255 lines
15 KiB
Plaintext
255 lines
15 KiB
Plaintext
tu-c0r2n72 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n72 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n72 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n72 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n78 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n78 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n78 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n78 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n75 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n75 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n75 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n75 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n87 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n87 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n87 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n87 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n93 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n93 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n93 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n93 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n90 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n90 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n90 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n90 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n81 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n81 - 1 device=1 binding=--interleave=2,3
|
|
tu-c0r2n81 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n81 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n84 - 0 device=0 binding=--interleave=0,1
|
|
tu-c0r2n84 - 3 device=3 binding=--interleave=6,7
|
|
tu-c0r2n84 - 2 device=2 binding=--interleave=4,5
|
|
tu-c0r2n84 - 1 device=1 binding=--interleave=2,3
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit[0]: ========================
|
|
AcceleratorCudaInit[0]: Device Number : 0
|
|
AcceleratorCudaInit[0]: ========================
|
|
AcceleratorCudaInit[0]: Device identifier: NVIDIA A100-SXM4-40GB
|
|
AcceleratorCudaInit[0]: totalGlobalMem: 42505273344
|
|
AcceleratorCudaInit[0]: managedMemory: 1
|
|
AcceleratorCudaInit[0]: isMultiGpuBoard: 0
|
|
AcceleratorCudaInit[0]: warpSize: 32
|
|
AcceleratorCudaInit[0]: pciBusID: 3
|
|
AcceleratorCudaInit[0]: pciDeviceID: 0
|
|
AcceleratorCudaInit[0]: maxGridSize (2147483647,65535,65535)
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
OPENMPI detected
|
|
AcceleratorCudaInit[0]: ========================
|
|
AcceleratorCudaInit[0]: Device Number : 0
|
|
AcceleratorCudaInit[0]: ========================
|
|
AcceleratorCudaInit[0]: Device identifier: NVIDIA A100-SXM4-40GB
|
|
AcceleratorCudaInit[0]: totalGlobalMem: 42505273344
|
|
AcceleratorCudaInit[0]: managedMemory: 1
|
|
AcceleratorCudaInit[0]: isMultiGpuBoard: 0
|
|
AcceleratorCudaInit[0]: warpSize: 32
|
|
AcceleratorCudaInit[0]: pciBusID: 3
|
|
AcceleratorCudaInit[0]: pciDeviceID: 0
|
|
AcceleratorCudaInit[0]: maxGridSize (2147483647,65535,65535)
|
|
AcceleratorCudaInit: using default device
|
|
AcceleratorCudaInit: assume user either uses
|
|
AcceleratorCudaInit: a) IBM jsrun, or
|
|
AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding
|
|
AcceleratorCudaInit: Configure options --enable-setdevice=no
|
|
AcceleratorCudaInit: ================================================
|
|
local rank 0 device 0 bus id: 0000:03:00.0
|
|
AcceleratorCudaInit: ================================================
|
|
AcceleratorCudaInit: ================================================
|
|
AcceleratorCudaInit: ================================================
|
|
local rank 2 device 0 bus id: 0000:84:00.0
|
|
local rank 0 device 0 bus id: 0000:03:00.0
|
|
AcceleratorCudaInit: ================================================
|
|
AcceleratorCudaInit: ================================================
|
|
AcceleratorCudaInit: ================================================
|
|
AcceleratorCudaInit: ================================================
|
|
local rank 3 device 0 bus id: 0000:C4:00.0
|
|
local rank 1 device 0 bus id: 0000:44:00.0
|
|
SharedMemoryMpi: World communicator of size 32
|
|
SharedMemoryMpi: Node communicator of size 4
|
|
0SharedMemoryMpi: SharedMemoryMPI.cc acceleratorAllocDevice 2147483648bytes at 0x14dde0000000 for comms buffers
|
|
Setting up IPC
|
|
|
|
__|__|__|__|__|__|__|__|__|__|__|__|__|__|__
|
|
__|__|__|__|__|__|__|__|__|__|__|__|__|__|__
|
|
__|_ | | | | | | | | | | | | _|__
|
|
__|_ _|__
|
|
__|_ GGGG RRRR III DDDD _|__
|
|
__|_ G R R I D D _|__
|
|
__|_ G R R I D D _|__
|
|
__|_ G GG RRRR I D D _|__
|
|
__|_ G G R R I D D _|__
|
|
__|_ GGGG R R III DDDD _|__
|
|
__|_ _|__
|
|
__|__|__|__|__|__|__|__|__|__|__|__|__|__|__
|
|
__|__|__|__|__|__|__|__|__|__|__|__|__|__|__
|
|
| | | | | | | | | | | | | |
|
|
|
|
|
|
Copyright (C) 2015 Peter Boyle, Azusa Yamaguchi, Guido Cossu, Antonin Portelli and other authors
|
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; either version 2 of the License, or
|
|
(at your option) any later version.
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
GNU General Public License for more details.
|
|
Current Grid git commit hash=188d2c7a4dc77807b545f5f2813cdb589b9e44ca: (HEAD -> develop, gh/develop, gh/HEAD) uncommited changes
|
|
|
|
Grid : Message : ================================================
|
|
Grid : Message : MPI is initialised and logging filters activated
|
|
Grid : Message : ================================================
|
|
Grid : Message : Requested 2147483648 byte stencil comms buffers
|
|
Grid : Message : MemoryManager Cache 34004218675 bytes
|
|
Grid : Message : MemoryManager::Init() setting up
|
|
Grid : Message : MemoryManager::Init() cache pool for recent allocations: SMALL 8 LARGE 2
|
|
Grid : Message : MemoryManager::Init() Non unified: Caching accelerator data in dedicated memory
|
|
Grid : Message : MemoryManager::Init() Using cudaMalloc
|
|
Grid : Message : 1.364287 s : Grid Layout
|
|
Grid : Message : 1.364291 s : Global lattice size : 64 64 64 128
|
|
Grid : Message : 1.364297 s : OpenMP threads : 4
|
|
Grid : Message : 1.364300 s : MPI tasks : 2 2 2 4
|
|
Grid : Message : 1.402623 s : Making s innermost grids
|
|
Grid : Message : 1.452917 s : Initialising 4d RNG
|
|
Grid : Message : 1.548452 s : Intialising parallel RNG with unique string 'The 4D RNG'
|
|
Grid : Message : 1.548478 s : Seed SHA256: 49db4542db694e3b1a74bf2592a8c1b83bfebbe18401693c2609a4c3af1
|
|
Grid : Message : 2.655959 s : Initialising 5d RNG
|
|
Grid : Message : 4.909760 s : Intialising parallel RNG with unique string 'The 5D RNG'
|
|
Grid : Message : 4.910540 s : Seed SHA256: b6316f2fac44ce14111f93e0296389330b077bfd0a7b359f781c58589f8a
|
|
Grid : Message : 21.205977 s : Initialised RNGs
|
|
Grid : Message : 24.842623 s : Drawing gauge field
|
|
Grid : Message : 25.835290 s : Random gauge initialised
|
|
Grid : Message : 25.856512 s : Setting up Cshift based reference
|
|
Grid : Message : 56.318033 s : *****************************************************************
|
|
Grid : Message : 56.318058 s : * Kernel options --dslash-generic, --dslash-unroll, --dslash-asm
|
|
Grid : Message : 56.318060 s : *****************************************************************
|
|
Grid : Message : 56.318065 s : *****************************************************************
|
|
Grid : Message : 56.318067 s : * Benchmarking DomainWallFermionR::Dhop
|
|
Grid : Message : 56.318069 s : * Vectorising space-time by 8
|
|
Grid : Message : 56.318071 s : * VComplexF size is 64 B
|
|
Grid : Message : 56.318073 s : * SINGLE precision
|
|
Grid : Message : 56.318075 s : * Using Overlapped Comms/Compute
|
|
Grid : Message : 56.318077 s : * Using GENERIC Nc WilsonKernels
|
|
Grid : Message : 56.318079 s : *****************************************************************
|
|
Grid : Message : 58.448027 s : Called warmup
|
|
Grid : Message : 579.884737 s : Called Dw 30000 times in 5.21434e+08 us
|
|
Grid : Message : 579.884799 s : mflop/s = 4.07724e+07
|
|
Grid : Message : 579.884801 s : mflop/s per rank = 1.27414e+06
|
|
Grid : Message : 579.884803 s : mflop/s per node = 5.09654e+06
|
|
Grid : Message : 579.884805 s : RF GiB/s (base 2) = 82848.5
|
|
Grid : Message : 579.884807 s : mem GiB/s (base 2) = 51780.3
|
|
Grid : Message : 579.890395 s : norm diff 1.07359e-13
|
|
Grid : Message : 579.961394 s : #### Dhop calls report
|
|
Grid : Message : 579.961401 s : WilsonFermion5D Number of DhopEO Calls : 60002
|
|
Grid : Message : 579.961404 s : WilsonFermion5D TotalTime /Calls : 8694.47 us
|
|
Grid : Message : 579.961406 s : WilsonFermion5D CommTime /Calls : 6578.27 us
|
|
Grid : Message : 579.961408 s : WilsonFermion5D FaceTime /Calls : 720.937 us
|
|
Grid : Message : 579.961410 s : WilsonFermion5D ComputeTime1/Calls : 5.01963 us
|
|
Grid : Message : 579.961412 s : WilsonFermion5D ComputeTime2/Calls : 1408.43 us
|
|
Grid : Message : 579.961423 s : Average mflops/s per call : 6.32207e+10
|
|
Grid : Message : 579.961427 s : Average mflops/s per call per rank : 1.97565e+09
|
|
Grid : Message : 579.961429 s : Average mflops/s per call per node : 7.90259e+09
|
|
Grid : Message : 579.961431 s : Average mflops/s per call (full) : 4.1495e+07
|
|
Grid : Message : 579.961433 s : Average mflops/s per call per rank (full): 1.29672e+06
|
|
Grid : Message : 579.961435 s : Average mflops/s per call per node (full): 5.18688e+06
|
|
Grid : Message : 579.961437 s : WilsonFermion5D Stencil
|
|
Grid : Message : 579.961438 s : WilsonFermion5D StencilEven
|
|
Grid : Message : 579.961439 s : WilsonFermion5D StencilOdd
|
|
Grid : Message : 579.961440 s : WilsonFermion5D Stencil Reporti()
|
|
Grid : Message : 579.961441 s : WilsonFermion5D StencilEven Reporti()
|
|
Grid : Message : 579.961443 s : WilsonFermion5D StencilOdd Reporti()
|
|
Grid : Message : 637.358875 s : Compare to naive wilson implementation Dag to verify correctness
|
|
Grid : Message : 637.358904 s : Called DwDag
|
|
Grid : Message : 637.358905 s : norm dag result 12.0421
|
|
Grid : Message : 637.379524 s : norm dag ref 12.0421
|
|
Grid : Message : 637.405427 s : norm dag diff 7.28475e-14
|
|
Grid : Message : 637.471220 s : Calling Deo and Doe and //assert Deo+Doe == Dunprec
|
|
Grid : Message : 637.865237 s : src_e0.499997
|
|
Grid : Message : 638.289523 s : src_o0.500003
|
|
Grid : Message : 638.406310 s : *********************************************************
|
|
Grid : Message : 638.406315 s : * Benchmarking DomainWallFermionF::DhopEO
|
|
Grid : Message : 638.406317 s : * Vectorising space-time by 8
|
|
Grid : Message : 638.406318 s : * SINGLE precision
|
|
Grid : Message : 638.406319 s : * Using Overlapped Comms/Compute
|
|
Grid : Message : 638.406320 s : * Using GENERIC Nc WilsonKernels
|
|
Grid : Message : 638.406321 s : *********************************************************
|
|
Grid : Message : 908.316704 s : Deo mflop/s = 3.93916e+07
|
|
Grid : Message : 908.316730 s : Deo mflop/s per rank 1.23099e+06
|
|
Grid : Message : 908.316732 s : Deo mflop/s per node 4.92395e+06
|
|
Grid : Message : 908.316734 s : #### Dhop calls report
|
|
Grid : Message : 908.316735 s : WilsonFermion5D Number of DhopEO Calls : 30001
|
|
Grid : Message : 908.316737 s : WilsonFermion5D TotalTime /Calls : 8996.53 us
|
|
Grid : Message : 908.316739 s : WilsonFermion5D CommTime /Calls : 6612.16 us
|
|
Grid : Message : 908.316741 s : WilsonFermion5D FaceTime /Calls : 846.374 us
|
|
Grid : Message : 908.316743 s : WilsonFermion5D ComputeTime1/Calls : 5.9754 us
|
|
Grid : Message : 908.316745 s : WilsonFermion5D ComputeTime2/Calls : 1562.46 us
|
|
Grid : Message : 908.316769 s : Average mflops/s per call : 5.29798e+10
|
|
Grid : Message : 908.316773 s : Average mflops/s per call per rank : 1.65562e+09
|
|
Grid : Message : 908.316775 s : Average mflops/s per call per node : 6.62248e+09
|
|
Grid : Message : 908.316778 s : Average mflops/s per call (full) : 4.01018e+07
|
|
Grid : Message : 908.316781 s : Average mflops/s per call per rank (full): 1.25318e+06
|
|
Grid : Message : 908.316783 s : Average mflops/s per call per node (full): 5.01273e+06
|
|
Grid : Message : 908.316786 s : WilsonFermion5D Stencil
|
|
Grid : Message : 908.316788 s : WilsonFermion5D StencilEven
|
|
Grid : Message : 908.316793 s : WilsonFermion5D StencilOdd
|
|
Grid : Message : 908.316795 s : WilsonFermion5D Stencil Reporti()
|
|
Grid : Message : 908.316796 s : WilsonFermion5D StencilEven Reporti()
|
|
Grid : Message : 908.316798 s : WilsonFermion5D StencilOdd Reporti()
|
|
Grid : Message : 908.402872 s : r_e6.02113
|
|
Grid : Message : 908.417198 s : r_o6.02101
|
|
Grid : Message : 908.427757 s : res12.0421
|
|
Grid : Message : 909.161595 s : norm diff 0
|
|
Grid : Message : 910.662440 s : norm diff even 0
|
|
Grid : Message : 910.480278 s : norm diff odd 0
|