Skip to content

Commit 7d38604

Browse files
authored
Merge pull request #1536 from Libensemble/docs/HPC_running
Docs/hpc running
2 parents 8c09098 + 97e6fdb commit 7d38604

10 files changed

Lines changed: 229 additions & 140 deletions

File tree

docs/data_structures/data_structures.rst

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -8,9 +8,9 @@ See :ref:`here<libe-module>` for instruction on constructing a complete workflow
88
:maxdepth: 2
99
:caption: libEnsemble Specifications:
1010

11-
sim_specs
12-
gen_specs
1311
libE_specs
12+
gen_specs
13+
sim_specs
1414
alloc_specs
1515
platform_specs
1616
persis_info

docs/data_structures/libE_specs.rst

Lines changed: 7 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
.. _datastruct-libe-specs:
22

3-
General Specs
4-
=============
3+
LibE Specs
4+
==========
55

66
libEnsemble is primarily customized by setting options within a ``LibeSpecs`` class or dictionary.
77

@@ -10,11 +10,10 @@ libEnsemble is primarily customized by setting options within a ``LibeSpecs`` cl
1010
from libensemble.specs import LibeSpecs
1111
1212
specs = LibeSpecs(
13-
comm=MPI.COMM_WORLD,
14-
comms="mpi",
15-
save_every_k_gens=1000,
13+
gen_on_manager=True,
14+
save_every_k_gens=100,
1615
sim_dirs_make=True,
17-
ensemble_dir_path="/scratch/ensemble",
16+
nworkers=4
1817
)
1918
2019
.. dropdown:: Settings by Category
@@ -272,8 +271,8 @@ libEnsemble is primarily customized by setting options within a ``LibeSpecs`` cl
272271
``disable_resource_manager`` is set.
273272

274273
**dedicated_mode** [bool] = ``False``:
275-
Disallow any resources running libEnsemble processes (manager and workers)
276-
from being valid targets for app submissions.
274+
Instructs libEnsemble’s MPI executor not to run applications on nodes where
275+
libEnsemble processes (manager and workers) are running.
277276

278277
**zero_resource_workers** [list of ints]:
279278
List of workers (by IDs) that require no resources. For when a fixed mapping of workers
101 KB
Loading

docs/platforms/example_scripts.rst

Lines changed: 72 additions & 53 deletions
Original file line numberDiff line numberDiff line change
@@ -2,86 +2,105 @@ Example Scheduler Submission Scripts
22
====================================
33

44
Below are example submission scripts used to configure and launch libEnsemble
5-
on a variety of high-powered systems. See :ref:`here<platform-index>` for more
6-
information about the respective systems and configuration.
5+
on a variety of high-powered systems. See :ref:`Running on HPC Systems<platform-index>`
6+
for more information about the respective systems and configuration.
7+
8+
.. note::
9+
It is **highly recommended** that the directive lines (e.g., #SBATCH) in batch
10+
submission scripts do **NOT** specify processor, task, or GPU configuration info
11+
--- these lines should only specify the number of nodes required.
12+
13+
For example, do not specify ``#SBATCH --gpus-per-node=4`` in order to use four
14+
GPUs on the node, when each worker may use less than this, as this may assign
15+
all of the GPUs to a single MPI invocation. Instead, the configuration should
16+
be supplied either
17+
:doc:`in the simulation function<../examples/sim_funcs/forces_simf_gpu>`
18+
or, if using dynamic resources,
19+
:doc:`in the generator<../examples/sim_funcs/forces_simf_gpu_vary_resources>`.
20+
721

822
General examples
923
----------------
1024

11-
.. dropdown:: Slurm - Basic
25+
Slurm - Basic
26+
^^^^^^^^^^^^^
1227

13-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_slurm_simple.sh
14-
:caption: /examples/libE_submission_scripts/submit_slurm_simple.sh
15-
:language: bash
28+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_slurm_simple.sh
29+
:caption: /examples/libE_submission_scripts/submit_slurm_simple.sh
30+
:language: bash
1631

17-
.. dropdown:: PBS - Basic
32+
PBS - Basic
33+
^^^^^^^^^^^
1834

19-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_simple.sh
20-
:caption: /examples/libE_submission_scripts/submit_pbs_simple.sh
21-
:language: bash
35+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_simple.sh
36+
:caption: /examples/libE_submission_scripts/submit_pbs_simple.sh
37+
:language: bash
2238

23-
.. dropdown:: LSF - Basic
39+
LSF - Basic
40+
^^^^^^^^^^^
2441

25-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_lsf_simple.sh
26-
:caption: /examples/libE_submission_scripts/submit_lsf_simple.sh
27-
:language: bash
42+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_lsf_simple.sh
43+
:caption: /examples/libE_submission_scripts/submit_lsf_simple.sh
44+
:language: bash
2845

2946

3047
System Examples
3148
---------------
3249

33-
.. dropdown:: Aurora
34-
35-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_aurora.sh
36-
:caption: /examples/libE_submission_scripts/submit_pbs_aurora.sh
37-
:language: bash
38-
39-
.. dropdown:: Frontier (Large WarpX Ensemble)
40-
41-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_frontier_large.sh
42-
:caption: /examples/libE_submission_scripts/submit_frontier_large.sh
43-
:language: bash
50+
Aurora
51+
^^^^^^
4452

53+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_aurora.sh
54+
:caption: /examples/libE_submission_scripts/submit_pbs_aurora.sh
55+
:language: bash
4556

46-
.. dropdown:: Perlmutter
57+
Frontier (Large WarpX Ensemble)
58+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4759

48-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_perlmutter.sh
49-
:caption: /examples/libE_submission_scripts/submit_perlmutter.sh
50-
:language: bash
60+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_frontier_large.sh
61+
:caption: /examples/libE_submission_scripts/submit_frontier_large.sh
62+
:language: bash
5163

52-
.. dropdown:: Polaris
64+
Perlmutter
65+
^^^^^^^^^^
5366

54-
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_polaris.sh
55-
:caption: /examples/libE_submission_scripts/submit_pbs_polaris.sh
56-
:language: bash
67+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_perlmutter.sh
68+
:caption: /examples/libE_submission_scripts/submit_perlmutter.sh
69+
:language: bash
5770

58-
.. dropdown:: Bridges - Central Mode
71+
Polaris
72+
^^^^^^^
5973

60-
.. literalinclude:: ../../examples/libE_submission_scripts/bridges_submit_slurm_central.sh
61-
:caption: /examples/libE_submission_scripts/bridges_submit_slurm_central.sh
62-
:language: bash
74+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_pbs_polaris.sh
75+
:caption: /examples/libE_submission_scripts/submit_pbs_polaris.sh
76+
:language: bash
6377

64-
.. dropdown:: Bebop - Central Mode
78+
Bebop
79+
^^^^^
6580

66-
.. literalinclude:: ../../examples/libE_submission_scripts/bebop_submit_slurm_central.sh
67-
:caption: /examples/libE_submission_scripts/bebop_submit_slurm_central.sh
68-
:language: bash
81+
.. literalinclude:: ../../examples/libE_submission_scripts/bebop_submit_pbs_central.sh
82+
:caption: /examples/libE_submission_scripts/bebop_submit_pbs_central.sh
83+
:language: bash
6984

70-
.. dropdown:: Bebop - Distributed Mode
85+
Bridges - MPI / Central Mode
86+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7187

72-
.. literalinclude:: ../../examples/libE_submission_scripts/bebop_submit_pbs_distrib.sh
73-
:caption: /examples/libE_submission_scripts/bebop_submit_pbs_distrib.sh
74-
:language: bash
88+
.. literalinclude:: ../../examples/libE_submission_scripts/bridges_submit_slurm_central.sh
89+
:caption: /examples/libE_submission_scripts/bridges_submit_slurm_central.sh
90+
:language: bash
7591

76-
.. dropdown:: Summit (Decommissioned) - On Launch Nodes with Multiprocessing
92+
.. _slurm_mpi_distributed:
7793

78-
.. literalinclude:: ../../examples/libE_submission_scripts/summit_submit_mproc.sh
79-
:caption: /examples/libE_submission_scripts/summit_submit_mproc.sh
80-
:language: bash
94+
SLURM - MPI / Distributed Mode (co-locate workers & MPI applications)
95+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8196

82-
.. dropdown:: Cobalt - Intermediate node with Multiprocessing
97+
.. literalinclude:: ../../examples/libE_submission_scripts/submit_distrib_mpi4py.sh
98+
:caption: /examples/libE_submission_scripts/submit_distrib_mpi4py.sh
99+
:language: bash
83100

84-
.. literalinclude:: ../../examples/libE_submission_scripts/cobalt_submit_mproc.sh
85-
:caption: /examples/libE_submission_scripts/cobalt_submit_mproc.sh
86-
:language: bash
101+
Summit (Decommissioned) - On Launch Nodes with Multiprocessing
102+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
87103

104+
.. literalinclude:: ../../examples/libE_submission_scripts/summit_submit_mproc.sh
105+
:caption: /examples/libE_submission_scripts/summit_submit_mproc.sh
106+
:language: bash

0 commit comments

Comments
 (0)