<p>This command sets parameters that affect the inter-processor
communication of atom information that occurs each timestep as
coordinates and other properties are exchanged between neighboring
processors and stored as properties of ghost atoms.</p>
<divclass="admonition note">
<pclass="first admonition-title">Note</p>
<pclass="last">These options apply to the currently defined comm style. When
you specify a <aclass="reference internal"href="comm_style.html"><spanclass="doc">comm_style</span></a> command, all communication
settings are restored to their default values, including those
previously reset by a comm_modify command. Thus if your input script
specifies a comm_style command, you should use the comm_modify command
after it.</p>
</div>
<p>The <em>mode</em> keyword determines whether a single or multiple cutoff
distances are used to determine which atoms to communicate.</p>
<p>The default mode is <em>single</em> which means each processor acquires
information for ghost atoms that are within a single distance from its
sub-domain. The distance is by default the maximum of the neighbor
cutoff across all atom type pairs.</p>
<p>For many systems this is an efficient algorithm, but for systems with
widely varying cutoffs for different type pairs, the <em>multi</em> mode can
be faster. In this case, each atom type is assigned its own distance
cutoff for communication purposes, and fewer atoms will be
communicated. See the <aclass="reference internal"href="neighbor.html"><spanclass="doc">neighbor multi</span></a> command for a
neighbor list construction option that may also be beneficial for
simulations of this kind.</p>
<p>The <em>cutoff</em> keyword allows you to extend the ghost cutoff distance
for communication mode <em>single</em>, which is the distance from the borders
of a processor’s sub-domain at which ghost atoms are acquired from other
processors. By default the ghost cutoff = neighbor cutoff = pairwise
force cutoff + neighbor skin. See the <aclass="reference internal"href="neighbor.html"><spanclass="doc">neighbor</span></a> command
for more information about the skin distance. If the specified Rcut is
greater than the neighbor cutoff, then extra ghost atoms will be acquired.
If the provided cutoff is smaller, the provided value will be ignored
and the ghost cutoff is set to the neighbor cutoff. Specifying a
cutoff value of 0.0 will reset any previous value to the default.</p>
<p>The <em>cutoff/multi</em> option is equivalent to <em>cutoff</em>, but applies to
communication mode <em>multi</em> instead. Since in this case the communication
cutoffs are determined per atom type, a type specifier is needed and
cutoff for one or multiple types can be extended. Also ranges of types
using the usual asterisk notation can be given.</p>
<p>These are simulation scenarios in which it may be useful or even
necessary to set a ghost cutoff > neighbor cutoff:</p>
<ulclass="simple">
<li>a single polymer chain with bond interactions, but no pairwise interactions</li>
<li>bonded interactions (e.g. dihedrals) extend further than the pairwise cutoff</li>
<li>ghost atoms beyond the pairwise cutoff are needed for some computation</li>
</ul>
<p>In the first scenario, a pairwise potential is not defined. Thus the
pairwise neighbor cutoff will be 0.0. But ghost atoms are still
needed for computing bond, angle, etc interactions between atoms on
different processors, or when the interaction straddles a periodic
boundary.</p>
<p>The appropriate ghost cutoff depends on the <aclass="reference internal"href="newton.html"><spanclass="doc">newton bond</span></a>
setting. For newton bond <em>off</em>, the distance needs to be the furthest
distance between any two atoms in the bond, angle, etc. E.g. the
distance between 1-4 atoms in a dihedral. For newton bond <em>on</em>, the
distance between the central atom in the bond, angle, etc and any
other atom is sufficient. E.g. the distance between 2-4 atoms in a
dihedral.</p>
<p>In the second scenario, a pairwise potential is defined, but its
neighbor cutoff is not sufficiently long enough to enable bond, angle,
etc terms to be computed. As in the previous scenario, an appropriate
ghost cutoff should be set.</p>
<p>In the last scenario, a <aclass="reference internal"href="fix.html"><spanclass="doc">fix</span></a> or <aclass="reference internal"href="compute.html"><spanclass="doc">compute</span></a> or
<aclass="reference internal"href="pair_style.html"><spanclass="doc">pairwise potential</span></a> needs to calculate with ghost
atoms beyond the normal pairwise cutoff for some computation it
performs (e.g. locate neighbors of ghost atoms in a multibody pair
potential). Setting the ghost cutoff appropriately can insure it will
find the needed atoms.</p>
<divclass="admonition note">
<pclass="first admonition-title">Note</p>
<pclass="last">In these scenarios, if you do not set the ghost cutoff long
enough, and if there is only one processor in a periodic dimension
(e.g. you are running in serial), then LAMMPS may “find” the atom it
is looking for (e.g. the partner atom in a bond), that is on the far
side of the simulation box, across a periodic boundary. This will
typically lead to bad dynamics (i.e. the bond length is now the
simulation box length). To detect if this is happening, see the
Built with <ahref="http://sphinx-doc.org/">Sphinx</a> using a <ahref="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <ahref="https://readthedocs.org">Read the Docs</a>.