add platform descriptions to Introduction, reference these properly i…

…n branson, miniem, and sparta. Start to get consistent naming accross results, we should all use Crossroads and Sierra as names, we caveat in the intro that testbeds are often used that have same architecture
lanl · Sep 21, 2023 · 058395a · 058395a
1 parent 17390e1
commit 058395a
Show file tree

Hide file tree

Showing 12 changed files with 58 additions and 73 deletions.
diff --git a/branson b/branson
diff --git a/doc/sphinx/00_intro/introduction.rst b/doc/sphinx/00_intro/introduction.rst
@@ -239,6 +239,37 @@ Where:
 *	w = weighting factor. 
 
 
+System Information
+==================
+
+The baseline platform for the ATS-5 procurement is the ATS-3 system (described below). 
+GPU performance is provided on the ATS-2 system and in some cases other GPU based systems 
+and is for information only, these are not to be used as baselines. 
+In most cases the performance numbers provided herein were collected on smaller scale 
+testbed systems that are the same architecture as that of ATS-3 and ATS-2 systems. 
+
+* Advanced Technology System 3 (ATS-3), also known as Crossroads (see :ref:`GlobalSystemATS3`)
+* Advanced Technology System 2 (ATS-2), also known as Sierra (see :ref:`GlobalSystemATS2`)
+
+
+.. _GlobalSystemATS3:
+
+ATS-3/Crossroads
+----------------
+
+This system has over 6,000 compute nodes that are made up of two Intel(R) Xeon(R) Max 9480 CPUs 
+interconnected with HPE Slingshot 11 interconnect. 
+
+.. _GlobalSystemATS2:
+
+ATS-2/Sierra
+------------
+
+This system has 4,284  compute nodes that are made up of two Power9
+CPUs with four NVIDIA V100 GPUs. Please refer to [Sierra-LLNL]_ for more
+detailed information.
+
+
 
 Approvals
 =========

diff --git a/doc/sphinx/01_branson/branson.rst b/doc/sphinx/01_branson/branson.rst
@@ -124,8 +124,10 @@ The memory footprint can be controlled by editing "photons" in the input file.
 
 Results from Branson are provided on the following systems:
 
-* Crossroads,
-* IBM Power9 with Nvidia V100 GPU, 
+* Crossroads (see
+  :ref:`GlobalSystemATS3`)
+* Sierra (see
+  :ref:`GlobalSystemATS2`)
 
 Crossroads
 ------------

diff --git a/doc/sphinx/07_miniem/miniem.rst b/doc/sphinx/07_miniem/miniem.rst
@@ -110,10 +110,8 @@ The platforms utilized for benchmarking activities are listed and described belo
 
 * Commodity Technology System 1 (CTS-1) with Intel Cascade Lake processors,
   known as Manzano at SNL (see :ref:`MiniEMSystemCTS1`)
-* Advanced Technology System 3 (ATS-3), also known as Crossroads (see
-  :ref:`MiniEMSystemATS3`)
-* Advanced Technology System 2 (ATS-2), also known as Sierra (see
-  :ref:`MiniEMSystemATS2`)
+* Crossroads (see :ref:`GlobalSystemATS3`)
+* Sierra (see :ref:`GlobalSystemATS2`)
 
 
 .. _MiniEMSystemCTS3:
@@ -145,27 +143,6 @@ instruction and data, and L3 being shared across all cores in the processor. The
 cache size is 1.5 MB/core, 35.75 MB/processor, or 71.5 MB/node.
 
 
-.. _MiniEMSystemATS3:
-
-ATS-3/Crossroads
-----------------
-
-This system is not available yet but is slated to be the reference platform.
-
-
-.. _MiniEMSystemATS2:
-
-ATS-2/Sierra
-------------
-
-This system has a plethora of compute nodes that are made up of Power9
-processors with four NVIDIA V100 GPUs. Please refer to [Sierra-LLNL]_ for more
-detailed information.
-
-A Sierra application and regression testbed system named Vortex, housed at SNL,
-was used for benchmarking for convenience. Vortex has the same compute node
-hardware as Sierra.
-
 
 Building
 ========

diff --git a/doc/sphinx/08_sparta/ats2.gp b/doc/sphinx/08_sparta/ats2.gp
@@ -2,7 +2,7 @@
 set terminal pngcairo enhanced size 1024, 768 dashed font 'Helvetica,18'
 set output "ats2.png"
 
-set title "SPARTA Throughput Performance on ATS-2/Vortex" font "serif,22"
+set title "SPARTA Throughput Performance on Sierra/Vortex" font "serif,22"
 set xlabel "No. Particles"
 set ylabel "Figure of Merit (M-particle-steps/sec)"
 

diff --git a/doc/sphinx/08_sparta/sparta.rst b/doc/sphinx/08_sparta/sparta.rst
@@ -114,7 +114,7 @@ Figure of Merit
 
 Each SPARTA simulation writes out a file named "log.sparta". At the end of this
 simulation is a block that resembles the following example (this is from the
-ATS-2/Sierra case discussed below with 57,143,091 particles whose full output is
+Sierra case discussed below with 57,143,091 particles whose full output is
 within :download:`log.sparta <log.sparta>`).
 
 .. code-block::
@@ -168,10 +168,8 @@ The platforms utilized for benchmarking activities are listed and described belo
 
 * Commodity Technology System 1 (CTS-1) with Intel Cascade Lake processors,
   known as Manzano at SNL (see :ref:`SystemCTS1`)
-* Advanced Technology System 3 (ATS-3), also known as Crossroads (see
-  :ref:`SystemATS3`)
-* Advanced Technology System 2 (ATS-2), also known as Sierra (see
-  :ref:`SystemATS2`)
+* Crossroads (see  :ref:`GlobalSystemATS3`)
+* Sierra (see :ref:`GlobalSystemATS2`)
 
 
 .. _SystemCTS3:
@@ -202,29 +200,6 @@ levels of cache, with L1 using separate instruction and data caches, L2 unifying
 instruction and data, and L3 being shared across all cores in the processor. The
 cache size is 1.5 MB/core, 35.75 MB/processor, or 71.5 MB/node.
 
-
-.. _SystemATS3:
-
-ATS-3/Crossroads
-----------------
-
-This system is not available yet but is slated to be the reference platform.
-
-
-.. _SystemATS2:
-
-ATS-2/Sierra
-------------
-
-This system has a plethora of compute nodes that are made up of Power9
-processors with four NVIDIA V100 GPUs. Please refer to [Sierra-LLNL]_ for more
-detailed information.
-
-A Sierra application and regression testbed system named Vortex, housed at SNL,
-was used for benchmarking for convenience. Vortex has the same compute node
-hardware as Sierra.
-
-
 Building
 ========
 
@@ -233,7 +208,7 @@ Instructions are provided on how to build SPARTA for the following systems:
 * Generic (see :ref:`BuildGeneric`)
 * Commodity Technology System 1 (CTS-1) with Intel Cascade Lake processors,
   known as Manzano at SNL (see :ref:`BuildCTS1`)
-* Advanced Technology System 2 (ATS-2), also known as Sierra (see
+* Sierra (see
   :ref:`BuildATS2`)
 
 If submodules were cloned within this repository, then the source code to build
@@ -269,7 +244,7 @@ at the top level of this repository.
 
 .. _BuildATS2:
 
-ATS-2/Vortex
+Sierra (Vortex testbed)
 ------------
 
 Instructions for building on Sierra are provided below.
@@ -293,7 +268,7 @@ Instructions are provided on how to run SPARTA for the following systems:
 
 * Commodity Technology System 1 (CTS-1) with Intel Cascade Lake processors,
   known as Manzano at SNL (see :ref:`RunCTS1`)
-* Advanced Technology System 2 (ATS-2), also known as Sierra (see
+* Sierra (see
   :ref:`RunATS2`)
 
 
@@ -324,7 +299,7 @@ An example of how to run the test case on Manzano is provided below.
 
 .. _RunATS2:
 
-ATS-2/Vortex
+Sierra (Vortex testbed)
 ------------
 
 An example of how to run the test case with a single GPU on Sierra is provided
@@ -350,7 +325,7 @@ Results from SPARTA are provided on the following systems:
 
 * Commodity Technology System 1 (CTS-1) with Intel Cascade Lake processors,
   known as Manzano at SNL (see :ref:`ResultsCTS1`)
-* Advanced Technology System 2 (ATS-2), also known as Sierra (see
+* Sierra (see
   :ref:`ResultsATS2`)
 
 
@@ -462,13 +437,13 @@ subsections.
 
 .. _ResultsATS2:
 
-ATS-2/Vortex
+Sierra (Vortex testbed)
 ------------
 
-Throughput performance of SPARTA on ATS-2/Vortex is provided within the
+Throughput performance of SPARTA on Sierra/Vortex is provided within the
 following table and figure.
 
-.. csv-table:: SPARTA Throughput Performance on ATS-2/Vortex
+.. csv-table:: SPARTA Throughput Performance on Sierra/Vortex
    :file: ats2.csv
    :align: center
    :widths: 10, 10
@@ -477,9 +452,9 @@ following table and figure.
 .. figure:: ats2.png
    :align: center
    :scale: 50%
-   :alt: SPARTA Throughput Performance on ATS-2/Vortex
+   :alt: SPARTA Throughput Performance on Sierra/Vortex
 
-   SPARTA Throughput Performance on ATS-2/Vortex
+   SPARTA Throughput Performance on Sierra/Vortex
 
 Output from the largest case is within :download:`log.sparta <log.sparta>`.
 

diff --git a/parthenon b/parthenon
diff --git a/sparta b/sparta
diff --git a/spatter b/spatter
diff --git a/trilinos b/trilinos
diff --git a/umt b/umt
diff --git a/utils/pavilion b/utils/pavilion
+25 −13		README.md
+1 −0		inputs/3D_hohlraum_multi_node.xml
+1 −0		inputs/3D_hohlraum_single_node.xml
+1 −0		inputs/big_cube.xml
+1 −0		inputs/cube_decomp_test.xml
+1 −0		inputs/hot_zone_input.xml
+1 −0		inputs/marshak_wave_dd.xml
+1 −0		inputs/marshak_wave_replicated.xml
+14 −10		src/CMakeLists.txt
+5 −0		src/cell_tally.h
+8 −0		src/config.h.in
+60 −0		src/config/find_tpls.cmake
+7 −1		src/imc_parameters.h
+0 −16		src/imc_state.h
+51 −20		src/input.h
+6 −2		src/main.cc
+56 −18		src/mesh.h
+1 −1		src/particle_pass_driver.h
+5 −8		src/particle_pass_transport.h
+1 −1		src/replicated_driver.h
+5 −4		src/replicated_transport.h
+2 −0		src/test/simple_input.xml
+16 −8		src/test/test_input.cc
+26 −10		src/transport_photon.h
+0 −8		CHANGELOG.md
+0 −3		doc/sphinx/src/interface/metadata.rst
+0 −28		doc/sphinx/src/interface/state.rst
+4 −5		example/advection/parthenon_app_inputs.cpp
+4 −4		example/calculate_pi/calculate_pi.cpp
+11 −5		example/calculate_pi/pi_driver.cpp
+6 −6		example/particle_tracers/particle_tracers.cpp
+5 −5		example/stochastic_subgrid/parthenon_app_inputs.cpp
+10 −4		example/stochastic_subgrid/stochastic_subgrid_driver.cpp
+0 −1		src/CMakeLists.txt
+1 −2		src/amr_criteria/amr_criteria.hpp
+2 −2		src/basic_types.hpp
+50 −0		src/bvals/boundary_conditions.cpp
+4 −4		src/bvals/bvals.cpp
+8 −8		src/bvals/bvals_base.cpp
+168 −253		src/bvals/comms/bnd_info.cpp
+25 −69		src/bvals/comms/bnd_info.hpp
+7 −16		src/bvals/comms/boundary_communication.cpp
+48 −11		src/bvals/comms/bvals_utils.hpp
+4 −4		src/bvals/comms/flux_correction.cpp
+9 −5		src/coordinates/uniform_cartesian.hpp
+17 −61		src/defs.hpp
+29 −23		src/interface/data_collection.cpp
+6 −16		src/interface/data_collection.hpp
+0 −20		src/interface/make_pack_descriptor.hpp
+0 −36		src/interface/mesh_data.cpp
+15 −21		src/interface/mesh_data.hpp
+61 −17		src/interface/meshblock_data.cpp
+35 −45		src/interface/meshblock_data.hpp
+1 −15		src/interface/metadata.cpp
+8 −31		src/interface/metadata.hpp
+2 −3		src/interface/sparse_pack.hpp
+72 −69		src/interface/sparse_pack_base.cpp
+39 −98		src/interface/sparse_pack_base.hpp
+0 −7		src/interface/sparse_pool.hpp
+0 −87		src/interface/state_descriptor.cpp
+3 −49		src/interface/state_descriptor.hpp
+6 −6		src/interface/swarm.cpp
+1 −3		src/interface/swarm_container.hpp
+0 −8		src/interface/variable.hpp
+66 −92		src/mesh/amr_loadbalance.cpp
+0 −11		src/mesh/logical_location.hpp
+276 −214		src/mesh/mesh.cpp
+66 −23		src/mesh/mesh.hpp
+66 −2		src/mesh/mesh_refinement.cpp
+6 −0		src/mesh/mesh_refinement.hpp
+6 −11		src/mesh/meshblock.cpp
+3 −4		src/mesh/meshblock.hpp
+11 −14		src/mesh/meshblock_tree.cpp
+2 −2		src/outputs/history.cpp
+7 −9		src/outputs/parthenon_hdf5.cpp
+1 −3		src/parthenon/driver.hpp
+10 −11		src/prolong_restrict/pr_loops.hpp
+10 −11		src/prolong_restrict/prolong_restrict.cpp
+14 −16		src/prolong_restrict/prolong_restrict.hpp
+60 −131		src/tasks/task_list.hpp
+1 −14		src/utils/unique_id.cpp
+0 −4		src/utils/unique_id.hpp
+2 −2		tst/unit/test_mesh_data.cpp
+11 −1		tst/unit/test_meshblock_data_iterator.cpp
+5 −8		tst/unit/test_metadata.cpp
+2 −12		tst/unit/test_sparse_pack.cpp
+3 −85		tst/unit/test_state_descriptor.cpp
+1 −1		scripts/build_cpu.sh
+1 −1		scripts/build_cuda.sh
+1 −1		spatter
+14 −26		BUILDING.md
+2 −2		DEPENDENCIES.md
+2 −3		benchmarks/README
+0 −41		benchmarks/generate_strong_scaling_runs_spp1.py
+7 −7		benchmarks/generate_weak_scaling_runs.py
+33 −0		benchmarks/run_weak_scaling_study.sh
+52 −34		build_and_run_umt.sh
+17 −16		spack/package.py
+7 −11		src/CMakeLists.txt
+1 −2		src/cmake/GetGPUInfo.cmake
+9 −19		src/cmake/InitBuildTypeCompilerFlags.cmake
+11 −14		src/teton/CMakeLists.txt
+13 −7		src/teton/aux/SurfaceEdit.F90
+6 −3		src/teton/aux/getAngleBins.F90
+2 −1		src/teton/aux/getZonalPsi.F90
+3 −27		src/teton/control/getPhiTotal_OMPOL.F90
+0 −26		src/teton/control/initPhiTotal_OMPOL.F90
+2 −3		src/teton/control/setBoundarySources.F90
+35 −45		src/teton/driver/makeUnstructuredBox.cc
+697 −911		src/teton/driver/test_driver.cc
+0 −1		src/teton/gpu/CMakeLists.txt
+0 −732		src/teton/gpu/CornerSweepUCBxyz_OMPOL.F90
+0 −4		src/teton/gpu/SetSweep_OMPOL.F90
+1 −17		src/teton/gpu/SweepGreyUCBrz_OMPOL.F90
+0 −2		src/teton/include/TetonConduitInterface.hh
+14 −14		src/teton/misc/mpif90_mod.F90
+2 −2		src/teton/mods/AngleSet_mod.F90
+2 −9		src/teton/mods/MemoryAllocator_mod.F90
+2 −4		src/teton/mods/Size_mod.F90
+10 −12		src/teton/snac/cyclebreaker.F90