@@ -64,13 +64,16 @@ args = arguments specific to the style :l
{no_affinity} values = none
{kokkos} args = keyword value ...
zero or more keyword/value pairs may be appended
keywords = {neigh} or {neigh/qeq} or {newton} or {binsize} or {comm} or {comm/exchange} or {comm/forward} or {comm/reverse} or {gpu/direct}
keywords = {neigh} or {neigh/qeq} or {neigh/thread} or {newton} or {binsize} or {comm} or {comm/exchange} or {comm/forward} or {comm/reverse} or {gpu/direct}
{neigh} value = {full} or {half}
full = full neighbor list
half = half neighbor list built in thread-safe manner
{neigh/qeq} value = {full} or {half}
full = full neighbor list
half = half neighbor list built in thread-safe manner
{neigh/thread} value = {off} or {on}
off = thread only over atoms
on = thread over both atoms and neighbors
{newton} = {off} or {on}
off = set Newton pairwise and bonded flags off
on = set Newton pairwise and bonded flags on
@@ -444,6 +447,18 @@ the {neigh/qeq} keyword determines how neighbor lists are built for "fix
qeq/reax/kk"_fix_qeq_reax.html. If not explicitly set, the value of
{neigh/qeq} will match {neigh}.
If the {neigh/thread} keyword is set to {off}, then the KOKKOS package
threads only over atoms. However, for small systems, this may not expose
enough parallelism to keep a GPU busy. When this keyword is set to {on},
the KOKKOS package threads over both atoms and neighbors of atoms. When
using {neigh/thread} {on}, a full neighbor list must also be used. Using
{neigh/thread} {on} may be slower for large systems, so this this option
is turned on by default only when there are 16K atoms or less owned by
an MPI rank and when using a full neighbor list. Not all KOKKOS-enabled
potentials support this keyword yet, and only thread over atoms. Many
simple pair-wise potentials such as Lennard-Jones do support threading
over both atoms and neighbors.
The {newton} keyword sets the Newton flags for pairwise and bonded
interactions to {off} or {on}, the same as the "newton"_newton.html
command allows. The default for GPUs is {off} because this will almost
@@ -498,14 +513,14 @@ identically. When using GPUs, the {device} value is the default since it
will typically be optimal if all of your styles used in your input
script are supported by the KOKKOS package. In this case data can stay
on the GPU for many timesteps without being moved between the host and
GPU, if you use the {device} value. This requires that your MPI is able
to access GPU memory directly. Currently that is true for OpenMPI 1.8
(or later versions), Mvapich2 1.9 (or later), and CrayMPI. If your
script uses styles (e.g. fixes) which are not yet supported by the
KOKKOS package, then data has to be move between the host and device
anyway, so it is typically faster to let the host handle communication,
by using the {host} value. Using {host} instead of {no} will enable use
of multiple threads to pack/unpack communicated data.
GPU, if you use the {device} value. If your script uses styles (e.g.
fixes) which are not yet supported by the KOKKOS package, then data has
to be move between the host and device anyway, so it is typically faster
to let the host handle communication, by using the {host} value. Using
{host} instead of {no} will enable use of multiple threads to
pack/unpack communicated data. When running small systems on a GPU,
performing the exchange pack/unpack on the host CPU can give speedup
since it reduces the number of CUDA kernel launches.
The {gpu/direct} keyword chooses whether GPU-direct will be used. When
this keyword is set to {on}, buffers in GPU memory are passed directly
@@ -518,7 +533,8 @@ the {gpu/direct} keyword is automatically set to {off} by default. When
the {gpu/direct} keyword is set to {off} while any of the {comm}
keywords are set to {device}, the value for these {comm} keywords will
be automatically changed to {host}. This setting has no effect if not
running on GPUs.
running on GPUs. GPU-direct is available for OpenMPI 1.8 (or later
versions), Mvapich2 1.9 (or later), and CrayMPI.
:line
@@ -630,11 +646,12 @@ neigh/qeq = full, newton = off, binsize for GPUs = 2x LAMMPS default
value, comm = device, gpu/direct = on. When LAMMPS can safely detect
that GPU-direct is not available, the default value of gpu/direct
becomes "off". For CPUs or Xeon Phis, the option defaults are neigh =
half, neigh/qeq = half, newton = on, binsize = 0.0, and comm = no. These
settings are made automatically by the required "-k on" "command-line
switch"_Run_options.html. You can change them by using the package
kokkos command in your input script or via the "-pk kokkos command-line
switch"_Run_options.html.
half, neigh/qeq = half, newton = on, binsize = 0.0, and comm = no. The
option neigh/thread = on when there are 16K atoms or less on an MPI
rank, otherwise it is "off". These settings are made automatically by
the required "-k on" "command-line switch"_Run_options.html. You can
change them by using the package kokkos command in your input script or
via the "-pk kokkos command-line switch"_Run_options.html.
For the OMP package, the default is Nthreads = 0 and the option
defaults are neigh = yes. These settings are made automatically if
