mirror of
https://github.com/paboyle/Grid.git
synced 2024-11-14 09:45:36 +00:00
157 lines
9.5 KiB
Plaintext
157 lines
9.5 KiB
Plaintext
|
SLURM detected
|
||
|
|
AcceleratorCudaInit[0]: ========================
|
||
|
|
AcceleratorCudaInit[0]: Device Number : 0
|
||
|
|
AcceleratorCudaInit[0]: ========================
|
||
|
|
AcceleratorCudaInit[0]: Device identifier: A100-SXM4-40GB
|
||
|
|
AcceleratorCudaInit[0]: totalGlobalMem: 42506321920
|
||
|
|
AcceleratorCudaInit[0]: managedMemory: 1
|
||
|
|
AcceleratorCudaInit[0]: isMultiGpuBoard: 0
|
||
|
|
AcceleratorCudaInit[0]: warpSize: 32
|
||
|
|
AcceleratorCudaInit[0]: pciBusID: 2
|
||
|
|
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-setdevice=no
|
||
|
|
AcceleratorCudaInit: ================================================
|
||
|
|
SharedMemoryMpi: World communicator of size 16
|
||
|
|
SharedMemoryMpi: Node communicator of size 4
|
||
|
|
0SharedMemoryMpi: SharedMemoryMPI.cc acceleratorAllocDevice 2147483648bytes at 0x7f4b80000000 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=b2ccaad761798e93a9314f97d8a4d1f851c6962a: (HEAD -> develop) 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 34005057536 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 : 0.648397 s : Grid Layout
|
||
|
|
Grid : Message : 0.648398 s : Global lattice size : 64 64 64 64
|
||
|
|
Grid : Message : 0.648401 s : OpenMP threads : 32
|
||
|
|
Grid : Message : 0.648402 s : MPI tasks : 2 2 2 2
|
||
|
|
Grid : Message : 0.663662 s : Making s innermost grids
|
||
|
|
Grid : Message : 0.682145 s : Initialising 4d RNG
|
||
|
|
Grid : Message : 0.754321 s : Intialising parallel RNG with unique string 'The 4D RNG'
|
||
|
|
Grid : Message : 0.754332 s : Seed SHA256: 49db4542db694e3b1a74bf2592a8c1b83bfebbe18401693c2609a4c3af1
|
||
|
|
Grid : Message : 0.863265 s : Initialising 5d RNG
|
||
|
|
Grid : Message : 1.967677 s : Intialising parallel RNG with unique string 'The 5D RNG'
|
||
|
|
Grid : Message : 1.967691 s : Seed SHA256: b6316f2fac44ce14111f93e0296389330b077bfd0a7b359f781c58589f8a
|
||
|
|
Grid : Message : 2.921676 s : Initialised RNGs
|
||
|
|
Grid : Message : 4.382384 s : Drawing gauge field
|
||
|
|
Grid : Message : 4.672590 s : Random gauge initialised
|
||
|
|
Grid : Message : 6.102697 s : Setting up Cshift based reference
|
||
|
|
Grid : Message : 7.185897 s : *****************************************************************
|
||
|
|
Grid : Message : 7.185906 s : * Kernel options --dslash-generic, --dslash-unroll, --dslash-asm
|
||
|
|
Grid : Message : 7.185907 s : *****************************************************************
|
||
|
|
Grid : Message : 7.185907 s : *****************************************************************
|
||
|
|
Grid : Message : 7.185907 s : * Benchmarking DomainWallFermionR::Dhop
|
||
|
|
Grid : Message : 7.185907 s : * Vectorising space-time by 8
|
||
|
|
Grid : Message : 7.185908 s : * VComplexF size is 64 B
|
||
|
|
Grid : Message : 7.185909 s : * SINGLE precision
|
||
|
|
Grid : Message : 7.185909 s : * Using Overlapped Comms/Compute
|
||
|
|
Grid : Message : 7.185909 s : * Using GENERIC Nc WilsonKernels
|
||
|
|
Grid : Message : 7.185909 s : *****************************************************************
|
||
|
|
Grid : Message : 8.114241 s : Called warmup
|
||
|
|
Grid : Message : 83.988100 s : Called Dw 3000 times in 7.48954e+07 us
|
||
|
|
Grid : Message : 83.992400 s : mflop/s = 1.41932e+07
|
||
|
|
Grid : Message : 83.992600 s : mflop/s per rank = 887074
|
||
|
|
Grid : Message : 83.992700 s : mflop/s per node = 3.5483e+06
|
||
|
|
Grid : Message : 83.992800 s : RF GiB/s (base 2) = 28840.2
|
||
|
|
Grid : Message : 83.992900 s : mem GiB/s (base 2) = 18025.1
|
||
|
|
Grid : Message : 83.134870 s : norm diff 1.03481e-13
|
||
|
|
Grid : Message : 83.493960 s : #### Dhop calls report
|
||
|
|
Grid : Message : 83.494000 s : WilsonFermion5D Number of DhopEO Calls : 6002
|
||
|
|
Grid : Message : 83.494030 s : WilsonFermion5D TotalTime /Calls : 12506 us
|
||
|
|
Grid : Message : 83.494040 s : WilsonFermion5D CommTime /Calls : 11071.5 us
|
||
|
|
Grid : Message : 83.494050 s : WilsonFermion5D FaceTime /Calls : 530.971 us
|
||
|
|
Grid : Message : 83.494060 s : WilsonFermion5D ComputeTime1/Calls : 23.6428 us
|
||
|
|
Grid : Message : 83.494070 s : WilsonFermion5D ComputeTime2/Calls : 911.864 us
|
||
|
|
Grid : Message : 83.494220 s : Average mflops/s per call : 7.6108e+09
|
||
|
|
Grid : Message : 83.494250 s : Average mflops/s per call per rank : 4.75675e+08
|
||
|
|
Grid : Message : 83.494260 s : Average mflops/s per call per node : 1.9027e+09
|
||
|
|
Grid : Message : 83.494270 s : Average mflops/s per call (full) : 1.44242e+07
|
||
|
|
Grid : Message : 83.494280 s : Average mflops/s per call per rank (full): 901513
|
||
|
|
Grid : Message : 83.494290 s : Average mflops/s per call per node (full): 3.60605e+06
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D Stencil
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D StencilEven
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D StencilOdd
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D Stencil Reporti()
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D StencilEven Reporti()
|
||
|
|
Grid : Message : 83.494300 s : WilsonFermion5D StencilOdd Reporti()
|
||
|
|
Grid : Message : 94.600488 s : Compare to naive wilson implementation Dag to verify correctness
|
||
|
|
Grid : Message : 94.600501 s : Called DwDag
|
||
|
|
Grid : Message : 94.600502 s : norm dag result 12.0421
|
||
|
|
Grid : Message : 94.613445 s : norm dag ref 12.0421
|
||
|
|
Grid : Message : 94.628514 s : norm dag diff 7.63236e-14
|
||
|
|
Grid : Message : 94.666370 s : Calling Deo and Doe and //assert Deo+Doe == Dunprec
|
||
|
|
Grid : Message : 95.136361 s : src_e0.499997
|
||
|
|
Grid : Message : 95.208108 s : src_o0.500003
|
||
|
|
Grid : Message : 95.271511 s : *********************************************************
|
||
|
|
Grid : Message : 95.271512 s : * Benchmarking DomainWallFermionF::DhopEO
|
||
|
|
Grid : Message : 95.271513 s : * Vectorising space-time by 8
|
||
|
|
Grid : Message : 95.271514 s : * SINGLE precision
|
||
|
|
Grid : Message : 95.271514 s : * Using Overlapped Comms/Compute
|
||
|
|
Grid : Message : 95.271515 s : * Using GENERIC Nc WilsonKernels
|
||
|
|
Grid : Message : 95.271515 s : *********************************************************
|
||
|
|
Grid : Message : 132.766274 s : Deo mflop/s = 1.41952e+07
|
||
|
|
Grid : Message : 132.766295 s : Deo mflop/s per rank 887201
|
||
|
|
Grid : Message : 132.766297 s : Deo mflop/s per node 3.5488e+06
|
||
|
|
Grid : Message : 132.766299 s : #### Dhop calls report
|
||
|
|
Grid : Message : 132.766300 s : WilsonFermion5D Number of DhopEO Calls : 3001
|
||
|
|
Grid : Message : 132.766301 s : WilsonFermion5D TotalTime /Calls : 12493.9 us
|
||
|
|
Grid : Message : 132.766302 s : WilsonFermion5D CommTime /Calls : 10990.2 us
|
||
|
|
Grid : Message : 132.766303 s : WilsonFermion5D FaceTime /Calls : 604.889 us
|
||
|
|
Grid : Message : 132.766304 s : WilsonFermion5D ComputeTime1/Calls : 13.7158 us
|
||
|
|
Grid : Message : 132.766305 s : WilsonFermion5D ComputeTime2/Calls : 920.096 us
|
||
|
|
Grid : Message : 132.766326 s : Average mflops/s per call : 1.31121e+10
|
||
|
|
Grid : Message : 132.766328 s : Average mflops/s per call per rank : 8.19504e+08
|
||
|
|
Grid : Message : 132.766329 s : Average mflops/s per call per node : 3.27802e+09
|
||
|
|
Grid : Message : 132.766330 s : Average mflops/s per call (full) : 1.44381e+07
|
||
|
|
Grid : Message : 132.766331 s : Average mflops/s per call per rank (full): 902382
|
||
|
|
Grid : Message : 132.766332 s : Average mflops/s per call per node (full): 3.60953e+06
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D Stencil
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D StencilEven
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D StencilOdd
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D Stencil Reporti()
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D StencilEven Reporti()
|
||
|
|
Grid : Message : 132.766333 s : WilsonFermion5D StencilOdd Reporti()
|
||
|
|
Grid : Message : 132.847999 s : r_e6.02111
|
||
|
|
Grid : Message : 132.854237 s : r_o6.02102
|
||
|
|
Grid : Message : 132.860309 s : res12.0421
|
||
|
|
Grid : Message : 133.458462 s : norm diff 0
|
||
|
|
Grid : Message : 133.832713 s : norm diff even 0
|
||
|
|
Grid : Message : 133.909147 s : norm diff odd 0
|