1
0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-17 23:37:06 +01:00
This commit is contained in:
Peter Boyle
2021-09-21 01:04:30 +02:00
parent 1fb6aaf150
commit e188c0512e
7 changed files with 529 additions and 0 deletions

View File

@ -0,0 +1,156 @@
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 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-summit, --enable-select-gpu=no
AcceleratorCudaInit: ================================================
SharedMemoryMpi: World communicator of size 16
SharedMemoryMpi: Node communicator of size 4
0SharedMemoryMpi: SharedMemoryMPI.cc acceleratorAllocDevice 2147483648bytes at 0x14b7a0000000 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=f660dc67e4b193afc4015bc5e5fe47cfdbb0356e: (HEAD -> develop, origin/develop, origin/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 32 LARGE 8
Grid : Message : MemoryManager::Init() Non unified: Caching accelerator data in dedicated memory
Grid : Message : MemoryManager::Init() Using cudaMalloc
Grid : Message : 0.751288 s : Grid Layout
Grid : Message : 0.751291 s : Global lattice size : 64 64 64 64
Grid : Message : 0.751296 s : OpenMP threads : 4
Grid : Message : 0.751297 s : MPI tasks : 2 2 2 2
Grid : Message : 0.792527 s : Making s innermost grids
Grid : Message : 0.835495 s : Initialising 4d RNG
Grid : Message : 0.940402 s : Intialising parallel RNG with unique string 'The 4D RNG'
Grid : Message : 0.940421 s : Seed SHA256: 49db4542db694e3b1a74bf2592a8c1b83bfebbe18401693c2609a4c3af1
Grid : Message : 1.336448 s : Initialising 5d RNG
Grid : Message : 2.956230 s : Intialising parallel RNG with unique string 'The 5D RNG'
Grid : Message : 2.956251 s : Seed SHA256: b6316f2fac44ce14111f93e0296389330b077bfd0a7b359f781c58589f8a
Grid : Message : 11.242345 s : Initialised RNGs
Grid : Message : 13.415017 s : Drawing gauge field
Grid : Message : 13.937529 s : Random gauge initialised
Grid : Message : 15.529056 s : Setting up Cshift based reference
Grid : Message : 21.472100 s : *****************************************************************
Grid : Message : 21.472690 s : * Kernel options --dslash-generic, --dslash-unroll, --dslash-asm
Grid : Message : 21.472700 s : *****************************************************************
Grid : Message : 21.472710 s : *****************************************************************
Grid : Message : 21.472720 s : * Benchmarking DomainWallFermionR::Dhop
Grid : Message : 21.472730 s : * Vectorising space-time by 8
Grid : Message : 21.472740 s : * VComplexF size is 64 B
Grid : Message : 21.472750 s : * SINGLE precision
Grid : Message : 21.472760 s : * Using Overlapped Comms/Compute
Grid : Message : 21.472770 s : * Using GENERIC Nc WilsonKernels
Grid : Message : 21.472780 s : *****************************************************************
Grid : Message : 22.302206 s : Called warmup
Grid : Message : 53.261410 s : Called Dw 3000 times in 3.07237e+07 us
Grid : Message : 53.261970 s : mflop/s = 3.45988e+07
Grid : Message : 53.261990 s : mflop/s per rank = 2.16243e+06
Grid : Message : 53.262000 s : mflop/s per node = 8.64971e+06
Grid : Message : 53.262010 s : RF GiB/s (base 2) = 70304
Grid : Message : 53.262020 s : mem GiB/s (base 2) = 43940
Grid : Message : 53.297930 s : norm diff 1.03481e-13
Grid : Message : 53.647310 s : #### Dhop calls report
Grid : Message : 53.647360 s : WilsonFermion5D Number of DhopEO Calls : 6002
Grid : Message : 53.647410 s : WilsonFermion5D TotalTime /Calls : 5149.34 us
Grid : Message : 53.647430 s : WilsonFermion5D CommTime /Calls : 3752.8 us
Grid : Message : 53.647440 s : WilsonFermion5D FaceTime /Calls : 474.579 us
Grid : Message : 53.647450 s : WilsonFermion5D ComputeTime1/Calls : 46.5708 us
Grid : Message : 53.647460 s : WilsonFermion5D ComputeTime2/Calls : 926.982 us
Grid : Message : 53.647550 s : Average mflops/s per call : 3.82643e+09
Grid : Message : 53.647580 s : Average mflops/s per call per rank : 2.39152e+08
Grid : Message : 53.647590 s : Average mflops/s per call per node : 9.56608e+08
Grid : Message : 53.647610 s : Average mflops/s per call (full) : 3.50314e+07
Grid : Message : 53.647620 s : Average mflops/s per call per rank (full): 2.18946e+06
Grid : Message : 53.647630 s : Average mflops/s per call per node (full): 8.75785e+06
Grid : Message : 53.647640 s : WilsonFermion5D Stencil
Grid : Message : 53.647650 s : WilsonFermion5D StencilEven
Grid : Message : 53.647660 s : WilsonFermion5D StencilOdd
Grid : Message : 53.647670 s : WilsonFermion5D Stencil Reporti()
Grid : Message : 53.647680 s : WilsonFermion5D StencilEven Reporti()
Grid : Message : 53.647690 s : WilsonFermion5D StencilOdd Reporti()
Grid : Message : 80.460422 s : Compare to naive wilson implementation Dag to verify correctness
Grid : Message : 80.460440 s : Called DwDag
Grid : Message : 80.460441 s : norm dag result 12.0421
Grid : Message : 80.481170 s : norm dag ref 12.0421
Grid : Message : 80.497179 s : norm dag diff 7.63236e-14
Grid : Message : 80.536902 s : Calling Deo and Doe and //assert Deo+Doe == Dunprec
Grid : Message : 80.917100 s : src_e0.499997
Grid : Message : 80.993190 s : src_o0.500003
Grid : Message : 81.634480 s : *********************************************************
Grid : Message : 81.634500 s : * Benchmarking DomainWallFermionF::DhopEO
Grid : Message : 81.634510 s : * Vectorising space-time by 8
Grid : Message : 81.634520 s : * SINGLE precision
Grid : Message : 81.634530 s : * Using Overlapped Comms/Compute
Grid : Message : 81.634540 s : * Using GENERIC Nc WilsonKernels
Grid : Message : 81.634550 s : *********************************************************
Grid : Message : 96.356262 s : Deo mflop/s = 3.49003e+07
Grid : Message : 96.356288 s : Deo mflop/s per rank 2.18127e+06
Grid : Message : 96.356290 s : Deo mflop/s per node 8.72508e+06
Grid : Message : 96.356297 s : #### Dhop calls report
Grid : Message : 96.356298 s : WilsonFermion5D Number of DhopEO Calls : 3001
Grid : Message : 96.356300 s : WilsonFermion5D TotalTime /Calls : 5095.74 us
Grid : Message : 96.356302 s : WilsonFermion5D CommTime /Calls : 3589.98 us
Grid : Message : 96.356303 s : WilsonFermion5D FaceTime /Calls : 552.086 us
Grid : Message : 96.356304 s : WilsonFermion5D ComputeTime1/Calls : 52.4692 us
Grid : Message : 96.356305 s : WilsonFermion5D ComputeTime2/Calls : 963.892 us
Grid : Message : 96.356324 s : Average mflops/s per call : 3.42222e+09
Grid : Message : 96.356326 s : Average mflops/s per call per rank : 2.13889e+08
Grid : Message : 96.356327 s : Average mflops/s per call per node : 8.55556e+08
Grid : Message : 96.356328 s : Average mflops/s per call (full) : 3.53999e+07
Grid : Message : 96.356329 s : Average mflops/s per call per rank (full): 2.21249e+06
Grid : Message : 96.356330 s : Average mflops/s per call per node (full): 8.84997e+06
Grid : Message : 96.356331 s : WilsonFermion5D Stencil
Grid : Message : 96.356332 s : WilsonFermion5D StencilEven
Grid : Message : 96.356333 s : WilsonFermion5D StencilOdd
Grid : Message : 96.356334 s : WilsonFermion5D Stencil Reporti()
Grid : Message : 96.356335 s : WilsonFermion5D StencilEven Reporti()
Grid : Message : 96.356336 s : WilsonFermion5D StencilOdd Reporti()
Grid : Message : 96.415562 s : r_e6.02111
Grid : Message : 96.422386 s : r_o6.02102
Grid : Message : 96.428364 s : res12.0421
Grid : Message : 97.865450 s : norm diff 0
Grid : Message : 97.490738 s : norm diff even 0
Grid : Message : 97.561591 s : norm diff odd 0