HPC/Applications/lammps/Package GPU - Revision history

Stern at 18:33, March 23, 2018

2018-03-23T18:33:45Z

← Older revision		Revision as of 18:33, March 23, 2018
Line 7:		Line 7:

	=== Usage ===		=== Usage ===
	# Use the ~~command [http://lammps.sandia.gov/doc/package.html~~ '''~~<code>package~~ gpu '~~'Ngpu~~''</code>~~'''] near~~ the ~~beginning of your LAMMPS control script. Since all Carbon GPU nodes have just~~ one ~~GPU per node, the ''Ngpu'' argument must be 1~~.		# Use the <code>lmp_mpi'''-gpu'''</code> binary instead of the MPI-only one.
	# Do one of the following: ~~<!-- In your LAMMPS control script, -->~~		# Use a GPU potential style. Do one of the following:
	#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).		#* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).
	#:* Use the [http://lammps.sandia.gov/doc/suffix.html ~~'''~~suffix gpu~~''' command~~].		#* Use the command [http://lammps.sandia.gov/doc/suffix.html <code>suffix gpu</code>].
	#:* On the command line, use the [http://lammps.sandia.gov/doc/Section_start.html#~~start_7~~ '''~~-suffix gpu~~''' ~~switch~~].		#* On the command line, use the [http://lammps.sandia.gov/doc/Section_start.html#command-line-options <code>-suffix gpu</code>].
	# In the job file or qsub command line, [http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources request a GPU] <code>#PBS -l nodes=...:gpus=1</code> (referring to the number of GPUs per node).		# Detail the GPU request. Do one of the following:
	~~# Call the <code>lmp_mpi'''-gpu'''</code> binary.~~		#* Near the beginning of your LAMMPS control script, use the command [http://lammps.sandia.gov/doc/package.html <code>package gpu ''Ngpu''</code>] . Since all Carbon GPU nodes have just one GPU per node, the ''Ngpu'' argument must be 1.
			#* Add [http://lammps.sandia.gov/doc/Section_start.html#command-line-options <code>-package gpu 1</code>] to the command line options. This is perhaps preferred because you do not need to alter your input file.
	~~=== Input file examples ===~~		# In the job file or qsub command line, [http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources request a node ''having'' a GPU], using <code>#PBS -l nodes=...:gpus=1</code> (referring to the number of GPUs per node).
	~~package gpu force 0 0 1.0~~
	~~package gpu force 0 0 0.75~~
	~~package gpu force/neigh 0 0 1.0~~
	~~package gpu force/neigh 0 1 -1.0~~

	…
	~~pair_style lj/charmm/coul/long'''/gpu''' 8.0 10.0~~

	~~Alternatively, use the following command line options:~~
	~~lmp_mpi-gpu '''''-sf gpu -pk gpu 1''''' -in ''file.in'' > ''file.out''~~

	=== Job file example ===		=== Job file example ===
	#PBS -l nodes=...''':gpus=1'''		#PBS -l nodes=...''':gpus=1'''
	…		…
	mpirun … lmp_mpi'''-gpu''' -in ''infile''		mpirun … lmp_mpi'''-gpu''' '''-sf gpu -pk gpu 1''' -in ''infile''

Stern: /* Usage */

2018-03-23T18:15:55Z

Usage

← Older revision		Revision as of 18:15, March 23, 2018
Line 7:		Line 7:

	=== Usage ===		=== Usage ===
	# Use the command [http://lammps.sandia.gov/doc/package.html '''<code>package gpu ''~~mode first last split~~''</code>'''] near the beginning of your LAMMPS control script. Since all Carbon GPU nodes have just one GPU per node, the ~~arguments~~ ''~~first'' and ''last~~'' must ~~always~~ be ~~zero; ''split'' is not restricted~~.		# Use the command [http://lammps.sandia.gov/doc/package.html '''<code>package gpu ''Ngpu''</code>'''] near the beginning of your LAMMPS control script. Since all Carbon GPU nodes have just one GPU per node, the ''Ngpu'' argument must be 1.
	# Do one of the following: <!-- In your LAMMPS control script, -->		# Do one of the following: <!-- In your LAMMPS control script, -->
	#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).		#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).
Line 13:		Line 13:
	#:* On the command line, use the [http://lammps.sandia.gov/doc/Section_start.html#start_7 '''-suffix gpu''' switch].		#:* On the command line, use the [http://lammps.sandia.gov/doc/Section_start.html#start_7 '''-suffix gpu''' switch].
	# In the job file or qsub command line, [http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources request a GPU] <code>#PBS -l nodes=...:gpus=1</code> (referring to the number of GPUs per node).		# In the job file or qsub command line, [http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources request a GPU] <code>#PBS -l nodes=...:gpus=1</code> (referring to the number of GPUs per node).
	# Call the <code>~~lmp_openmpi~~'''-gpu'''</code> binary.		# Call the <code>lmp_mpi'''-gpu'''</code> binary.

	=== Input file examples ===		=== Input file examples ===

Stern: /* Job file example */

2018-03-23T18:12:35Z

Job file example

← Older revision		Revision as of 18:12, March 23, 2018
Line 30:		Line 30:
	#PBS -l nodes=...''':gpus=1'''		#PBS -l nodes=...''':gpus=1'''
	…		…
	mpirun … ~~lmp_openmpi~~'''-gpu''' -in ''infile''		mpirun … lmp_mpi'''-gpu''' -in ''infile''

Stern: /* Input file examples */

2018-03-23T18:11:27Z

Input file examples

← Older revision		Revision as of 18:11, March 23, 2018
Line 23:		Line 23:
	…		…
	pair_style lj/charmm/coul/long'''/gpu''' 8.0 10.0		pair_style lj/charmm/coul/long'''/gpu''' 8.0 10.0

			Alternatively, use the following command line options:
			lmp_mpi-gpu '''''-sf gpu -pk gpu 1''''' -in ''file.in'' > ''file.out''

	=== Job file example ===		=== Job file example ===

Stern: Stern moved page HPC/Modules/lammps/Package GPU to HPC/Applications/lammps/Package GPU

2013-06-25T15:21:28Z

Stern moved page HPC/Modules/lammps/Package GPU to HPC/Applications/lammps/Package GPU

← Older revision	Revision as of 15:21, June 25, 2013
(No difference)

Stern: /* Usage */

2012-11-05T23:05:36Z

Usage

← Older revision		Revision as of 23:05, November 5, 2012
Line 7:		Line 7:

	=== Usage ===		=== Usage ===
	# Use the command [http://lammps.sandia.gov/doc/package.html '''<code>package gpu</code>'''] near the beginning of your LAMMPS control script. Since all Carbon GPU nodes have just one GPU per node, the ~~first two~~ arguments ~~(called~~ ''first'' and ''last'') must always be zero; ~~the~~ ''split'' ~~argument~~ is not restricted.		# Use the command [http://lammps.sandia.gov/doc/package.html '''<code>package gpu ''mode first last split''</code>'''] near the beginning of your LAMMPS control script. Since all Carbon GPU nodes have just one GPU per node, the arguments ''first'' and ''last'' must always be zero; ''split'' is not restricted.
	# Do one of the following: <!-- In your LAMMPS control script, -->		# Do one of the following: <!-- In your LAMMPS control script, -->
	#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).		#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).

Stern: Created page with "== Package GPU == * Provides multi-threaded versions of most pair styles, all dihedral styles and a few fixes in LAMMPS; for the full list: *# In your browser, open http://lam..."

2012-11-05T22:41:42Z

Created page with "== Package GPU == * Provides multi-threaded versions of most pair styles, all dihedral styles and a few fixes in LAMMPS; for the full list: *# In your browser, open http://lam..."

New page

== Package GPU ==
* Provides multi-threaded versions of most pair styles, all dihedral styles and a few fixes in LAMMPS; for the full list:
*# In your browser, open http://lammps.sandia.gov/doc/Section_commands.html#comm
*# Search for the string '''/cuda'''.
* Supports one physical GPU per LAMMPS MPI process (CPU core).
* Multiple MPI processes (CPU cores) can share a single GPU, and in many cases it will be more efficient to run this way.

=== Usage ===
# Use the command [http://lammps.sandia.gov/doc/package.html '''<code>package gpu</code>'''] near the beginning of your LAMMPS control script. Since all Carbon GPU nodes have just one GPU per node, the first two arguments (called ''first'' and ''last'') must always be zero; the ''split'' argument is not restricted.
# Do one of the following: 
#:* Append '''/gpu''' to the style name (e.g. pair_style lj/cut/gpu).
#:* Use the [http://lammps.sandia.gov/doc/suffix.html '''suffix gpu''' command].
#:* On the command line, use the [http://lammps.sandia.gov/doc/Section_start.html#start_7 '''-suffix gpu''' switch].
# In the job file or qsub command line, [http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources request a GPU] <code>#PBS -l nodes=...:gpus=1</code> (referring to the number of GPUs per node).
# Call the <code>lmp_openmpi'''-gpu'''</code> binary.

=== Input file examples ===
package gpu force 0 0 1.0
package gpu force 0 0 0.75
package gpu force/neigh 0 0 1.0
package gpu force/neigh 0 1 -1.0

…
pair_style lj/charmm/coul/long'''/gpu''' 8.0 10.0

=== Job file example ===
#PBS -l nodes=...''':gpus=1'''
…
mpirun … lmp_openmpi'''-gpu''' -in ''infile''