_pair_class2.html - COMPASS with long-range Coulomb
"pair_style lj/cut"_pair_lj.html - cutoff Lennard-Jones potential with no Coulomb
"pair_style lj/cut/gpu"_pair_lj.html - GPU-enabled version of cutoff LJ
"pair_style lj/cut/opt"_pair_lj.html - optimized version of cutoff LJ
"pair_style lj/cut/coul/cut"_pair_lj.html - LJ with cutoff Coulomb
"pair_style lj/cut/coul/debye"_pair_lj.html - LJ with Debye screening added to Coulomb

61
doc/pair_gayberne.html
View File

@ -11,11 +11,18 @@
<H3>pair_style gayberne command
</H3>
<H3>pair_style gayberne/gpu command
</H3>
<P><B>Syntax:</B>
</P>
<PRE>pair_style gayberne gamma upsilon mu cutoff
</PRE>
<UL><LI>gamma = shift for potential minimum (typically 1)
<PRE>pair_style gayberne/gpu gpuflag gpunum gamma upsilon mu cutoff
</PRE>
<UL><LI>style = <I>gayberne</I> or <I>gayberne/gpu</I>
<LI>gpuflag = <I>one/node</I> or <I>one/gpu</I> or <I>multi/gpu</I>, only used with gayberne/gpu
<LI>gpunum = gpu ID or quantity, only used with gayberne/gpu
<LI>gamma = shift for potential minimum (typically 1)
<LI>upsilon = exponent for eta orientation-dependent energy function
<LI>mu = exponent for chi orientation-dependent energy function
<LI>cutoff = global cutoff for interactions (distance units)
@ -23,11 +30,12 @@
<P><B>Examples:</B>
</P>
<PRE>pair_style gayberne 1.0 1.0 1.0 10.0
pair_style gayberne/gpu 1.0 1.0 1.0 10.0
pair_coeff * * 1.0 1.7 1.7 3.4 3.4 1.0 1.0 1.0
</PRE>
<P><B>Description:</B>
</P>
<P>Style <I>gayberne</I> computes a Gay-Berne anisotropic LJ interaction
<P>The <I>gayberne</I> styles compute a Gay-Berne anisotropic LJ interaction
<A HREF = "#Berardi">(Berardi)</A> between pairs of ellipsoidal particles or an
ellipsoidal and spherical particle via the formulas
</P>
@ -42,6 +50,11 @@ both particles are spherical, the formula reduces to the usual
Lennard-Jones interaction (see details below for when Gay-Berne treats
a particle as "spherical").
</P>
<P>Style <I>gayberne/gpu</I> is a GPU-enabled version of style <I>gayberne</I> that
should give identical answers. Depending on system size and the GPU
processor you have on your system, it may be 100x faster (for the
pairwise portion of the run time). See more details below.
</P>
<P>For large uniform molecules it has been shown that the energy
parameters are approximately representable in terms of local contact
curvatures <A HREF = "#Everaers">(Everaers)</A>:
@ -124,6 +137,45 @@ to specify its interaction with other spherical particles.
</P>
<HR>
<P>The <I>gayberne/gpu</I> style is identical to the <I>gayberne</I> style, except
that each processor off-loads its pairwise calculations to a GPU chip.
Depending on the hardware available on your system this can provide a
significant speed-up, espcially for the relatively expensive
computations inherent in Gay-Berne interactions. See the <A HREF = "Section_start.html#2_8">Running on
GPUs</A> section of the manual for more details
about hardware and software requirements for using GPUs.
</P>
<P>The <I>gpuflag</I> and <I>gpunum</I> settings in the pair_style command refer to
how the GPUs on your system are configured.
</P>
<P>Set <I>gpuflag</I> to <I>one/node</I> if you have multiple gpus on 1 CPU node of
your system. In this case, <I>gpunum</I> should be the starting GPU ID.
</P>
<P>Set <I>gpuflag</I> to <I>one/node</I> if you have multiple gpus on 1 node
of your system. In this case, <I>gpunum</I> should be the starting GPU ID.
</P>
<P> // set multi_gpu_mode to one/gpu to select the same gpu id on every node
// -- param is id of gpu
// set multi_gpu_mode to multi/gpu to get ma
// -- param is number of gpus per node
</P>
<P>Additional requirements in your input script to run with style
<I>gayberne/gpu</I> are as follows:
</P>
<P>The <A HREF = "newton.html">newton pair</A> setting must be <I>off</I>.
</P>
<P>You should use the <A HREF = "neigh_modify.html">neigh_modify one</A> command and
set its value to something close (but slightly larger) than the number
of pairwise neighbors/atom you expect to have in your model. This is
a function of the pairwise cutoff. Note that the default for this
setting is 2000, which is much larger than most models need. Unlike
neighbor lists in LAMMPS itself, the GPU version of this pair style
uses that setting to allocate memory on the GPU for neighbor
information. If the setting is too large, it will limit the number of
atoms that can be stored on the GPU.
</P>
<HR>
<P><B>Mixing, shift, table, tail correction, restart, rRESPA info</B>:
</P>
<P>For atom type pairs I,J and I != J, the epsilon and sigma coefficients
@ -155,8 +207,9 @@ to be specified in an input script that reads a restart file.
<P><B>Restrictions:</B>
</P>
<P>This style is part of the "asphere" package. It is only enabled if
LAMMPS was built with that package. See the <A HREF = "Section_start.html#2_3">Making
<P>The <I>gayberne</I> style is part of the "asphere" package. The
<I>gayberne/gpu</I> style is part of the "gpu" package. They are only
enabled if LAMMPS was built with the those packages. See the <A HREF = "Section_start.html#2_3">Making
LAMMPS</A> section for more info.
</P>
<P>This pair style requires that atoms store torque and a quaternion to

57
doc/pair_gayberne.txt
View File

@ -7,11 +7,16 @@
:line
pair_style gayberne command :h3
pair_style gayberne/gpu command :h3
[Syntax:]
pair_style gayberne gamma upsilon mu cutoff :pre
pair_style gayberne/gpu gpuflag gpunum gamma upsilon mu cutoff :pre
style = {gayberne} or {gayberne/gpu}
gpuflag = {one/node} or {one/gpu} or {multi/gpu}, only used with gayberne/gpu
gpunum = gpu ID or quantity, only used with gayberne/gpu
gamma = shift for potential minimum (typically 1)
upsilon = exponent for eta orientation-dependent energy function
mu = exponent for chi orientation-dependent energy function
@ -20,11 +25,12 @@ cutoff = global cutoff for interactions (distance units) :ul
[Examples:]
pair_style gayberne 1.0 1.0 1.0 10.0
pair_style gayberne/gpu 1.0 1.0 1.0 10.0
pair_coeff * * 1.0 1.7 1.7 3.4 3.4 1.0 1.0 1.0 :pre
[Description:]
Style {gayberne} computes a Gay-Berne anisotropic LJ interaction
The {gayberne} styles compute a Gay-Berne anisotropic LJ interaction
"(Berardi)"_#Berardi between pairs of ellipsoidal particles or an
ellipsoidal and spherical particle via the formulas
@ -39,6 +45,11 @@ both particles are spherical, the formula reduces to the usual
Lennard-Jones interaction (see details below for when Gay-Berne treats
a particle as "spherical").
Style {gayberne/gpu} is a GPU-enabled version of style {gayberne} that
should give identical answers. Depending on system size and the GPU
processor you have on your system, it may be 100x faster (for the
pairwise portion of the run time). See more details below.
For large uniform molecules it has been shown that the energy
parameters are approximately representable in terms of local contact
curvatures "(Everaers)"_#Everaers:
@ -121,6 +132,45 @@ to specify its interaction with other spherical particles.
:line
The {gayberne/gpu} style is identical to the {gayberne} style, except
that each processor off-loads its pairwise calculations to a GPU chip.
Depending on the hardware available on your system this can provide a
significant speed-up, espcially for the relatively expensive
computations inherent in Gay-Berne interactions. See the "Running on
GPUs"_Section_start.html#2_8 section of the manual for more details
about hardware and software requirements for using GPUs.
The {gpuflag} and {gpunum} settings in the pair_style command refer to
how the GPUs on your system are configured.
Set {gpuflag} to {one/node} if you have multiple gpus on 1 CPU node of
your system. In this case, {gpunum} should be the starting GPU ID.
Set {gpuflag} to {one/node} if you have multiple gpus on 1 node
of your system. In this case, {gpunum} should be the starting GPU ID.
// set multi_gpu_mode to one/gpu to select the same gpu id on every node
// -- param is id of gpu
// set multi_gpu_mode to multi/gpu to get ma
// -- param is number of gpus per node
Additional requirements in your input script to run with style
{gayberne/gpu} are as follows:
The "newton pair"_newton.html setting must be {off}.
You should use the "neigh_modify one"_neigh_modify.html command and
set its value to something close (but slightly larger) than the number
of pairwise neighbors/atom you expect to have in your model. This is
a function of the pairwise cutoff. Note that the default for this
setting is 2000, which is much larger than most models need. Unlike
neighbor lists in LAMMPS itself, the GPU version of this pair style
uses that setting to allocate memory on the GPU for neighbor
information. If the setting is too large, it will limit the number of
atoms that can be stored on the GPU.
:line
[Mixing, shift, table, tail correction, restart, rRESPA info]:
For atom type pairs I,J and I != J, the epsilon and sigma coefficients
@ -152,8 +202,9 @@ This pair style can only be used via the {pair} keyword of the
[Restrictions:]
This style is part of the "asphere" package. It is only enabled if
LAMMPS was built with that package. See the "Making
The {gayberne} style is part of the "asphere" package. The
{gayberne/gpu} style is part of the "gpu" package. They are only
enabled if LAMMPS was built with the those packages. See the "Making
LAMMPS"_Section_start.html#2_3 section for more info.
This pair style requires that atoms store torque and a quaternion to

46
doc/pair_lj.html
View File

@ -11,6 +11,8 @@
<H3>pair_style lj/cut command
</H3>
<H3>pair_style lj/cut/gpu command
</H3>
<H3>pair_style lj/cut/opt command
</H3>
<H3>pair_style lj/cut/coul/cut command
@ -25,11 +27,13 @@
</P>
<PRE>pair_style style args
</PRE>
<UL><LI>style = <I>lj/cut</I> or <I>lj/cut/opt</I> or <I>lj/cut/coul/cut</I> or <I>lj/cut/coul/debye</I> or <I>lj/cut/coul/long</I> or <I>lj/cut/coul/long/tip4p</I>
<UL><LI>style = <I>lj/cut</I> or <I>lj/cut/gpu</I> or <I>lj/cut/opt</I> or <I>lj/cut/coul/cut</I> or <I>lj/cut/coul/debye</I> or <I>lj/cut/coul/long</I> or <I>lj/cut/coul/long/tip4p</I>
<LI>args = list of arguments for a particular style
</UL>
<PRE> <I>lj/cut</I> args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
<I>lj/cut/gpu</I> args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
<I>lj/cut/opt</I> args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
<I>lj/cut/coul/cut</I> args = cutoff (cutoff2)
@ -52,6 +56,7 @@
<P><B>Examples:</B>
</P>
<PRE>pair_style lj/cut 2.5
pair_style lj/cut/gpu 2.5
pair_style lj/cut/opt 2.5
pair_coeff * * 1 1
pair_coeff 1 1 1 1.1 2.8
@ -87,6 +92,11 @@ given by
</CENTER>
<P>Rc is the cutoff.
</P>
<P>Style <I>gayberne/gpu</I> is a GPU-enabled version of style <I>lj/cut</I> that
should give identical answers. Depending on system size and the GPU
processor you have on your system, it may be 4x faster (for the
pairwise portion of the run time). See more details below.
</P>
<P>Style <I>lj/cut/opt</I> is an optimized version of style <I>lj/cut</I> that
should give identical answers. Depending on system size and the
processor you are running on, it may be 5-25% faster (for the pairwise
@ -165,6 +175,30 @@ Coulombic cutoff specified in the pair_style command.
</P>
<HR>
<P>The <I>lj/cut/gpu</I> style is identical to the <I>lj/cut</I> style, except that
each processor off-loads its pairwise calculations to a GPU chip.
Depending on the hardware available on your system this can provide a
speed-up. See the <A HREF = "Section_start.html#2_8">Running on GPUs</A> section of
the manual for more details about hardware and software requirements
for using GPUs.
</P>
<P>Additional requirements in your input script to run with style
<I>gayberne/gpu</I> are as follows:
</P>
<P>The <A HREF = "newton.html">newton pair</A> setting must be <I>off</I>.
</P>
<P>You should use the <A HREF = "neigh_modify.html">neigh_modify one</A> command and
set its value to something close (but slightly larger) than the number
of pairwise neighbors/atom you expect to have in your model. This is
a function of the pairwise cutoff. Note that the default for this
setting is 2000, which is much larger than most models need. Unlike
neighbor lists in LAMMPS itself, the GPU version of this pair style
uses that setting to allocate memory on the GPU for neighbor
information. If the setting is too large, it will limit the number of
atoms that can be stored on the GPU.
</P>
<HR>
<P><B>Mixing, shift, table, tail correction, restart, rRESPA info</B>:
</P>
<P>For atom type pairs I,J and I != J, the epsilon and sigma coefficients
@ -202,11 +236,11 @@ See the <A HREF = "run_style.html">run_style</A> command for details.
<P><B>Restrictions:</B>
</P>
<P>The <I>lj/cut/coul/long</I> and <I>lj/cut/coul/long/tip4p</I> styles are part of
the "kspace" package. They are only enabled if LAMMPS was built with
that package (which it is by default). The <I>lj/cut/opt</I> style is part
of the "opt" package. It is only enabled if LAMMPS was built with
that package. See the <A HREF = "Section_start.html#2_3">Making LAMMPS</A> section
for more info.
the "kspace" package. The <I>lj/cut/gpu</I> style is part of the "gpu"
package. The <I>lj/cut/opt</I> style is part of the "opt" package. They
are only enabled if LAMMPS was built with those packages. See the
<A HREF = "Section_start.html#2_3">Making LAMMPS</A> section for more info. Note
that the "kspace" package is installed by default.
</P>
<P>On some 64-bit machines, compiling with -O3 appears to break the
Coulombic tabling option used by the <I>lj/cut/coul/long</I> style. See

45
doc/pair_lj.txt
View File

@ -7,6 +7,7 @@
:line
pair_style lj/cut command :h3
pair_style lj/cut/gpu command :h3
pair_style lj/cut/opt command :h3
pair_style lj/cut/coul/cut command :h3
pair_style lj/cut/coul/debye command :h3
@ -17,11 +18,13 @@ pair_style lj/cut/coul/long/tip4p command :h3
pair_style style args :pre
style = {lj/cut} or {lj/cut/opt} or {lj/cut/coul/cut} or {lj/cut/coul/debye} \
style = {lj/cut} or {lj/cut/gpu} or {lj/cut/opt} or {lj/cut/coul/cut} or {lj/cut/coul/debye} \
or {lj/cut/coul/long} or {lj/cut/coul/long/tip4p}
args = list of arguments for a particular style :ul
{lj/cut} args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
{lj/cut/gpu} args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
{lj/cut/opt} args = cutoff
cutoff = global cutoff for Lennard Jones interactions (distance units)
{lj/cut/coul/cut} args = cutoff (cutoff2)
@ -44,6 +47,7 @@ args = list of arguments for a particular style :ul
[Examples:]
pair_style lj/cut 2.5
pair_style lj/cut/gpu 2.5
pair_style lj/cut/opt 2.5
pair_coeff * * 1 1
pair_coeff 1 1 1 1.1 2.8 :pre
@ -79,6 +83,11 @@ given by
Rc is the cutoff.
Style {gayberne/gpu} is a GPU-enabled version of style {lj/cut} that
should give identical answers. Depending on system size and the GPU
processor you have on your system, it may be 4x faster (for the
pairwise portion of the run time). See more details below.
Style {lj/cut/opt} is an optimized version of style {lj/cut} that
should give identical answers. Depending on system size and the
processor you are running on, it may be 5-25% faster (for the pairwise
@ -157,6 +166,30 @@ Coulombic cutoff specified in the pair_style command.
:line
The {lj/cut/gpu} style is identical to the {lj/cut} style, except that
each processor off-loads its pairwise calculations to a GPU chip.
Depending on the hardware available on your system this can provide a
speed-up. See the "Running on GPUs"_Section_start.html#2_8 section of
the manual for more details about hardware and software requirements
for using GPUs.
Additional requirements in your input script to run with style
{gayberne/gpu} are as follows:
The "newton pair"_newton.html setting must be {off}.
You should use the "neigh_modify one"_neigh_modify.html command and
set its value to something close (but slightly larger) than the number
of pairwise neighbors/atom you expect to have in your model. This is
a function of the pairwise cutoff. Note that the default for this
setting is 2000, which is much larger than most models need. Unlike
neighbor lists in LAMMPS itself, the GPU version of this pair style
uses that setting to allocate memory on the GPU for neighbor
information. If the setting is too large, it will limit the number of
atoms that can be stored on the GPU.
:line
[Mixing, shift, table, tail correction, restart, rRESPA info]:
For atom type pairs I,J and I != J, the epsilon and sigma coefficients
@ -194,11 +227,11 @@ See the "run_style"_run_style.html command for details.
[Restrictions:]
The {lj/cut/coul/long} and {lj/cut/coul/long/tip4p} styles are part of
the "kspace" package. They are only enabled if LAMMPS was built with
that package (which it is by default). The {lj/cut/opt} style is part
of the "opt" package. It is only enabled if LAMMPS was built with
that package. See the "Making LAMMPS"_Section_start.html#2_3 section
for more info.
the "kspace" package. The {lj/cut/gpu} style is part of the "gpu"
package. The {lj/cut/opt} style is part of the "opt" package. They
are only enabled if LAMMPS was built with those packages. See the
"Making LAMMPS"_Section_start.html#2_3 section for more info. Note
that the "kspace" package is installed by default.
On some 64-bit machines, compiling with -O3 appears to break the
Coulombic tabling option used by the {lj/cut/coul/long} style. See

2
doc/pair_style.html
View File

@ -108,6 +108,7 @@ the pair_style command, and coefficients specified by the associated
<LI><A HREF = "pair_eam.html">pair_style eam/fs</A> - Finnis-Sinclair EAM
<LI><A HREF = "pair_eam.html">pair_style eam/fs/opt</A> - optimized version of Finnis-Sinclair EAM
<LI><A HREF = "pair_gayberne.html">pair_style gayberne</A> - Gay-Berne ellipsoidal potential
<LI><A HREF = "pair_gayberne.html">pair_style gayberne/gpu</A> - GPU-enabled Gay-Berne ellipsoidal potential
<LI><A HREF = "pair_gran.html">pair_style gran/hertz/history</A> - granular potential with Hertzian interactions
<LI><A HREF = "pair_gran.html">pair_style gran/hooke</A> - granular potential with history effects
<LI><A HREF = "pair_gran.html">pair_style gran/hooke/history</A> - granular potential without history effects
@ -119,6 +120,7 @@ the pair_style command, and coefficients specified by the associated
<LI><A HREF = "pair_class2.html">pair_style lj/class2/coul/cut</A> - COMPASS with cutoff Coulomb
<LI><A HREF = "pair_class2.html">pair_style lj/class2/coul/long</A> - COMPASS with long-range Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut</A> - cutoff Lennard-Jones potential with no Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut/gpu</A> - GPU-enabled version of cutoff LJ
<LI><A HREF = "pair_lj.html">pair_style lj/cut/opt</A> - optimized version of cutoff LJ
<LI><A HREF = "pair_lj.html">pair_style lj/cut/coul/cut</A> - LJ with cutoff Coulomb
<LI><A HREF = "pair_lj.html">pair_style lj/cut/coul/debye</A> - LJ with Debye screening added to Coulomb

2
doc/pair_style.txt
View File

@ -104,6 +104,7 @@ the pair_style command, and coefficients specified by the associated
"pair_style eam/fs"_pair_eam.html - Finnis-Sinclair EAM
"pair_style eam/fs/opt"_pair_eam.html - optimized version of Finnis-Sinclair EAM
"pair_style gayberne"_pair_gayberne.html - Gay-Berne ellipsoidal potential
"pair_style gayberne/gpu"_pair_gayberne.html - GPU-enabled Gay-Berne ellipsoidal potential
"pair_style gran/hertz/history"_pair_gran.html - granular potential with Hertzian interactions
"pair_style gran/hooke"_pair_gran.html - granular potential with history effects
"pair_style gran/hooke/history"_pair_gran.html - granular potential without history effects
@ -115,6 +116,7 @@ the pair_style command, and coefficients specified by the associated
"pair_style lj/class2/coul/cut"_pair_class2.html - COMPASS with cutoff Coulomb
"pair_style lj/class2/coul/long"_pair_class2.html - COMPASS with long-range Coulomb
"pair_style lj/cut"_pair_lj.html - cutoff Lennard-Jones potential with no Coulomb
"pair_style lj/cut/gpu"_pair_lj.html - GPU-enabled version of cutoff LJ
"pair_style lj/cut/opt"_pair_lj.html - optimized version of cutoff LJ
"pair_style lj/cut/coul/cut"_pair_lj.html - LJ with cutoff Coulomb
"pair_style lj/cut/coul/debye"_pair_lj.html - LJ with Debye screening added to Coulomb