=== This is the CP2K Performance-Test ===
Already up to date.
Current branch master is up to date.
Already up to date.
Current branch master is up to date.
GIT Revision: 18d5c6eecb1583739a723fdc877dd6e49c5a3a43
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.6, ELPA 2022.11.001, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.2.2, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.2,
# SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Last update: 21.06.2023
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2022.11.001
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.2
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
HDF5_VER := $(strip $(HDF5_VER))
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(HDF5_VER)/lib
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__HDF5
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/01
job id: 47814358
--- Point ---
name: 10
plot: h2o_32_ri_rpa_mp2
regex: Total RI-RPA Time=
label: RI-RPA (8n/2r/6t)
--- Point ---
name: 11
plot: h2o_32_ri_rpa_mp2_mem
regex: Estimated peak process memory
label: RI-RPA (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-MP2.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-HF-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-MP2-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/02
job id: 47814362
--- Point ---
name: 20
plot: h2o_32_ri_rpa_mp2
regex: Total MP2 Time=
label: RI-MP2 (8n/6r/2t)
--- Point ---
name: 21
plot: h2o_32_ri_rpa_mp2_mem
regex: Estimated peak process memory
label: RI-MP2 (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/03
job id: 47814367
--- Point ---
name: 100
plot: h2o_64_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 101
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/04
job id: 47814370
--- Point ---
name: 102
plot: h2o_64_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 103
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/05
job id: 47814373
--- Point ---
name: 104
plot: h2o_64_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 105
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/06
job id: 47814375
--- Point ---
name: 106
plot: h2o_64_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 107
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/07
job id: 47814378
--- Point ---
name: 108
plot: h2o_64_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 109
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/08
job id: 47814379
--- Point ---
name: 110
plot: h2o_64_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 111
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/09
job id: 47814380
--- Point ---
name: 200
plot: h2o_128_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 201
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/10
job id: 47814381
--- Point ---
name: 202
plot: h2o_128_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 203
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/11
job id: 47814382
--- Point ---
name: 204
plot: h2o_128_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 205
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/12
job id: 47814383
--- Point ---
name: 206
plot: h2o_128_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 207
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/13
job id: 47814385
--- Point ---
name: 208
plot: h2o_128_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 209
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/14
job id: 47814386
--- Point ---
name: 210
plot: h2o_128_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 211
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/15
job id: 47814387
--- Point ---
name: 400
plot: h2o_256_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 401
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/16
job id: 47814388
--- Point ---
name: 402
plot: h2o_256_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 403
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/17
job id: 47814389
--- Point ---
name: 404
plot: h2o_256_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 405
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/18
job id: 47814390
--- Point ---
name: 406
plot: h2o_256_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 407
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/19
job id: 47814391
--- Point ---
name: 408
plot: h2o_256_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 409
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/20
job id: 47814392
--- Point ---
name: 410
plot: h2o_256_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 411
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/21
job id: 47814393
--- Point ---
name: 500
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 501
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/22
job id: 47814394
--- Point ---
name: 502
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 503
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/23
job id: 47814395
--- Point ---
name: 504
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 505
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/24
job id: 47814396
--- Point ---
name: 506
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 507
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/25
job id: 47814398
--- Point ---
name: 508
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 509
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/26
job id: 47814399
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: 512 H2O (4 NVE MD steps on 64 nodes)
input file: benchmarks/QS/00512_H2O/H2O-512_md.inp
required files: []
output file: result.log
# nodes = 64
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/27
job id: 47814400
--- Point ---
name: 601
plot: h2o_512_md
regex: CP2K
label: (64n/12r/1t)
--- Point ---
name: 602
plot: h2o_512_md_mem
regex: Estimated peak process memory
label: (64n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 424 8.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.040 134.671 134.672
farming_run 1 2.0 134.230 134.233 134.637 134.640
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 4194304 0.0% 0.0% 100.0%
flops 14 x 32 x 32 154140672 0.0% 0.0% 100.0%
flops 29 x 32 x 32 159645696 0.0% 0.0% 100.0%
flops 14 x 14 x 32 208732160 0.0% 0.0% 100.0%
flops 29 x 14 x 32 212860928 0.0% 0.0% 100.0%
flops 14 x 29 x 32 212860928 0.0% 0.0% 100.0%
flops 29 x 29 x 32 227352576 0.0% 0.0% 100.0%
flops 14 x 32 x 14 896801644032 0.0% 0.0% 100.0%
flops 29 x 32 x 14 928925089792 0.0% 0.0% 100.0%
flops 14 x 32 x 29 928925089792 0.0% 0.0% 100.0%
flops 29 x 32 x 29 962100985856 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693169221632 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753639550976 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.164741E+12 0.0% 0.0% 100.0%
flops max/rank 447.801317E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249492158 0.0% 0.0% 100.0%
number of processed stacks 164328 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1518.3
marketing flops 7.165779E+12
-------------------------------------------------------------------------------
# multiplications 1160
max memory usage/rank 1.490190E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1134128.
MP_Allreduce 571 1938539.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.027 115.231 115.231
qs_energies 1 2.0 0.000 0.000 115.033 115.037
mp2_main 1 3.0 0.000 0.000 112.719 112.722
mp2_gpw_main 1 4.0 0.027 0.033 111.616 111.620
mp2_ri_gpw_compute_in 1 5.0 0.172 0.174 92.684 93.062
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 55.002 55.380
mp2_eri_3c_integrate_gpw 272 7.0 0.151 0.170 41.395 46.467
get_2c_integrals 1 6.0 0.008 0.009 37.002 37.512
integrate_v_rspace 273 8.0 0.441 0.457 24.790 29.687
pw_transfer 6555 10.6 0.374 0.394 27.451 28.098
fft_wrap_pw1pw2 5465 11.4 0.045 0.048 26.059 26.625
grid_integrate_task_list 273 9.0 20.589 25.979 20.589 25.979
fft_wrap_pw1pw2_100 2178 12.4 1.188 1.427 23.588 24.148
compute_2c_integrals 1 7.0 0.002 0.002 19.712 19.714
compute_2c_integrals_loop_lm 1 8.0 0.002 0.003 18.944 19.431
mp2_eri_2c_integrate_gpw 1 9.0 2.367 2.431 18.942 19.428
rpa_ri_compute_en 1 5.0 0.019 0.023 18.816 19.051
cp_fm_cholesky_decompose 12 8.2 17.254 17.753 17.254 17.753
fft3d_s 5443 13.4 16.189 16.651 16.211 16.672
cholesky_decomp 1 7.0 0.000 0.000 16.123 16.627
ao_to_mo_and_store_B_mult_1 272 7.0 10.769 15.303 10.769 15.303
calculate_wavefunction 272 8.0 5.400 5.556 12.502 13.108
rpa_num_int 1 6.0 0.000 0.000 10.679 10.689
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.641 10.669
calc_mat_Q 8 8.0 0.000 0.000 9.469 9.580
contract_S_to_Q 8 9.0 0.000 0.000 8.890 9.003
calc_potential_gpw 544 9.5 0.005 0.005 8.311 8.745
parallel_gemm_fm 14 9.1 0.000 0.000 8.500 8.601
parallel_gemm_fm_cosma 14 10.1 8.500 8.601 8.500 8.601
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.001 8.253 8.563
potential_pw2rs 545 10.0 0.107 0.109 7.706 8.455
collocate_single_gaussian 272 10.0 0.039 0.042 7.509 7.778
create_integ_mat 1 6.0 0.014 0.028 7.667 7.676
array2fm 1 7.0 0.000 0.000 6.610 7.083
pw_scatter_s 2720 13.7 4.404 4.656 4.404 4.656
pw_gather_s 2722 13.2 3.873 4.234 3.873 4.234
array2fm_buffer_send 1 8.0 2.849 3.101 2.849 3.101
pw_poisson_solve 545 10.5 1.098 1.134 2.198 2.430
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=111.615714, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2808.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 424 9.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.055 396.423 396.425
farming_run 1 2.0 395.585 395.590 396.374 396.384
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 16777216 0.0% 0.0% 100.0%
flops 14 x 32 x 32 565182464 0.0% 0.0% 100.0%
flops 29 x 32 x 32 585367552 0.0% 0.0% 100.0%
flops 14 x 14 x 32 626196480 0.0% 0.0% 100.0%
flops 29 x 14 x 32 638582784 0.0% 0.0% 100.0%
flops 14 x 29 x 32 638582784 0.0% 0.0% 100.0%
flops 29 x 29 x 32 682057728 0.0% 0.0% 100.0%
flops 14 x 32 x 14 897827141120 0.0% 0.0% 100.0%
flops 29 x 32 x 14 929989394432 0.0% 0.0% 100.0%
flops 14 x 32 x 29 929989394432 0.0% 0.0% 100.0%
flops 29 x 32 x 29 963203301376 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693481172992 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753962643456 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.172206E+12 0.0% 0.0% 100.0%
flops max/rank 150.696064E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249788822 0.0% 0.0% 100.0%
number of processed stacks 98736 0.0% 0.0% 100.0%
average stack size 0.0 0.0 2529.9
marketing flops 7.174951E+12
-------------------------------------------------------------------------------
# multiplications 1140
max memory usage/rank 1.234612E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 757 478553.
MP_Allreduce 2021 21391.
MP_Sync 37
MP_Alltoall 77
MP_SendRecv 2876 2171486.
MP_ISendRecv 1034 172620.
MP_Wait 1346
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.033 209.128 209.129
qs_energies 1 2.0 0.004 0.031 208.870 208.884
scf_env_do_scf 1 3.0 0.000 0.000 106.209 106.210
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.271 105.280
rebuild_ks_matrix 4 6.0 0.000 0.000 105.270 105.278
qs_ks_build_kohn_sham_matrix 4 7.0 0.054 0.063 105.270 105.278
hfx_ks_matrix 4 8.0 0.001 0.001 104.883 104.887
integrate_four_center 4 9.0 0.142 0.464 104.882 104.887
mp2_main 1 3.0 0.007 0.057 102.351 102.359
mp2_gpw_main 1 4.0 0.054 0.122 101.344 101.358
integrate_four_center_main 4 10.0 0.082 0.488 96.706 99.169
integrate_four_center_bin 266 11.0 96.623 99.159 96.623 99.159
init_scf_loop 1 4.0 0.000 0.000 91.841 91.841
mp2_ri_gpw_compute_in 1 5.0 0.070 0.117 74.438 75.548
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.013 54.023 55.133
mp2_eri_3c_integrate_gpw 91 7.0 0.144 0.166 41.437 46.842
integrate_v_rspace 95 8.0 0.396 0.565 27.814 32.966
pw_transfer 2240 10.6 0.143 0.166 29.943 30.303
fft_wrap_pw1pw2 1868 11.4 0.018 0.020 28.952 29.345
ao_to_mo_and_store_B_mult_1 91 7.0 10.907 28.556 10.907 28.556
grid_integrate_task_list 95 9.0 23.105 28.460 23.105 28.460
mp2_ri_gpw_compute_en 1 5.0 0.062 0.135 26.675 28.304
fft_wrap_pw1pw2_100 730 12.4 1.270 1.431 26.664 27.050
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.835 1.895 24.857 24.869
get_2c_integrals 1 6.0 0.001 0.012 20.319 20.352
compute_2c_integrals 1 7.0 0.003 0.004 19.297 19.309
compute_2c_integrals_loop_lm 1 8.0 0.002 0.008 18.862 19.176
mp2_eri_2c_integrate_gpw 1 9.0 1.739 1.842 18.860 19.174
fft3d_s 1823 13.4 18.444 18.938 18.458 18.951
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.366 14.366
calculate_wavefunction 91 8.0 2.017 2.054 9.743 9.977
mp2_ri_gpw_compute_en_expansio 172 7.0 0.559 0.594 8.745 9.319
potential_pw2rs 186 10.0 0.033 0.036 8.653 9.257
local_gemm 172 8.0 8.186 8.726 8.186 8.726
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.002 0.022 8.246 8.624
mp2_ri_gpw_compute_en_comm 22 7.0 0.501 0.520 7.892 8.359
collocate_single_gaussian 91 10.0 0.017 0.020 7.886 8.152
calc_potential_gpw 182 9.5 0.002 0.002 7.900 8.084
mp2_ri_gpw_compute_en_ener 172 7.0 6.341 6.449 6.341 6.449
mp_sendrecv_dm3 2068 8.0 5.920 6.382 5.920 6.382
pw_gather_s 912 13.2 4.926 5.481 4.926 5.481
mp_sync 37 10.5 2.852 5.222 2.852 5.222
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.349914, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1494.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/03/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 29.277748E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 5055360 0.0% 0.0% 100.0%
average stack size 0.0 0.0 29.1
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 454.041600E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62379.
MP_Allreduce 10329 270.
MP_Sync 530
MP_Alltoall 2083 588782.
MP_SendRecv 22610 5520.
MP_ISendRecv 22610 5520.
MP_Wait 37876
MP_comm_split 50
MP_ISend 20771 42672.
MP_IRecv 20771 42672.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.022 0.056 56.225 56.227
qs_mol_dyn_low 1 2.0 0.004 0.009 55.847 55.879
qs_forces 11 3.9 0.003 0.009 54.062 54.066
qs_energies 11 4.9 0.003 0.011 52.504 52.525
scf_env_do_scf 11 5.9 0.001 0.001 46.450 46.450
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 44.353 44.353
qs_scf_new_mos 108 7.5 0.000 0.001 33.753 34.051
qs_scf_loop_do_ot 108 8.5 0.001 0.001 33.752 34.051
dbcsr_multiply_generic 2286 12.5 0.094 0.098 33.285 33.659
ot_scf_mini 108 9.5 0.002 0.003 32.084 32.310
velocity_verlet 10 3.0 0.001 0.002 27.595 27.613
multiply_cannon 2286 13.5 0.190 0.196 26.011 27.440
multiply_cannon_loop 2286 14.5 1.818 1.934 25.296 26.752
ot_mini 108 10.5 0.001 0.001 19.222 19.472
qs_ot_get_derivative 108 11.5 0.001 0.002 16.233 16.410
mp_waitall_1 245248 16.5 8.332 14.445 8.332 14.445
multiply_cannon_metrocomm3 54864 15.5 0.073 0.080 5.884 12.862
multiply_cannon_multrec 54864 15.5 3.696 5.791 7.616 11.023
qs_ot_get_p 119 10.4 0.001 0.001 8.194 8.477
rebuild_ks_matrix 119 8.3 0.000 0.000 8.184 8.317
qs_ks_build_kohn_sham_matrix 119 9.3 0.011 0.017 8.183 8.317
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.200 7.324
mp_sum_l 7287 12.8 5.271 6.794 5.271 6.794
multiply_cannon_sync_h2d 54864 15.5 5.183 5.919 5.183 5.919
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.646 5.756
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.257 5.702
dbcsr_mm_accdrv_process 76910 16.1 1.792 2.852 3.834 5.524
qs_ot_p2m_diag 50 11.0 0.004 0.007 5.413 5.447
sum_up_and_integrate 119 10.3 0.012 0.017 4.720 4.727
integrate_v_rspace 119 11.3 0.002 0.003 4.708 4.716
init_scf_run 11 5.9 0.000 0.001 4.678 4.678
scf_env_initial_rho_setup 11 6.9 0.000 0.002 4.678 4.678
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.492 4.662
calculate_rho_elec 119 8.7 0.012 0.018 4.491 4.661
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.559 4.559
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.327 4.328
cp_fm_redistribute_end 50 14.0 2.205 4.295 2.215 4.300
cp_fm_diag_elpa_base 50 14.0 2.077 4.182 2.082 4.189
apply_preconditioner_dbcsr 119 12.6 0.000 0.001 2.935 3.136
apply_single 119 13.6 0.000 0.000 2.935 3.136
rs_pw_transfer 974 11.9 0.012 0.014 2.995 3.120
calculate_dm_sparse 119 9.5 0.000 0.001 2.886 3.047
multiply_cannon_metrocomm1 54864 15.5 0.057 0.062 1.666 2.840
acc_transpose_blocks 54864 15.5 0.227 0.255 2.195 2.745
ot_diis_step 108 11.5 0.009 0.033 2.717 2.738
jit_kernel_multiply 13 15.8 1.981 2.592 1.981 2.592
density_rs2pw 119 9.7 0.004 0.004 2.390 2.527
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.429 2.488
calculate_first_density_matrix 1 7.0 0.000 0.003 2.431 2.435
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 2.233 2.236
wfi_extrapolate 11 7.9 0.001 0.002 2.173 2.173
grid_integrate_task_list 119 12.3 2.014 2.130 2.014 2.130
init_scf_loop 11 6.9 0.001 0.004 2.078 2.078
potential_pw2rs 119 12.3 0.004 0.004 2.056 2.078
mp_sum_d 4135 12.0 1.380 2.029 1.380 2.029
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.936 1.990
pw_transfer 1439 11.6 0.052 0.057 1.897 1.983
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 1.821 1.911
mp_sum_dm 438 4.9 1.809 1.862 1.809 1.862
md_write_output 11 3.9 0.019 1.770 0.021 1.785
md_output 10 3.0 0.000 0.000 0.022 1.784
update_particle_set 20 4.0 0.000 0.001 1.740 1.759
make_m2s 4572 13.5 0.053 0.055 1.712 1.757
make_images 4572 14.5 0.134 0.139 1.630 1.674
mp_alltoall_d11v 2130 13.8 1.367 1.555 1.367 1.555
fft3d_ps 1201 14.6 0.370 0.480 1.461 1.549
fft_wrap_pw1pw2_140 487 13.2 0.187 0.203 1.410 1.499
acc_transpose_blocks_sync 164592 16.5 1.186 1.443 1.186 1.443
grid_collocate_task_list 119 9.7 1.355 1.438 1.355 1.438
mp_waitany 12084 13.8 1.283 1.426 1.283 1.426
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.001 1.387 1.415
dbcsr_dot_sd 1205 11.9 0.049 0.059 0.761 1.172
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=56.227000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=432.818182, yerr=1.585054
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 487.337984E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10306 303.
MP_Sync 54
MP_Alltoall 2060 1034122.
MP_SendRecv 16779 37093.
MP_ISendRecv 16779 37093.
MP_Wait 23539
MP_comm_split 50
MP_ISend 5720 128509.
MP_IRecv 5720 128509.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.026 39.869 39.870
qs_mol_dyn_low 1 2.0 0.003 0.003 39.684 39.705
qs_forces 11 3.9 0.002 0.003 38.888 38.889
qs_energies 11 4.9 0.001 0.002 37.179 37.183
scf_env_do_scf 11 5.9 0.001 0.001 31.990 31.991
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 29.585 29.585
dbcsr_multiply_generic 2286 12.5 0.101 0.103 22.130 22.478
qs_scf_new_mos 108 7.5 0.001 0.001 20.503 20.736
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.502 20.736
ot_scf_mini 108 9.5 0.002 0.003 19.593 19.761
velocity_verlet 10 3.0 0.001 0.002 19.195 19.211
multiply_cannon 2286 13.5 0.210 0.220 17.109 18.636
multiply_cannon_loop 2286 14.5 1.195 1.259 15.966 17.381
ot_mini 108 10.5 0.001 0.001 12.103 12.334
mp_waitall_1 200699 16.5 5.523 10.700 5.523 10.700
qs_ot_get_derivative 108 11.5 0.001 0.001 9.745 9.914
multiply_cannon_metrocomm3 27432 15.5 0.071 0.074 4.068 9.499
multiply_cannon_multrec 27432 15.5 1.835 4.266 6.282 9.087
rebuild_ks_matrix 119 8.3 0.000 0.000 7.191 7.332
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 7.191 7.332
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.331 6.460
dbcsr_mm_accdrv_process 47894 16.0 3.559 5.769 4.366 6.460
qs_ot_get_p 119 10.4 0.001 0.001 4.593 4.813
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.715 4.565
sum_up_and_integrate 119 10.3 0.025 0.028 4.101 4.109
integrate_v_rspace 119 11.3 0.002 0.003 4.076 4.084
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.014 4.070
apply_single 119 13.6 0.000 0.000 3.013 4.070
mp_sum_l 7287 12.8 2.003 4.065 2.003 4.065
init_scf_run 11 5.9 0.000 0.001 4.005 4.006
scf_env_initial_rho_setup 11 6.9 0.000 0.001 4.005 4.005
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.743 3.776
calculate_rho_elec 119 8.7 0.021 0.024 3.743 3.776
qs_ot_p2m_diag 50 11.0 0.009 0.013 2.980 2.999
make_m2s 4572 13.5 0.052 0.054 2.564 2.899
rs_pw_transfer 974 11.9 0.011 0.012 2.431 2.851
make_images 4572 14.5 0.205 0.245 2.476 2.813
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.584 2.585
calculate_first_density_matrix 1 7.0 0.000 0.001 2.499 2.500
density_rs2pw 119 9.7 0.004 0.004 2.004 2.443
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.300 2.387
init_scf_loop 11 6.9 0.000 0.000 2.387 2.387
multiply_cannon_sync_h2d 27432 15.5 1.708 2.322 1.708 2.322
ot_diis_step 108 11.5 0.011 0.011 2.312 2.312
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.274 2.275
cp_fm_redistribute_end 50 14.0 1.152 2.250 1.156 2.252
calculate_dm_sparse 119 9.5 0.000 0.001 2.154 2.236
cp_fm_diag_elpa_base 50 14.0 1.064 2.159 1.093 2.199
jit_kernel_multiply 10 16.4 0.749 2.077 0.749 2.077
acc_transpose_blocks 27432 15.5 0.115 0.121 1.678 2.035
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.974 1.976
grid_integrate_task_list 119 12.3 1.844 1.955 1.844 1.955
pw_transfer 1439 11.6 0.065 0.069 1.877 1.905
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.787 1.826
potential_pw2rs 119 12.3 0.006 0.006 1.818 1.825
fft_wrap_pw1pw2 1201 12.6 0.008 0.009 1.785 1.814
make_images_data 4572 15.5 0.048 0.054 1.232 1.583
prepare_preconditioner 11 7.9 0.000 0.000 1.470 1.496
make_preconditioner 11 8.9 0.000 0.000 1.470 1.496
fft_wrap_pw1pw2_140 487 13.2 0.206 0.216 1.432 1.463
hybrid_alltoall_any 4725 16.4 0.053 0.116 1.096 1.455
wfi_extrapolate 11 7.9 0.001 0.001 1.449 1.449
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.373 1.428
grid_collocate_task_list 119 9.7 1.282 1.419 1.282 1.419
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.352 1.360
fft3d_ps 1201 14.6 0.519 0.575 1.331 1.354
mp_alltoall_d11v 2130 13.8 1.176 1.274 1.176 1.274
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.188 1.234
mp_sum_d 4135 12.0 0.522 0.994 0.522 0.994
mp_allgather_i34 2286 14.5 0.560 0.970 0.560 0.970
rs_pw_transfer_RS2PW_140 130 11.5 0.141 0.150 0.533 0.951
acc_transpose_blocks_sync 82296 16.5 0.824 0.949 0.824 0.949
acc_transpose_blocks_kernels 27432 16.5 0.187 0.275 0.714 0.943
mp_waitany 5720 13.7 0.509 0.929 0.509 0.929
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.916 0.929
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.924 0.925
mp_sum_dm 438 4.9 0.789 0.832 0.789 0.832
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=39.870000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=464.545455, yerr=1.157084
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 521.224192E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63490.
MP_Allreduce 10155 305.
MP_Sync 54
MP_Alltoall 1821 972933.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.053 0.088 34.616 34.617
qs_mol_dyn_low 1 2.0 0.003 0.004 34.293 34.301
qs_forces 11 3.9 0.002 0.002 34.228 34.234
qs_energies 11 4.9 0.002 0.006 32.578 32.584
scf_env_do_scf 11 5.9 0.001 0.003 27.667 27.667
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 24.897 24.898
dbcsr_multiply_generic 2286 12.5 0.095 0.097 17.378 17.469
qs_scf_new_mos 108 7.5 0.001 0.001 16.303 16.318
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.303 16.317
velocity_verlet 10 3.0 0.001 0.002 16.251 16.253
ot_scf_mini 108 9.5 0.002 0.003 15.528 15.545
multiply_cannon 2286 13.5 0.194 0.199 13.787 14.599
multiply_cannon_loop 2286 14.5 0.858 0.888 12.930 13.799
ot_mini 108 10.5 0.001 0.001 9.359 9.378
qs_ot_get_derivative 108 11.5 0.001 0.001 7.826 7.841
multiply_cannon_multrec 18288 15.5 1.873 2.887 7.041 7.421
rebuild_ks_matrix 119 8.3 0.000 0.000 6.715 6.736
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.015 6.715 6.736
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.922 5.940
dbcsr_mm_accdrv_process 38222 16.0 4.915 5.836 5.074 5.901
sum_up_and_integrate 119 10.3 0.031 0.031 4.172 4.178
integrate_v_rspace 119 11.3 0.003 0.003 4.141 4.151
mp_waitall_1 158411 16.6 2.814 3.984 2.814 3.984
qs_ot_get_p 119 10.4 0.001 0.001 3.880 3.903
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.721 3.729
calculate_rho_elec 119 8.7 0.031 0.032 3.720 3.729
init_scf_run 11 5.9 0.000 0.001 3.679 3.679
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.678 3.679
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.912 3.528
rs_pw_transfer 974 11.9 0.010 0.011 2.540 2.794
init_scf_loop 11 6.9 0.001 0.005 2.746 2.748
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.612 2.619
multiply_cannon_metrocomm3 18288 15.5 0.047 0.048 1.422 2.494
density_rs2pw 119 9.7 0.004 0.005 2.198 2.470
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.050 2.373
apply_single 119 13.6 0.000 0.000 2.050 2.373
calculate_first_density_matrix 1 7.0 0.000 0.003 2.344 2.345
cp_dbcsr_syevd 50 12.0 0.003 0.006 2.290 2.291
make_m2s 4572 13.5 0.044 0.045 2.010 2.136
pw_transfer 1439 11.6 0.065 0.069 2.070 2.099
make_images 4572 14.5 0.193 0.206 1.925 2.050
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.978 2.010
calculate_dm_sparse 119 9.5 0.000 0.000 1.983 1.993
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.976 1.987
cp_fm_diag_elpa_base 50 14.0 1.948 1.964 1.972 1.983
potential_pw2rs 119 12.3 0.007 0.008 1.902 1.916
grid_integrate_task_list 119 12.3 1.796 1.884 1.796 1.884
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.815 1.827
prepare_preconditioner 11 7.9 0.000 0.000 1.810 1.813
make_preconditioner 11 8.9 0.000 0.002 1.810 1.813
make_full_inverse_cholesky 11 9.9 0.000 0.001 1.662 1.745
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.708 1.709
acc_transpose_blocks 18288 15.5 0.079 0.082 1.626 1.670
mp_sum_l 7287 12.8 1.223 1.594 1.223 1.594
fft_wrap_pw1pw2_140 487 13.2 0.256 0.264 1.531 1.558
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.520 1.529
ot_diis_step 108 11.5 0.011 0.011 1.511 1.511
fft3d_ps 1201 14.6 0.530 0.547 1.462 1.494
grid_collocate_task_list 119 9.7 1.246 1.380 1.246 1.380
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.313 1.318
wfi_extrapolate 11 7.9 0.001 0.001 1.261 1.261
multiply_cannon_sync_h2d 18288 15.5 1.030 1.151 1.030 1.151
make_images_data 4572 15.5 0.047 0.051 0.863 1.021
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.987 1.008
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 0.978 0.983
mp_alltoall_d11v 2130 13.8 0.783 0.932 0.783 0.932
hybrid_alltoall_any 4725 16.4 0.058 0.116 0.753 0.928
jit_kernel_multiply 4 16.1 0.101 0.876 0.101 0.876
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.867 0.869
mp_alltoall_z22v 1201 16.6 0.764 0.864 0.764 0.864
acc_transpose_blocks_kernels 18288 16.5 0.216 0.222 0.795 0.814
mp_waitany 9880 13.7 0.529 0.786 0.529 0.786
rs_pw_transfer_RS2PW_140 130 11.5 0.121 0.126 0.525 0.779
cp_fm_cholesky_invert 11 10.9 0.775 0.779 0.775 0.779
acc_transpose_blocks_sync 54864 16.5 0.733 0.773 0.733 0.773
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.003 0.677 0.737
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=34.617000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.363636, yerr=1.610913
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/06/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 114.044384E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3805952 0.0% 0.0% 100.0%
average stack size 0.0 0.0 38.6
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 557.182976E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63489.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.025 0.048 39.278 39.279
qs_mol_dyn_low 1 2.0 0.003 0.004 39.085 39.092
qs_forces 11 3.9 0.002 0.003 37.079 37.081
qs_energies 11 4.9 0.002 0.002 35.319 35.329
scf_env_do_scf 11 5.9 0.000 0.001 30.175 30.177
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 26.743 26.743
velocity_verlet 10 3.0 0.001 0.002 20.779 20.862
dbcsr_multiply_generic 2286 12.5 0.098 0.101 19.968 20.112
qs_scf_new_mos 108 7.5 0.001 0.001 18.080 18.133
qs_scf_loop_do_ot 108 8.5 0.001 0.001 18.079 18.132
ot_scf_mini 108 9.5 0.002 0.002 17.048 17.096
multiply_cannon 2286 13.5 0.218 0.226 16.200 16.965
multiply_cannon_loop 2286 14.5 1.504 1.607 15.230 15.894
ot_mini 108 10.5 0.001 0.001 10.371 10.431
multiply_cannon_multrec 27432 15.5 2.445 3.133 9.012 9.332
qs_ot_get_derivative 108 11.5 0.001 0.001 8.499 8.548
dbcsr_mm_accdrv_process 47916 15.9 5.912 7.802 6.465 7.862
rebuild_ks_matrix 119 8.3 0.000 0.000 6.722 6.773
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.721 6.773
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.952 5.999
sum_up_and_integrate 119 10.3 0.035 0.037 3.874 3.885
qs_ot_get_p 119 10.4 0.001 0.001 3.817 3.879
integrate_v_rspace 119 11.3 0.003 0.003 3.839 3.850
init_scf_run 11 5.9 0.000 0.001 3.797 3.797
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.796 3.797
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.651 3.692
calculate_rho_elec 119 8.7 0.040 0.046 3.651 3.692
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.173 3.549
init_scf_loop 11 6.9 0.000 0.000 3.412 3.412
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.095 2.615
apply_single 119 13.6 0.000 0.000 2.095 2.615
acc_transpose_blocks 27432 15.5 0.118 0.122 2.463 2.596
prepare_preconditioner 11 7.9 0.000 0.000 2.546 2.555
make_preconditioner 11 8.9 0.000 0.000 2.546 2.555
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.152 2.479
make_m2s 4572 13.5 0.054 0.055 2.353 2.470
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.352 2.379
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.355 2.367
calculate_first_density_matrix 1 7.0 0.000 0.000 2.363 2.364
make_images 4572 14.5 0.271 0.332 2.245 2.362
mp_waitall_1 137007 16.6 1.858 2.355 1.858 2.355
rs_pw_transfer 974 11.9 0.009 0.010 2.083 2.277
density_rs2pw 119 9.7 0.004 0.004 2.006 2.208
calculate_dm_sparse 119 9.5 0.000 0.000 2.136 2.190
pw_transfer 1439 11.6 0.065 0.070 2.038 2.072
mp_sum_dm 438 4.9 1.928 2.018 1.928 2.018
md_write_output 11 3.9 0.077 1.840 0.085 1.991
update_particle_set 20 4.0 0.000 0.000 1.905 1.988
md_output 10 3.0 0.000 0.000 0.085 1.987
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.945 1.984
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.951 1.951
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.943 1.944
grid_integrate_task_list 119 12.3 1.823 1.918 1.823 1.918
ot_diis_step 108 11.5 0.012 0.012 1.831 1.832
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.773 1.787
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.683 1.694
cp_fm_diag_elpa_base 50 14.0 1.648 1.667 1.681 1.692
mp_sum_l 7287 12.8 0.996 1.665 0.996 1.665
fft_wrap_pw1pw2_140 487 13.2 0.292 0.306 1.591 1.633
potential_pw2rs 119 12.3 0.008 0.009 1.610 1.614
acc_transpose_blocks_sync 82296 16.5 1.406 1.514 1.406 1.514
fft3d_ps 1201 14.6 0.556 0.608 1.374 1.398
wfi_extrapolate 11 7.9 0.001 0.001 1.380 1.380
multiply_cannon_metrocomm3 27432 15.5 0.041 0.041 0.808 1.375
grid_collocate_task_list 119 9.7 1.249 1.357 1.249 1.357
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.285 1.304
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.232 1.245
cp_fm_upper_to_full 72 14.2 0.818 1.163 0.818 1.163
jit_kernel_multiply 6 15.8 0.488 1.103 0.488 1.103
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.080 1.083
dbcsr_complete_redistribute 329 12.2 0.120 0.145 0.784 1.053
make_images_data 4572 15.5 0.047 0.051 0.871 1.000
acc_transpose_blocks_kernels 27432 16.5 0.269 0.277 0.911 0.936
mp_alltoall_d11v 2130 13.8 0.784 0.928 0.784 0.928
hybrid_alltoall_any 4725 16.4 0.065 0.154 0.744 0.926
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.799 0.874
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.860 0.864
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.571 0.835
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=39.279000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=529.090909, yerr=2.842999
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/07/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 117.977176E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1384136 0.0% 0.0% 100.0%
average stack size 0.0 0.0 106.2
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 619.565056E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63488.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.043 30.184 30.185
qs_mol_dyn_low 1 2.0 0.003 0.003 29.953 29.962
qs_forces 11 3.9 0.002 0.002 29.884 29.884
qs_energies 11 4.9 0.002 0.002 28.137 28.139
scf_env_do_scf 11 5.9 0.000 0.001 23.227 23.227
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 20.716 20.717
velocity_verlet 10 3.0 0.014 0.017 15.268 15.271
dbcsr_multiply_generic 2286 12.5 0.091 0.096 13.574 13.694
qs_scf_new_mos 108 7.5 0.001 0.001 12.462 12.487
qs_scf_loop_do_ot 108 8.5 0.001 0.001 12.461 12.487
ot_scf_mini 108 9.5 0.002 0.002 11.740 11.765
multiply_cannon 2286 13.5 0.223 0.231 10.699 11.289
multiply_cannon_loop 2286 14.5 0.644 0.671 9.735 9.956
ot_mini 108 10.5 0.001 0.001 6.721 6.752
multiply_cannon_multrec 9144 15.5 1.679 1.927 5.950 6.237
rebuild_ks_matrix 119 8.3 0.000 0.000 6.152 6.178
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.152 6.177
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.474 5.497
qs_ot_get_derivative 108 11.5 0.001 0.001 5.416 5.438
dbcsr_mm_accdrv_process 12550 15.8 3.561 4.202 4.159 4.244
sum_up_and_integrate 119 10.3 0.037 0.041 3.782 3.787
integrate_v_rspace 119 11.3 0.003 0.003 3.744 3.750
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.671 3.678
calculate_rho_elec 119 8.7 0.060 0.061 3.670 3.678
init_scf_run 11 5.9 0.000 0.001 3.445 3.445
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.445 3.445
qs_ot_get_p 119 10.4 0.001 0.001 3.103 3.136
init_scf_loop 11 6.9 0.000 0.000 2.489 2.490
calculate_first_density_matrix 1 7.0 0.000 0.000 2.242 2.243
make_m2s 4572 13.5 0.034 0.035 2.001 2.160
pw_transfer 1439 11.6 0.066 0.068 2.142 2.151
mp_waitall_1 115863 16.7 1.658 2.119 1.658 2.119
density_rs2pw 119 9.7 0.004 0.004 1.960 2.101
make_images 4572 14.5 0.270 0.300 1.913 2.070
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.048 2.059
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.049 2.052
rs_pw_transfer 974 11.9 0.008 0.009 1.807 1.932
grid_integrate_task_list 119 12.3 1.863 1.929 1.863 1.929
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.797 1.797
calculate_dm_sparse 119 9.5 0.000 0.000 1.769 1.787
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.762 1.782
prepare_preconditioner 11 7.9 0.000 0.000 1.729 1.735
make_preconditioner 11 8.9 0.000 0.000 1.729 1.735
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.703 1.704
fft_wrap_pw1pw2_140 487 13.2 0.367 0.377 1.661 1.673
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.620 1.648
acc_transpose_blocks 9144 15.5 0.042 0.043 1.507 1.565
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.523 1.534
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.511 1.526
potential_pw2rs 119 12.3 0.010 0.011 1.485 1.488
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.480 1.488
cp_fm_diag_elpa_base 50 14.0 1.452 1.469 1.478 1.486
grid_collocate_task_list 119 9.7 1.299 1.418 1.299 1.418
fft3d_ps 1201 14.6 0.561 0.572 1.377 1.386
ot_diis_step 108 11.5 0.012 0.013 1.288 1.288
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.223 1.256
apply_single 119 13.6 0.000 0.000 1.223 1.256
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.229 1.230
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.146 1.151
wfi_extrapolate 11 7.9 0.001 0.001 1.148 1.148
hybrid_alltoall_any 4725 16.4 0.065 0.175 0.835 1.089
make_images_data 4572 15.5 0.042 0.046 0.847 1.051
jit_kernel_multiply 6 15.7 0.558 1.029 0.558 1.029
mp_alltoall_d11v 2130 13.8 0.835 0.961 0.835 0.961
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.871 0.923
cp_fm_cholesky_invert 11 10.9 0.862 0.865 0.862 0.865
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.838 0.845
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.802 0.804
acc_transpose_blocks_kernels 9144 16.5 0.116 0.119 0.732 0.781
acc_transpose_blocks_sync 27432 16.5 0.717 0.757 0.717 0.757
mp_allgather_i34 2286 14.5 0.280 0.738 0.280 0.738
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.677 0.724
multiply_cannon_metrocomm3 9144 15.5 0.020 0.021 0.421 0.724
mp_alltoall_z22v 1201 16.6 0.690 0.719 0.690 0.719
jit_kernel_transpose 5 15.6 0.616 0.666 0.616 0.666
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=30.185000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=587.727273, yerr=4.114076
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/08/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 235.585836E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1388964 0.0% 0.0% 100.0%
average stack size 0.0 0.0 105.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 771.006464E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63723.
MP_Allreduce 10154 429.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.054 44.765 44.765
qs_mol_dyn_low 1 2.0 0.003 0.003 43.414 44.556
qs_forces 11 3.9 0.002 0.002 42.676 42.687
qs_energies 11 4.9 0.002 0.003 40.658 40.669
scf_env_do_scf 11 5.9 0.001 0.001 35.025 35.025
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 27.168 27.170
velocity_verlet 10 3.0 0.002 0.002 24.649 24.718
dbcsr_multiply_generic 2286 12.5 0.098 0.101 18.961 19.189
qs_scf_new_mos 108 7.5 0.001 0.001 17.357 17.461
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.356 17.461
ot_scf_mini 108 9.5 0.002 0.002 16.204 16.306
multiply_cannon 2286 13.5 0.296 0.303 14.947 15.880
multiply_cannon_loop 2286 14.5 0.877 0.904 13.695 14.658
ot_mini 108 10.5 0.001 0.001 9.883 9.996
multiply_cannon_multrec 9144 15.5 3.391 4.898 8.764 8.849
qs_ot_get_derivative 108 11.5 0.001 0.001 7.844 7.945
init_scf_loop 11 6.9 0.000 0.000 7.830 7.833
rebuild_ks_matrix 119 8.3 0.000 0.000 7.337 7.475
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.337 7.475
prepare_preconditioner 11 7.9 0.000 0.000 6.855 6.868
make_preconditioner 11 8.9 0.000 0.000 6.855 6.868
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.368 6.748
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.605 6.730
dbcsr_mm_accdrv_process 12550 15.8 4.482 5.913 5.231 6.523
cp_fm_upper_to_full 72 14.2 3.221 4.654 3.221 4.654
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.290 4.295
calculate_rho_elec 119 8.7 0.118 0.121 4.289 4.294
sum_up_and_integrate 119 10.3 0.064 0.065 4.037 4.042
integrate_v_rspace 119 11.3 0.004 0.004 3.972 3.979
init_scf_run 11 5.9 0.000 0.001 3.575 3.575
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.575 3.575
qs_ot_get_p 119 10.4 0.001 0.001 3.413 3.549
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.768 3.254
mp_waitall_1 94719 16.7 2.174 3.197 2.174 3.197
dbcsr_complete_redistribute 329 12.2 0.289 0.295 2.027 2.908
pw_transfer 1439 11.6 0.068 0.069 2.669 2.673
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.720 2.579
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.570 2.574
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.273 2.531
apply_single 119 13.6 0.000 0.000 2.272 2.531
make_m2s 4572 13.5 0.037 0.038 2.343 2.485
make_images 4572 14.5 0.354 0.388 2.223 2.365
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.482 2.333
mp_alltoall_i22 627 13.8 1.460 2.307 1.460 2.307
calculate_dm_sparse 119 9.5 0.000 0.000 2.226 2.237
density_rs2pw 119 9.7 0.004 0.004 2.176 2.187
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.246 2.186
fft_wrap_pw1pw2_140 487 13.2 0.620 0.624 2.171 2.175
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.079 2.130
grid_integrate_task_list 119 12.3 2.072 2.100 2.072 2.100
calculate_first_density_matrix 1 7.0 0.000 0.000 2.091 2.093
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.037 2.040
ot_diis_step 108 11.5 0.014 0.014 2.016 2.016
acc_transpose_blocks 9144 15.5 0.044 0.045 1.842 1.896
md_output 10 3.0 0.000 0.000 0.233 1.846
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.827 1.827
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.775 1.787
mp_sum_l 7287 12.8 1.128 1.720 1.128 1.720
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.678 1.679
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.572 1.611
fft3d_ps 1201 14.6 0.593 0.603 1.582 1.585
grid_collocate_task_list 119 9.7 1.517 1.538 1.517 1.538
rs_pw_transfer 974 11.9 0.009 0.009 1.479 1.503
wfi_extrapolate 11 7.9 0.001 0.001 1.417 1.417
cp_fm_cholesky_invert 11 10.9 1.398 1.401 1.398 1.401
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.392 1.393
potential_pw2rs 119 12.3 0.014 0.014 1.388 1.391
cp_fm_diag_elpa_base 50 14.0 1.244 1.299 1.390 1.391
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.303 1.321
mp_max_l 33 2.8 1.142 1.305 1.142 1.305
write_restart 10 4.0 0.162 1.292 0.163 1.293
hybrid_alltoall_any 4725 16.4 0.089 0.149 1.048 1.255
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.236 1.242
acc_transpose_blocks_sync 27432 16.5 1.121 1.168 1.121 1.168
make_images_data 4572 15.5 0.045 0.048 0.969 1.151
mp_alltoall_d11v 2130 13.8 1.112 1.147 1.112 1.147
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.101 1.121
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.993 1.028
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.943 0.957
qs_create_task_list 11 7.9 0.000 0.000 0.930 0.941
generate_qs_task_list 11 8.9 0.367 0.386 0.929 0.941
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=44.765000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=725.000000, yerr=14.142136
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/09/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 198.287135E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 8410880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 117.0
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 500.203520E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66212.
MP_Allreduce 9776 488.
MP_Sync 52
MP_Alltoall 1938 2296993.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.029 83.436 83.437
qs_mol_dyn_low 1 2.0 0.026 0.032 83.183 83.194
qs_forces 11 3.9 0.009 0.022 83.084 83.097
qs_energies 11 4.9 0.002 0.002 80.167 80.180
scf_env_do_scf 11 5.9 0.000 0.001 71.284 71.287
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 65.858 65.858
dbcsr_multiply_generic 2055 12.4 0.105 0.110 52.376 52.691
qs_scf_new_mos 99 7.5 0.000 0.001 48.394 48.553
qs_scf_loop_do_ot 99 8.5 0.001 0.001 48.393 48.553
ot_scf_mini 99 9.5 0.002 0.002 46.125 46.249
multiply_cannon 2055 13.4 0.186 0.190 42.771 45.326
multiply_cannon_loop 2055 14.4 1.782 1.839 41.819 44.339
velocity_verlet 10 3.0 0.001 0.002 42.558 42.559
ot_mini 99 10.5 0.001 0.001 28.216 28.312
qs_ot_get_derivative 99 11.5 0.001 0.001 21.361 21.492
multiply_cannon_multrec 49320 15.4 11.630 12.202 17.673 19.162
rebuild_ks_matrix 110 8.3 0.000 0.000 14.698 14.862
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.697 14.861
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.868 13.010
mp_waitall_1 220248 16.4 10.504 12.978 10.504 12.978
multiply_cannon_sync_h2d 49320 15.4 9.850 10.484 9.850 10.484
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.815 10.176
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 9.080 10.160
apply_single 110 13.6 0.000 0.000 9.080 10.160
qs_ot_get_p 110 10.4 0.001 0.001 9.555 9.697
multiply_cannon_metrocomm3 49320 15.4 0.085 0.090 6.464 8.749
dbcsr_mm_accdrv_process 87628 16.1 2.929 3.085 5.910 7.496
sum_up_and_integrate 110 10.3 0.036 0.044 7.118 7.133
integrate_v_rspace 110 11.3 0.003 0.004 7.081 7.106
init_scf_run 11 5.9 0.000 0.001 6.807 6.807
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.807 6.807
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.661 6.802
calculate_rho_elec 110 8.6 0.021 0.026 6.661 6.802
mp_sum_l 6594 12.7 5.239 6.792 5.239 6.792
ot_diis_step 99 11.5 0.005 0.006 6.640 6.640
qs_ot_p2m_diag 48 11.0 0.012 0.019 6.457 6.472
cp_dbcsr_syevd 48 12.0 0.003 0.003 5.697 5.697
init_scf_loop 11 6.9 0.000 0.000 5.398 5.399
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.200 5.251
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.127 5.163
cp_fm_diag_elpa_base 48 14.0 5.114 5.149 5.125 5.161
rs_pw_transfer 902 11.9 0.012 0.013 3.709 4.420
jit_kernel_multiply 13 15.9 2.707 4.277 2.707 4.277
density_rs2pw 110 9.6 0.004 0.005 3.440 4.154
wfi_extrapolate 11 7.9 0.001 0.001 4.073 4.073
make_m2s 4110 13.4 0.060 0.065 3.910 4.033
make_images 4110 14.4 0.177 0.189 3.814 3.943
multiply_cannon_metrocomm1 49320 15.4 0.066 0.069 2.199 3.894
calculate_dm_sparse 110 9.5 0.001 0.001 3.649 3.767
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.591 3.594
grid_integrate_task_list 110 12.3 3.240 3.415 3.240 3.415
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.231 3.293
prepare_preconditioner 11 7.9 0.000 0.000 3.231 3.261
make_preconditioner 11 8.9 0.000 0.000 3.231 3.261
pw_transfer 1331 11.6 0.055 0.062 3.149 3.228
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.060 3.141
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.013 3.068
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.996 3.047
fft_wrap_pw1pw2_140 451 13.1 0.457 0.502 2.604 2.697
acc_transpose_blocks 49320 15.4 0.209 0.224 2.534 2.650
mp_alltoall_d11v 2046 13.8 2.052 2.646 2.052 2.646
potential_pw2rs 110 12.3 0.006 0.007 2.620 2.641
calculate_first_density_matrix 1 7.0 0.000 0.000 2.629 2.632
mp_waitany 14300 13.8 1.824 2.568 1.824 2.568
grid_collocate_task_list 110 9.6 2.159 2.360 2.159 2.360
fft3d_ps 1111 14.6 0.782 0.870 2.248 2.308
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.959 1.989
make_images_data 4110 15.4 0.044 0.047 1.796 1.928
hybrid_alltoall_any 4261 16.3 0.086 0.488 1.564 1.819
mp_sum_d 3889 11.9 1.320 1.805 1.320 1.805
cp_fm_cholesky_invert 11 10.9 1.734 1.738 1.734 1.738
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.657 1.693
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=83.437000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=475.636364, yerr=2.772355
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/10/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 390.715586E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 5019072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 196.1
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 590.073856E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66430.
MP_Allreduce 9775 566.
MP_Sync 52
MP_Alltoall 1717 2210383.
MP_SendRecv 10340 26400.
MP_ISendRecv 10340 26400.
MP_Wait 22352
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.028 70.781 70.781
qs_mol_dyn_low 1 2.0 0.004 0.004 70.547 70.556
qs_forces 11 3.9 0.003 0.003 70.468 70.479
qs_energies 11 4.9 0.002 0.002 67.100 67.114
scf_env_do_scf 11 5.9 0.001 0.001 57.771 57.774
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 50.353 50.354
dbcsr_multiply_generic 2055 12.4 0.115 0.119 39.730 39.858
velocity_verlet 10 3.0 0.001 0.002 36.058 36.059
multiply_cannon 2055 13.4 0.224 0.242 32.990 34.247
qs_scf_new_mos 99 7.5 0.001 0.001 34.041 34.157
qs_scf_loop_do_ot 99 8.5 0.001 0.001 34.040 34.157
ot_scf_mini 99 9.5 0.003 0.003 32.376 32.480
multiply_cannon_loop 2055 14.4 1.163 1.191 31.730 32.368
ot_mini 99 10.5 0.001 0.001 19.378 19.496
multiply_cannon_multrec 24660 15.4 7.031 8.636 14.108 16.354
rebuild_ks_matrix 110 8.3 0.000 0.001 13.790 13.903
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 13.790 13.902
qs_ot_get_derivative 99 11.5 0.001 0.001 13.609 13.717
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.115 12.209
mp_waitall_1 176588 16.5 8.962 11.739 8.962 11.739
multiply_cannon_metrocomm3 24660 15.4 0.072 0.076 6.488 9.751
dbcsr_mm_accdrv_process 52282 16.1 5.466 6.354 6.906 7.897
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.651 7.606
apply_single 110 13.6 0.000 0.001 6.651 7.606
init_scf_loop 11 6.9 0.000 0.000 7.383 7.384
multiply_cannon_sync_h2d 24660 15.4 6.328 7.356 6.328 7.356
init_scf_run 11 5.9 0.000 0.001 6.949 6.950
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.949 6.949
sum_up_and_integrate 110 10.3 0.053 0.060 6.606 6.622
integrate_v_rspace 110 11.3 0.003 0.003 6.553 6.570
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 5.617 6.538
qs_ot_get_p 110 10.4 0.001 0.001 6.367 6.529
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.181 6.336
calculate_rho_elec 110 8.6 0.040 0.048 6.181 6.336
ot_diis_step 99 11.5 0.010 0.010 5.722 5.722
prepare_preconditioner 11 7.9 0.000 0.000 5.362 5.379
make_preconditioner 11 8.9 0.000 0.000 5.362 5.379
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.947 5.100
make_m2s 4110 13.4 0.057 0.059 4.188 4.660
make_images 4110 14.4 0.407 0.465 4.081 4.552
qs_ot_p2m_diag 48 11.0 0.028 0.044 4.490 4.510
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.028 4.029
density_rs2pw 110 9.6 0.004 0.005 3.251 3.876
calculate_dm_sparse 110 9.5 0.001 0.001 3.801 3.835
pw_transfer 1331 11.6 0.066 0.072 3.608 3.739
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 3.501 3.634
rs_pw_transfer 902 11.9 0.013 0.015 2.918 3.547
wfi_extrapolate 11 7.9 0.001 0.001 3.509 3.509
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.472 3.483
cp_fm_diag_elpa_base 48 14.0 3.424 3.438 3.469 3.480
grid_integrate_task_list 110 12.3 3.169 3.357 3.169 3.357
calculate_first_density_matrix 1 7.0 0.000 0.001 3.337 3.343
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.255 3.314
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.240 3.242
fft_wrap_pw1pw2_140 451 13.1 0.527 0.542 3.015 3.154
mp_sum_l 6594 12.7 1.919 3.138 1.919 3.138
acc_transpose_blocks 24660 15.4 0.117 0.120 2.137 3.130
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.906 2.948
make_images_data 4110 15.4 0.049 0.054 2.287 2.753
hybrid_alltoall_any 4261 16.3 0.107 0.456 1.990 2.748
fft3d_ps 1111 14.6 1.107 1.333 2.465 2.610
grid_collocate_task_list 110 9.6 2.103 2.541 2.103 2.541
cp_fm_cholesky_invert 11 10.9 2.480 2.487 2.480 2.487
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 2.371 2.385
potential_pw2rs 110 12.3 0.008 0.008 2.371 2.378
jit_kernel_multiply 9 16.4 1.085 2.144 1.085 2.144
acc_transpose_blocks_kernels 24660 16.4 0.317 0.328 1.059 2.036
mp_alltoall_d11v 2046 13.8 1.712 2.018 1.712 2.018
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.951 1.971
mp_waitany 10164 13.8 1.235 1.897 1.235 1.897
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.876 1.888
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.831 1.846
jit_kernel_transpose 5 15.6 0.741 1.717 0.741 1.717
multiply_cannon_metrocomm4 22605 15.4 0.078 0.082 0.783 1.629
rs_pw_transfer_RS2PW_140 121 11.5 0.208 0.219 0.983 1.608
mp_allgather_i34 2055 14.4 0.568 1.608 0.568 1.608
mp_irecv_dv 57340 16.2 0.654 1.505 0.654 1.505
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.386 1.492
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=70.781000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=558.454545, yerr=5.726551
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/11/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 404.681598E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 3346752 0.0% 0.0% 100.0%
average stack size 0.0 0.0 294.1
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 668.860416E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66421.
MP_Allreduce 9774 562.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.031 63.188 63.188
qs_mol_dyn_low 1 2.0 0.003 0.004 62.803 62.933
qs_forces 11 3.9 0.003 0.003 61.282 61.292
qs_energies 11 4.9 0.002 0.002 58.074 58.088
scf_env_do_scf 11 5.9 0.000 0.001 50.263 50.263
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 41.866 41.867
velocity_verlet 10 3.0 0.018 0.025 33.954 34.000
dbcsr_multiply_generic 2055 12.4 0.105 0.110 30.354 30.632
qs_scf_new_mos 99 7.5 0.001 0.001 26.739 26.821
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.739 26.820
ot_scf_mini 99 9.5 0.002 0.002 25.498 25.605
multiply_cannon 2055 13.4 0.213 0.224 23.671 24.952
multiply_cannon_loop 2055 14.4 0.816 0.870 22.516 24.151
ot_mini 99 10.5 0.001 0.001 15.384 15.490
rebuild_ks_matrix 110 8.3 0.000 0.000 12.408 12.541
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.408 12.541
mp_waitall_1 139946 16.5 7.784 11.783 7.784 11.783
qs_ot_get_derivative 99 11.5 0.001 0.001 10.966 11.068
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.902 11.018
multiply_cannon_multrec 16440 15.4 3.705 4.755 10.057 10.961
multiply_cannon_metrocomm3 16440 15.4 0.045 0.049 5.049 9.125
init_scf_loop 11 6.9 0.000 0.000 8.364 8.365
dbcsr_mm_accdrv_process 34862 16.1 5.476 7.396 6.198 7.913
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.411 6.622
apply_single 110 13.6 0.000 0.000 5.411 6.622
prepare_preconditioner 11 7.9 0.000 0.000 6.600 6.618
make_preconditioner 11 8.9 0.000 0.000 6.600 6.618
sum_up_and_integrate 110 10.3 0.061 0.062 6.544 6.560
integrate_v_rspace 110 11.3 0.003 0.003 6.483 6.499
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.905 6.274
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.052 6.061
calculate_rho_elec 110 8.6 0.059 0.059 6.051 6.060
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.403 5.458
init_scf_run 11 5.9 0.000 0.001 5.429 5.429
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.428 5.428
qs_ot_get_p 110 10.4 0.001 0.001 5.283 5.411
make_m2s 4110 13.4 0.049 0.051 4.113 4.494
ot_diis_step 99 11.5 0.011 0.011 4.381 4.381
make_images 4110 14.4 0.397 0.519 3.999 4.379
density_rs2pw 110 9.6 0.004 0.005 3.099 4.340
rs_pw_transfer 902 11.9 0.010 0.011 2.638 3.852
multiply_cannon_sync_h2d 16440 15.4 3.260 3.722 3.260 3.722
qs_ot_p2m_diag 48 11.0 0.041 0.044 3.679 3.684
pw_transfer 1331 11.6 0.066 0.073 3.643 3.653
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.536 3.548
grid_integrate_task_list 110 12.3 3.184 3.381 3.184 3.381
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.312 3.312
fft_wrap_pw1pw2_140 451 13.1 0.645 0.654 3.069 3.081
wfi_extrapolate 11 7.9 0.001 0.001 2.959 2.959
mp_sum_l 6594 12.7 1.948 2.818 1.948 2.818
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.786 2.793
cp_fm_diag_elpa_base 48 14.0 2.718 2.748 2.784 2.792
make_images_data 4110 15.4 0.046 0.051 2.313 2.787
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.771 2.771
hybrid_alltoall_any 4261 16.3 0.110 0.389 2.066 2.732
cp_fm_cholesky_invert 11 10.9 2.554 2.560 2.554 2.560
calculate_dm_sparse 110 9.5 0.001 0.001 2.510 2.543
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.455 2.502
grid_collocate_task_list 110 9.6 2.122 2.497 2.122 2.497
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.412 2.471
mp_waitany 17072 13.8 1.185 2.432 1.185 2.432
multiply_cannon_metrocomm4 14385 15.4 0.047 0.051 0.875 2.405
calculate_first_density_matrix 1 7.0 0.000 0.000 2.377 2.378
fft3d_ps 1111 14.6 1.088 1.096 2.347 2.358
mp_irecv_dv 48980 15.7 0.802 2.273 0.802 2.273
mp_alltoall_d11v 2046 13.8 1.788 2.254 1.788 2.254
potential_pw2rs 110 12.3 0.011 0.011 2.247 2.253
rs_pw_transfer_RS2PW_140 121 11.5 0.177 0.185 0.910 2.122
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.953 1.963
dbcsr_complete_redistribute 325 12.2 0.328 0.354 1.455 1.893
cp_fm_upper_to_full 70 14.2 1.371 1.765 1.371 1.765
acc_transpose_blocks 16440 15.4 0.080 0.086 1.611 1.736
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.699 1.717
multiply_cannon_metrocomm1 16440 15.4 0.029 0.030 0.551 1.715
md_write_output 11 3.9 0.045 1.413 0.052 1.619
mp_sum_dm 438 4.9 1.465 1.543 1.465 1.543
cp_fm_cholesky_decompose 22 10.9 1.513 1.532 1.513 1.532
mp_allgather_i34 2055 14.4 0.465 1.503 0.465 1.503
update_particle_set 20 4.0 0.000 0.000 1.427 1.490
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.454 1.467
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.368 1.458
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.990 1.432
rs_gather_matrices 110 12.3 0.234 0.261 0.974 1.377
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.274 1.285
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=63.188000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=631.181818, yerr=8.642581
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/12/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 601.317074E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 4916280 0.0% 0.0% 100.0%
average stack size 0.0 0.0 200.2
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 736.448512E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66419.
MP_Allreduce 9774 603.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.030 69.682 69.683
qs_mol_dyn_low 1 2.0 0.010 0.012 69.314 69.325
qs_forces 11 3.9 0.005 0.010 68.493 68.512
qs_energies 11 4.9 0.003 0.006 64.995 65.018
scf_env_do_scf 11 5.9 0.001 0.002 56.632 56.635
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 45.088 45.089
velocity_verlet 10 3.0 0.001 0.002 38.128 38.161
dbcsr_multiply_generic 2055 12.4 0.113 0.119 33.169 33.368
qs_scf_new_mos 99 7.5 0.001 0.001 29.935 30.062
qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.935 30.061
ot_scf_mini 99 9.5 0.003 0.003 28.256 28.382
multiply_cannon 2055 13.4 0.240 0.261 25.558 27.017
multiply_cannon_loop 2055 14.4 1.408 1.501 24.012 25.137
ot_mini 99 10.5 0.001 0.001 17.259 17.407
multiply_cannon_multrec 24660 15.4 4.085 6.665 13.137 15.074
qs_ot_get_derivative 99 11.5 0.001 0.001 13.077 13.206
rebuild_ks_matrix 110 8.3 0.000 0.000 12.343 12.444
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.018 12.343 12.443
init_scf_loop 11 6.9 0.003 0.011 11.504 11.506
dbcsr_mm_accdrv_process 52304 16.0 7.878 11.023 8.894 11.321
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.840 10.930
prepare_preconditioner 11 7.9 0.000 0.001 9.714 9.733
make_preconditioner 11 8.9 0.000 0.002 9.714 9.733
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.945 9.408
mp_waitall_1 121746 16.5 5.637 8.332 5.637 8.332
sum_up_and_integrate 110 10.3 0.067 0.071 6.518 6.531
integrate_v_rspace 110 11.3 0.003 0.003 6.451 6.464
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.166 6.178
calculate_rho_elec 110 8.6 0.078 0.082 6.166 6.178
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 5.417 6.044
qs_ot_get_p 110 10.4 0.001 0.001 5.853 6.022
make_m2s 4110 13.4 0.059 0.061 5.369 5.857
make_images 4110 14.4 0.580 0.701 5.229 5.712
multiply_cannon_metrocomm3 24660 15.4 0.038 0.039 2.632 5.643
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.625 5.627
apply_single 110 13.6 0.000 0.000 4.625 5.626
init_scf_run 11 5.9 0.000 0.001 5.591 5.591
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.590 5.591
cp_fm_upper_to_full 70 14.2 3.313 4.822 3.313 4.822
ot_diis_step 99 11.5 0.011 0.012 4.145 4.145
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.983 3.999
dbcsr_complete_redistribute 325 12.2 0.421 0.472 2.658 3.806
pw_transfer 1331 11.6 0.065 0.075 3.761 3.791
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.655 3.689
density_rs2pw 110 9.6 0.004 0.004 3.037 3.557
grid_integrate_task_list 110 12.3 3.268 3.435 3.268 3.435
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.363 3.434
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.413 3.414
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.123 3.255
acc_transpose_blocks 24660 15.4 0.112 0.114 2.649 3.245
fft_wrap_pw1pw2_140 451 13.1 0.674 0.696 3.135 3.173
multiply_cannon_metrocomm1 24660 15.4 0.035 0.036 0.531 3.121
hybrid_alltoall_any 4261 16.3 0.123 0.459 2.260 3.098
make_images_data 4110 15.4 0.048 0.052 2.627 3.090
calculate_dm_sparse 110 9.5 0.001 0.001 3.000 3.038
wfi_extrapolate 11 7.9 0.001 0.001 3.002 3.002
rs_pw_transfer 902 11.9 0.016 0.054 2.418 2.919
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.877 2.887
cp_fm_diag_elpa_base 48 14.0 2.728 2.782 2.874 2.884
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.760 2.877
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.850 2.852
mp_sum_l 6594 12.7 1.566 2.847 1.566 2.847
mp_alltoall_i22 605 13.7 1.672 2.837 1.672 2.837
cp_fm_cholesky_invert 11 10.9 2.605 2.613 2.605 2.613
multiply_cannon_sync_h2d 24660 15.4 2.370 2.602 2.370 2.602
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.483 2.523
calculate_first_density_matrix 1 7.0 0.001 0.005 2.490 2.493
grid_collocate_task_list 110 9.6 2.225 2.492 2.225 2.492
fft3d_ps 1111 14.6 1.082 1.109 2.407 2.428
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 2.273 2.294
potential_pw2rs 110 12.3 0.012 0.013 2.155 2.163
mp_alltoall_d11v 2046 13.8 1.743 1.999 1.743 1.999
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.784 1.823
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.746 1.758
acc_transpose_blocks_kernels 24660 16.4 0.306 0.312 1.099 1.735
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.612 1.707
cp_fm_cholesky_decompose 22 10.9 1.608 1.654 1.608 1.654
acc_transpose_blocks_sync 73980 16.4 1.396 1.594 1.396 1.594
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.565 1.588
mp_allgather_i34 2055 14.4 0.629 1.559 0.629 1.559
mp_waitany 13376 13.8 1.091 1.542 1.091 1.542
multiply_cannon_metrocomm4 20550 15.4 0.060 0.065 0.863 1.458
jit_kernel_transpose 5 15.6 0.793 1.432 0.793 1.432
jit_kernel_multiply 8 15.7 0.675 1.432 0.675 1.432
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=69.683000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=696.818182, yerr=11.914988
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/13/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 807.299199E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 1438408 0.0% 0.0% 100.0%
average stack size 0.0 0.0 684.2
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 851.279872E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66417.
MP_Allreduce 9774 644.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.031 56.595 56.596
qs_mol_dyn_low 1 2.0 0.011 0.013 56.324 56.334
qs_forces 11 3.9 0.022 0.024 55.916 55.919
qs_energies 11 4.9 0.001 0.002 52.198 52.204
scf_env_do_scf 11 5.9 0.000 0.001 43.973 43.973
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 36.376 36.377
velocity_verlet 10 3.0 0.002 0.002 31.624 31.647
dbcsr_multiply_generic 2055 12.4 0.104 0.109 23.893 24.025
qs_scf_new_mos 99 7.5 0.001 0.001 21.301 21.357
qs_scf_loop_do_ot 99 8.5 0.001 0.001 21.300 21.357
ot_scf_mini 99 9.5 0.002 0.002 20.055 20.082
multiply_cannon 2055 13.4 0.235 0.249 18.348 19.778
multiply_cannon_loop 2055 14.4 0.609 0.626 17.076 17.417
rebuild_ks_matrix 110 8.3 0.000 0.000 11.919 11.961
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 11.919 11.961
ot_mini 99 10.5 0.001 0.001 11.390 11.409
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.571 10.608
multiply_cannon_multrec 8220 15.4 3.698 5.141 8.226 9.082
qs_ot_get_derivative 99 11.5 0.001 0.001 7.704 7.730
mp_waitall_1 103326 16.6 5.882 7.646 5.882 7.646
init_scf_loop 11 6.9 0.000 0.000 7.548 7.550
sum_up_and_integrate 110 10.3 0.080 0.081 6.521 6.533
integrate_v_rspace 110 11.3 0.003 0.003 6.441 6.453
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.326 6.338
calculate_rho_elec 110 8.6 0.115 0.116 6.325 6.338
prepare_preconditioner 11 7.9 0.000 0.000 5.891 5.898
make_preconditioner 11 8.9 0.000 0.000 5.891 5.898
dbcsr_mm_accdrv_process 17442 15.9 3.127 4.705 4.389 5.772
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.474 5.558
init_scf_run 11 5.9 0.000 0.001 5.103 5.103
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.102 5.103
qs_ot_get_p 110 10.4 0.001 0.001 4.709 4.737
make_m2s 4110 13.4 0.038 0.040 4.230 4.514
make_images 4110 14.4 0.650 0.709 4.101 4.387
multiply_cannon_metrocomm3 8220 15.4 0.019 0.019 3.008 4.224
pw_transfer 1331 11.6 0.066 0.071 3.911 3.919
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.803 3.816
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.722 3.771
apply_single 110 13.6 0.000 0.000 3.721 3.771
ot_diis_step 99 11.5 0.012 0.012 3.664 3.664
grid_integrate_task_list 110 12.3 3.363 3.541 3.363 3.541
density_rs2pw 110 9.6 0.004 0.004 3.025 3.400
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.358 3.362
fft_wrap_pw1pw2_140 451 13.1 0.839 0.852 3.286 3.300
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.021 3.021
cp_fm_cholesky_invert 11 10.9 2.819 2.823 2.819 2.823
hybrid_alltoall_any 4261 16.3 0.202 0.861 2.221 2.752
wfi_extrapolate 11 7.9 0.001 0.001 2.741 2.741
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 2.667 2.671
make_images_data 4110 15.4 0.041 0.047 2.263 2.657
grid_collocate_task_list 110 9.6 2.316 2.576 2.316 2.576
rs_pw_transfer 902 11.9 0.010 0.010 2.172 2.571
calculate_dm_sparse 110 9.5 0.001 0.001 2.475 2.512
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.503 2.511
cp_fm_diag_elpa_base 48 14.0 2.447 2.473 2.502 2.510
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.506 2.507
multiply_cannon_sync_h2d 8220 15.4 2.374 2.471 2.374 2.471
fft3d_ps 1111 14.6 1.133 1.148 2.339 2.350
calculate_first_density_matrix 1 7.0 0.000 0.000 2.261 2.263
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.150 2.163
potential_pw2rs 110 12.3 0.015 0.015 2.052 2.057
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.004 2.032
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.783 1.986
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.907 1.923
mp_alltoall_d11v 2046 13.8 1.631 1.901 1.631 1.901
jit_kernel_multiply 7 15.6 0.955 1.781 0.955 1.781
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.678 1.683
cp_fm_cholesky_decompose 22 10.9 1.638 1.655 1.638 1.655
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.490 1.611
mp_allgather_i34 2055 14.4 0.443 1.537 0.443 1.537
dbcsr_complete_redistribute 325 12.2 0.563 0.587 1.421 1.519
mp_waitany 9240 13.8 1.062 1.484 1.062 1.484
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.434 1.446
acc_transpose_blocks 8220 15.4 0.039 0.041 1.375 1.395
qs_create_task_list 11 7.9 0.000 0.001 1.213 1.318
generate_qs_task_list 11 8.9 0.374 0.442 1.212 1.318
rs_pw_transfer_RS2PW_140 121 11.5 0.162 0.167 0.801 1.205
rs_gather_matrices 110 12.3 0.327 0.367 0.936 1.176
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.143 1.164
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=56.596000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=806.909091, yerr=9.839203
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/14/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 1.612391E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 1464624 0.0% 0.0% 100.0%
average stack size 0.0 0.0 672.0
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 1.454686E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67098.
MP_Allreduce 9752 812.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.037 90.832 90.833
qs_mol_dyn_low 1 2.0 0.003 0.003 90.490 90.499
qs_forces 11 3.9 0.003 0.003 90.403 90.404
qs_energies 11 4.9 0.002 0.002 86.199 86.202
scf_env_do_scf 11 5.9 0.001 0.001 76.031 76.031
velocity_verlet 10 3.0 0.002 0.002 57.528 57.536
scf_env_do_scf_inner_loop 99 6.5 0.003 0.006 47.003 47.004
dbcsr_multiply_generic 2055 12.4 0.119 0.124 30.480 30.572
init_scf_loop 11 6.9 0.000 0.000 28.950 28.952
qs_scf_new_mos 99 7.5 0.001 0.001 28.491 28.540
qs_scf_loop_do_ot 99 8.5 0.001 0.001 28.491 28.540
prepare_preconditioner 11 7.9 0.000 0.000 26.863 26.871
make_preconditioner 11 8.9 0.000 0.000 26.863 26.871
ot_scf_mini 99 9.5 0.002 0.002 26.663 26.700
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.947 26.335
multiply_cannon 2055 13.4 0.332 0.368 23.375 23.940
multiply_cannon_loop 2055 14.4 0.817 0.834 21.543 22.019
cp_fm_upper_to_full 70 14.2 12.698 18.334 12.698 18.334
ot_mini 99 10.5 0.001 0.001 15.002 15.039
rebuild_ks_matrix 110 8.3 0.000 0.000 14.324 14.377
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 14.323 14.376
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.948 12.995
dbcsr_complete_redistribute 325 12.2 1.016 1.035 7.560 10.897
qs_ot_get_derivative 99 11.5 0.001 0.001 10.317 10.356
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.489 9.826
multiply_cannon_multrec 8220 15.4 4.049 4.215 9.717 9.788
mp_waitall_1 84994 16.7 8.321 9.634 8.321 9.634
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.901 9.200
mp_alltoall_i22 605 13.7 5.526 8.882 5.526 8.882
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.861 7.902
calculate_rho_elec 110 8.6 0.226 0.227 7.860 7.902
sum_up_and_integrate 110 10.3 0.150 0.151 7.399 7.414
integrate_v_rspace 110 11.3 0.004 0.004 7.248 7.263
qs_ot_get_p 110 10.4 0.001 0.001 6.273 6.323
multiply_cannon_metrocomm3 8220 15.4 0.019 0.020 5.576 6.205
make_m2s 4110 13.4 0.043 0.044 5.474 5.975
init_scf_run 11 5.9 0.000 0.001 5.866 5.866
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.866 5.866
make_images 4110 14.4 0.886 0.941 5.286 5.785
dbcsr_mm_accdrv_process 11614 15.7 3.837 4.082 5.519 5.695
cp_fm_cholesky_invert 11 10.9 5.550 5.554 5.550 5.554
pw_transfer 1331 11.6 0.075 0.075 5.153 5.156
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 5.036 5.039
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.589 4.824
apply_single 110 13.6 0.000 0.000 4.589 4.824
ot_diis_step 99 11.5 0.015 0.015 4.658 4.658
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.608 4.617
fft_wrap_pw1pw2_140 451 13.1 1.340 1.342 4.366 4.373
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.081 4.082
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.828 4.063
density_rs2pw 110 9.6 0.004 0.004 3.841 3.879
grid_integrate_task_list 110 12.3 3.687 3.770 3.687 3.770
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.685 3.686
hybrid_alltoall_any 4261 16.3 0.263 0.565 2.905 3.537
make_images_data 4110 15.4 0.045 0.047 2.834 3.488
wfi_extrapolate 11 7.9 0.001 0.001 3.471 3.471
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.452 3.452
cp_fm_diag_elpa_base 48 14.0 2.911 3.117 3.450 3.450
calculate_dm_sparse 110 9.5 0.001 0.001 3.215 3.244
multiply_cannon_sync_h2d 8220 15.4 3.140 3.155 3.140 3.155
fft3d_ps 1111 14.6 1.299 1.306 2.960 2.966
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.923 2.924
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.748 2.769
grid_collocate_task_list 110 9.6 2.676 2.696 2.676 2.696
acc_transpose_blocks 8220 15.4 0.041 0.041 1.934 2.600
potential_pw2rs 110 12.3 0.021 0.021 2.468 2.472
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.400 2.417
rs_pw_transfer 902 11.9 0.011 0.011 2.359 2.390
calculate_first_density_matrix 1 7.0 0.000 0.000 2.269 2.269
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.196 2.261
cp_fm_cholesky_decompose 22 10.9 2.158 2.185 2.158 2.185
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.097 2.167
mp_alltoall_d11v 2046 13.8 2.054 2.162 2.054 2.162
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.976 1.982
qs_create_task_list 11 7.9 0.016 0.020 1.901 1.948
generate_qs_task_list 11 8.9 0.730 0.782 1.885 1.931
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.841 1.879
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=90.833000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1300.909091, yerr=72.754200
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 631.074816E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175954870160
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 57905.
MP_Allreduce 11059 797.
MP_Sync 87
MP_Alltoall 2226 2090273.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.033 214.705 214.706
qs_mol_dyn_low 1 2.0 0.004 0.004 214.284 214.295
qs_forces 11 3.9 0.005 0.005 213.946 213.946
qs_energies 11 4.9 0.002 0.003 208.247 208.259
scf_env_do_scf 11 5.9 0.003 0.011 189.134 189.138
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 168.218 168.220
dbcsr_multiply_generic 2507 12.6 0.177 0.179 128.588 129.265
qs_scf_new_mos 117 7.6 0.001 0.001 127.291 127.546
qs_scf_loop_do_ot 117 8.6 0.001 0.001 127.290 127.545
velocity_verlet 10 3.0 0.001 0.002 126.657 126.660
ot_scf_mini 117 9.6 0.003 0.003 120.522 120.760
multiply_cannon 2507 13.6 0.240 0.250 104.172 106.059
multiply_cannon_loop 2507 14.6 2.423 2.469 101.862 103.467
ot_mini 117 10.6 0.001 0.001 66.699 66.933
multiply_cannon_multrec 60168 15.6 31.963 33.659 42.333 44.892
qs_ot_get_derivative 117 11.6 0.001 0.001 41.799 42.016
rebuild_ks_matrix 128 8.3 0.001 0.001 34.118 34.447
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.016 34.117 34.446
mp_waitall_1 267128 16.5 30.209 33.404 30.209 33.404
qs_ot_get_p 128 10.4 0.001 0.001 31.587 31.869
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.608 30.892
multiply_cannon_sync_h2d 60168 15.6 26.259 27.534 26.259 27.534
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.440 25.301
apply_single 128 13.6 0.001 0.001 24.440 25.301
qs_ot_p2m_diag 83 11.4 0.079 0.091 24.799 24.851
ot_diis_step 117 11.6 0.008 0.008 24.647 24.648
cp_dbcsr_syevd 83 12.4 0.005 0.005 21.966 21.967
init_scf_loop 11 6.9 0.000 0.001 20.835 20.840
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.197 20.372
multiply_cannon_metrocomm3 60168 15.6 0.118 0.122 16.528 18.976
cp_fm_diag_elpa 83 13.4 0.000 0.000 18.840 18.873
cp_fm_diag_elpa_base 83 14.4 18.762 18.802 18.835 18.870
prepare_preconditioner 11 7.9 0.000 0.000 16.199 16.246
make_preconditioner 11 8.9 0.000 0.000 16.199 16.246
make_full_inverse_cholesky 11 9.9 0.000 0.000 15.427 15.596
init_scf_run 11 5.9 0.000 0.001 14.941 14.942
scf_env_initial_rho_setup 11 6.9 0.000 0.001 14.941 14.942
sum_up_and_integrate 128 10.3 0.089 0.106 14.531 14.545
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.363 14.533
calculate_rho_elec 128 8.7 0.046 0.065 14.362 14.533
integrate_v_rspace 128 11.3 0.004 0.005 14.442 14.459
make_m2s 5014 13.6 0.103 0.112 14.134 14.452
make_images 5014 14.6 0.403 0.425 13.958 14.289
dbcsr_mm_accdrv_process 124484 16.2 4.684 4.821 9.937 11.366
density_rs2pw 128 9.7 0.007 0.008 7.537 10.892
mp_sum_l 7950 12.9 9.016 9.919 9.016 9.919
rs_pw_transfer 1046 11.9 0.017 0.020 6.161 9.542
multiply_cannon_metrocomm1 60168 15.6 0.095 0.098 6.888 9.310
cp_fm_cholesky_invert 11 10.9 9.274 9.282 9.274 9.282
wfi_extrapolate 11 7.9 0.001 0.001 9.245 9.245
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 9.123 9.135
calculate_dm_sparse 128 9.5 0.001 0.001 8.586 8.679
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 8.228 8.345
pw_transfer 1547 11.6 0.075 0.090 8.086 8.304
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.129 8.256
qs_ot_get_orbitals 117 10.6 0.001 0.001 8.112 8.193
fft_wrap_pw1pw2 1291 12.7 0.011 0.013 7.883 8.102
make_images_data 5014 15.6 0.070 0.076 6.936 7.819
grid_integrate_task_list 128 12.3 7.066 7.592 7.066 7.592
hybrid_alltoall_any 5200 16.5 0.298 2.285 6.076 7.408
fft_wrap_pw1pw2_140 523 13.2 1.274 1.327 6.760 6.955
mp_alltoall_d11v 2415 14.1 4.445 6.081 4.445 6.081
mp_waitany 16020 13.9 2.646 6.031 2.646 6.031
fft3d_ps 1291 14.7 2.139 2.780 5.643 5.996
grid_collocate_task_list 128 9.7 4.688 5.924 4.688 5.924
calculate_first_density_matrix 1 7.0 0.000 0.001 5.506 5.513
rs_pw_transfer_RS2PW_140 139 11.5 0.282 0.297 2.135 5.502
potential_pw2rs 128 12.3 0.009 0.010 4.857 4.896
cp_fm_cholesky_decompose 22 10.9 4.728 4.742 4.728 4.742
acc_transpose_blocks 60168 15.6 0.297 0.312 4.201 4.404
jit_kernel_multiply 13 15.9 3.213 4.313 3.213 4.313
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=214.706000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=597.818182, yerr=6.264158
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022950912 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963542011904 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444706349056 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019182452736 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019182452736 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796564E+12 0.0% 0.0% 100.0%
flops max/rank 2.166473E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499488 0.0% 0.0% 100.0%
number of processed stacks 5925696 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.6
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 841.416704E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2385600
MPI messages size (bytes):
total size 4.069299E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.705776E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70188 2295595008
32768 < size <= 131072 716032 54973693952
131072 < size <= 4194304 1363760 1386318135296
4194304 < size <= 16777216 153648 1453842592400
16777216 < size 67056 1171888537600
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3995 58313.
MP_Allreduce 11066 961.
MP_Sync 86
MP_Alltoall 1955 5290472.
MP_SendRecv 11938 47072.
MP_ISendRecv 11938 47072.
MP_Wait 25718
MP_ISend 11660 212488.
MP_IRecv 11660 212488.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.020 0.060 192.414 192.416
qs_mol_dyn_low 1 2.0 0.004 0.007 191.884 191.913
qs_forces 11 3.9 0.004 0.005 191.796 191.797
qs_energies 11 4.9 0.002 0.003 184.859 184.869
scf_env_do_scf 11 5.9 0.001 0.001 168.271 168.281
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 135.245 135.247
velocity_verlet 10 3.0 0.001 0.002 120.156 120.158
dbcsr_multiply_generic 2485 12.5 0.188 0.193 96.924 98.196
qs_scf_new_mos 116 7.6 0.001 0.001 95.290 95.788
qs_scf_loop_do_ot 116 8.6 0.001 0.001 95.289 95.787
ot_scf_mini 116 9.6 0.004 0.004 90.486 91.052
multiply_cannon 2485 13.5 0.497 0.549 76.691 81.285
multiply_cannon_loop 2485 14.5 1.552 1.617 73.368 76.083
ot_mini 116 10.6 0.001 0.001 49.886 50.384
multiply_cannon_multrec 29820 15.5 20.966 27.211 31.424 37.919
mp_waitall_1 212858 16.6 24.012 37.773 24.012 37.773
rebuild_ks_matrix 127 8.3 0.001 0.001 33.093 33.801
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.019 33.092 33.801
init_scf_loop 11 6.9 0.000 0.000 32.935 32.936
qs_ks_update_qs_env 127 7.6 0.001 0.001 29.673 30.314
qs_ot_get_derivative 116 11.6 0.001 0.002 28.133 28.689
prepare_preconditioner 11 7.9 0.000 0.000 28.543 28.615
make_preconditioner 11 8.9 0.000 0.000 28.543 28.615
multiply_cannon_metrocomm3 29820 15.5 0.095 0.100 15.402 28.185
make_full_inverse_cholesky 11 9.9 0.000 0.000 27.207 27.740
qs_ot_get_p 127 10.4 0.001 0.001 22.679 23.272
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 21.898 23.015
apply_single 127 13.6 0.001 0.001 21.898 23.015
ot_diis_step 116 11.6 0.014 0.015 21.574 21.575
multiply_cannon_sync_h2d 29820 15.5 17.960 20.289 17.960 20.289
qs_ot_p2m_diag 82 11.4 0.186 0.213 17.740 17.773
cp_fm_cholesky_invert 11 10.9 16.662 16.675 16.662 16.675
cp_dbcsr_syevd 82 12.4 0.005 0.006 16.537 16.538
make_m2s 4970 13.5 0.091 0.097 14.313 15.890
make_images 4970 14.5 1.163 1.371 14.105 15.679
sum_up_and_integrate 127 10.3 0.116 0.133 14.808 14.840
integrate_v_rspace 127 11.3 0.004 0.004 14.692 14.729
qs_rho_update_rho_low 127 7.7 0.001 0.001 14.289 14.319
calculate_rho_elec 127 8.7 0.088 0.105 14.288 14.318
cp_fm_diag_elpa 82 13.4 0.000 0.001 13.324 13.357
cp_fm_diag_elpa_base 82 14.4 13.065 13.167 13.318 13.346
init_scf_run 11 5.9 0.000 0.001 11.737 11.738
scf_env_initial_rho_setup 11 6.9 0.000 0.001 11.737 11.738
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 11.113 11.522
multiply_cannon_metrocomm4 27335 15.5 0.106 0.122 3.806 10.686
make_images_data 4970 15.5 0.068 0.078 8.481 10.474
dbcsr_mm_accdrv_process 61726 16.2 5.390 6.147 9.911 10.449
mp_irecv_dv 68888 16.3 3.602 10.290 3.602 10.290
density_rs2pw 127 9.7 0.007 0.007 7.600 9.988
hybrid_alltoall_any 5155 16.4 0.351 1.547 7.182 9.728
pw_transfer 1535 11.6 0.086 0.097 8.838 8.894
fft_wrap_pw1pw2 1281 12.7 0.011 0.012 8.615 8.672
wfi_extrapolate 11 7.9 0.001 0.001 8.490 8.490
rs_pw_transfer 1038 11.9 0.014 0.017 5.958 8.331
fft_wrap_pw1pw2_140 519 13.2 1.334 1.353 7.565 7.640
grid_integrate_task_list 127 12.3 7.138 7.634 7.138 7.634
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.310 7.126
cp_fm_cholesky_decompose 22 10.9 6.981 7.071 6.981 7.071
calculate_dm_sparse 127 9.5 0.001 0.001 6.465 6.618
mp_sum_l 7884 12.9 4.149 6.325 4.149 6.325
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.004 6.164 6.174
fft3d_ps 1281 14.7 2.789 2.965 5.925 5.964
grid_collocate_task_list 127 9.7 4.751 5.938 4.751 5.938
qs_ot_get_orbitals 116 10.6 0.001 0.001 5.337 5.409
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.261 5.408
potential_pw2rs 127 12.3 0.015 0.017 5.074 5.099
mp_allgather_i34 2485 14.5 1.856 5.082 1.856 5.082
mp_waitany 11660 13.9 2.529 4.882 2.529 4.882
mp_alltoall_d11v 2401 14.1 4.201 4.789 4.201 4.789
rs_pw_transfer_RS2PW_140 138 11.5 0.350 0.375 2.122 4.493
mp_sum_d 4448 12.1 2.626 4.083 2.626 4.083
dbcsr_complete_redistribute 393 12.7 0.774 0.849 3.206 4.051
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.809 3.871
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=192.416000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=800.545455, yerr=4.207431
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430460020736 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613089636352 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228663E+12 0.0% 0.0% 100.0%
flops max/rank 2.950886E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806316384 0.0% 0.0% 100.0%
number of processed stacks 4015680 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1694.9
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 966.942720E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1051232
MPI messages size (bytes):
total size 2.737039E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.603648E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 284160 37245419520
131072 < size <= 4194304 665408 1004401590272
4194304 < size <= 16777216 66080 937889764224
16777216 < size 28896 757491302400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4024 57897.
MP_Allreduce 11137 997.
MP_Sync 87
MP_Alltoall 1724 9394172.
MP_SendRecv 7998 75008.
MP_ISendRecv 7998 75008.
MP_Wait 21986
MP_ISend 11836 275177.
MP_IRecv 11836 275177.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.040 180.510 180.511
qs_mol_dyn_low 1 2.0 0.003 0.005 180.081 180.103
qs_forces 11 3.9 0.005 0.012 179.912 179.918
qs_energies 11 4.9 0.002 0.003 173.210 173.225
scf_env_do_scf 11 5.9 0.001 0.002 157.555 157.556
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 122.669 122.670
velocity_verlet 10 3.0 0.001 0.002 115.339 115.344
dbcsr_multiply_generic 2527 12.6 0.181 0.185 82.776 84.006
qs_scf_new_mos 118 7.6 0.001 0.001 83.349 83.720
qs_scf_loop_do_ot 118 8.6 0.001 0.001 83.348 83.719
ot_scf_mini 118 9.6 0.003 0.004 79.089 79.518
multiply_cannon 2527 13.6 0.513 0.538 62.732 67.404
multiply_cannon_loop 2527 14.6 1.137 1.203 59.582 62.725
ot_mini 118 10.6 0.001 0.001 43.073 43.522
init_scf_loop 11 6.9 0.000 0.000 34.786 34.787
mp_waitall_1 171878 16.6 24.962 34.473 24.962 34.473
rebuild_ks_matrix 129 8.3 0.001 0.001 31.422 31.933
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.018 31.422 31.932
prepare_preconditioner 11 7.9 0.000 0.000 30.701 30.750
make_preconditioner 11 8.9 0.000 0.000 30.701 30.750
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.339 29.743
qs_ks_update_qs_env 129 7.6 0.001 0.001 28.306 28.777
multiply_cannon_multrec 20216 15.6 13.238 16.481 22.559 25.738
multiply_cannon_metrocomm3 20216 15.6 0.062 0.065 15.285 25.182
qs_ot_get_derivative 118 11.6 0.001 0.002 23.330 23.772
qs_ot_get_p 129 10.4 0.001 0.001 21.179 21.666
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.766 20.759
apply_single 129 13.6 0.001 0.001 19.766 20.759
ot_diis_step 118 11.6 0.018 0.019 19.627 19.628
qs_ot_p2m_diag 83 11.4 0.265 0.272 16.690 16.697
make_m2s 5054 13.6 0.082 0.086 14.946 16.233
make_images 5054 14.6 1.153 1.253 14.712 15.995
cp_dbcsr_syevd 83 12.4 0.005 0.006 15.633 15.634
multiply_cannon_sync_h2d 20216 15.6 13.790 15.341 13.790 15.341
sum_up_and_integrate 129 10.3 0.133 0.146 14.942 14.975
qs_rho_update_rho_low 129 7.7 0.001 0.001 14.877 14.921
calculate_rho_elec 129 8.7 0.133 0.149 14.877 14.921
integrate_v_rspace 129 11.3 0.004 0.005 14.809 14.836
cp_fm_cholesky_invert 11 10.9 14.304 14.314 14.304 14.314
cp_fm_diag_elpa 83 13.4 0.000 0.001 12.305 12.328
cp_fm_diag_elpa_base 83 14.4 11.886 12.045 12.301 12.323
make_images_data 5054 15.6 0.064 0.075 8.952 10.651
init_scf_run 11 5.9 0.000 0.001 10.605 10.605
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.605 10.605
hybrid_alltoall_any 5240 16.5 0.454 2.051 7.775 10.047
density_rs2pw 129 9.7 0.006 0.007 7.496 9.803
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.186 9.504
multiply_cannon_metrocomm4 17689 15.6 0.066 0.076 3.506 9.358
pw_transfer 1559 11.6 0.087 0.102 9.112 9.217
mp_irecv_dv 50625 16.2 3.379 9.104 3.379 9.104
fft_wrap_pw1pw2 1301 12.7 0.010 0.012 8.886 8.995
dbcsr_mm_accdrv_process 41830 16.2 5.641 5.994 8.773 8.893
fft_wrap_pw1pw2_140 527 13.2 1.431 1.457 7.796 7.918
grid_integrate_task_list 129 12.3 7.411 7.811 7.411 7.811
rs_pw_transfer 1054 12.0 0.014 0.015 5.486 7.803
wfi_extrapolate 11 7.9 0.001 0.001 7.578 7.578
cp_fm_cholesky_decompose 22 10.9 7.328 7.394 7.328 7.394
cp_fm_upper_to_full 105 14.8 5.683 7.219 5.683 7.219
dbcsr_complete_redistribute 395 12.7 1.180 1.200 4.658 6.415
fft3d_ps 1301 14.7 2.762 2.990 5.970 6.042
calculate_dm_sparse 129 9.5 0.001 0.001 5.854 5.963
grid_collocate_task_list 129 9.7 5.018 5.896 5.018 5.896
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.481 5.491
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 4.797 5.379
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.457 5.210
mp_sum_l 8010 12.9 3.244 5.038 3.244 5.038
mp_alltoall_d11v 2423 14.1 4.420 4.998 4.420 4.998
mp_allgather_i34 2527 14.6 1.629 4.990 1.629 4.990
potential_pw2rs 129 12.3 0.020 0.022 4.923 4.950
mp_waitany 11836 13.9 2.387 4.728 2.387 4.728
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.555 4.684
rs_pw_transfer_RS2PW_140 140 11.5 0.334 0.354 1.965 4.241
transfer_fm_to_dbcsr 11 9.9 0.018 0.022 2.343 4.080
qs_ot_get_orbitals 118 10.6 0.001 0.001 3.970 4.008
mp_alltoall_i22 718 14.1 1.962 3.906 1.962 3.906
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 3.774 3.779
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.622 3.669
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=180.511000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=920.818182, yerr=6.899515
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 4.353788E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5977344 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.142301E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1143192
MPI messages size (bytes):
total size 2.023815E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.770319E+06
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 0 0
8192 < size <= 32768 396 8650752
32768 < size <= 131072 319024 36042702848
131072 < size <= 4194304 715736 785529176064
4194304 < size <= 16777216 70320 665379241840
16777216 < size 30720 536870912000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58348.
MP_Allreduce 11057 1083.
MP_Sync 87
MP_Alltoall 1712 12503107.
MP_SendRecv 5888 75008.
MP_ISendRecv 5888 75008.
MP_Wait 22442
MP_ISend 14952 244818.
MP_IRecv 14952 244818.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.044 0.159 191.828 191.829
qs_mol_dyn_low 1 2.0 0.004 0.005 191.255 191.292
qs_forces 11 3.9 0.005 0.012 191.147 191.150
qs_energies 11 4.9 0.004 0.020 183.881 183.889
scf_env_do_scf 11 5.9 0.001 0.001 166.974 166.982
velocity_verlet 10 3.0 0.002 0.003 126.090 126.093
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 120.807 120.809
qs_scf_new_mos 117 7.6 0.001 0.001 82.911 83.250
qs_scf_loop_do_ot 117 8.6 0.001 0.001 82.910 83.249
dbcsr_multiply_generic 2507 12.6 0.187 0.195 80.847 81.776
ot_scf_mini 117 9.6 0.003 0.005 78.328 78.682
multiply_cannon 2507 13.6 0.551 0.587 55.747 58.556
multiply_cannon_loop 2507 14.6 1.811 1.929 51.996 54.126
init_scf_loop 11 6.9 0.002 0.015 46.044 46.046
ot_mini 117 10.6 0.001 0.003 43.317 43.665
prepare_preconditioner 11 7.9 0.000 0.001 41.878 41.915
make_preconditioner 11 8.9 0.000 0.001 41.878 41.915
make_full_inverse_cholesky 11 9.9 0.013 0.024 35.725 40.688
multiply_cannon_multrec 30084 15.6 13.445 18.871 26.286 31.390
rebuild_ks_matrix 128 8.3 0.001 0.001 30.179 30.524
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.020 30.178 30.523
mp_waitall_1 147882 16.7 17.492 27.506 17.492 27.506
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.176 27.492
qs_ot_get_derivative 117 11.6 0.002 0.002 23.494 23.853
make_m2s 5014 13.6 0.097 0.103 20.620 21.803
qs_ot_get_p 128 10.4 0.001 0.002 21.138 21.509
make_images 5014 14.6 1.962 2.273 20.312 21.494
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 19.178 19.875
apply_single 128 13.6 0.001 0.001 19.177 19.875
ot_diis_step 117 11.6 0.019 0.025 19.693 19.695
qs_ot_p2m_diag 83 11.4 0.343 0.390 16.826 16.881
cp_fm_upper_to_full 105 14.8 11.410 16.779 11.410 16.779
cp_fm_cholesky_invert 11 10.9 16.122 16.131 16.122 16.131
cp_dbcsr_syevd 83 12.4 0.005 0.006 15.417 15.419
multiply_cannon_metrocomm3 30084 15.6 0.051 0.053 6.487 14.975
sum_up_and_integrate 128 10.3 0.140 0.151 14.835 14.857
qs_rho_update_rho_low 128 7.7 0.001 0.002 14.768 14.818
calculate_rho_elec 128 8.7 0.176 0.193 14.767 14.818
integrate_v_rspace 128 11.3 0.004 0.005 14.695 14.721
dbcsr_mm_accdrv_process 62264 16.2 8.322 9.173 12.415 13.004
make_images_data 5014 15.6 0.066 0.071 11.116 12.683
dbcsr_complete_redistribute 395 12.7 1.511 1.623 8.907 12.621
cp_fm_diag_elpa 83 13.4 0.000 0.000 12.114 12.131
cp_fm_diag_elpa_base 83 14.4 11.140 11.469 12.107 12.122
hybrid_alltoall_any 5200 16.5 0.531 2.225 9.799 11.961
multiply_cannon_sync_h2d 30084 15.6 10.446 11.606 10.446 11.606
copy_fm_to_dbcsr 209 11.7 0.001 0.002 7.535 11.231
init_scf_run 11 5.9 0.000 0.001 10.977 10.979
scf_env_initial_rho_setup 11 6.9 0.001 0.003 10.977 10.979
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.883 10.136
transfer_fm_to_dbcsr 11 9.9 0.001 0.005 6.131 9.739
pw_transfer 1547 11.6 0.087 0.102 9.263 9.359
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 9.036 9.138
mp_alltoall_i22 716 14.1 5.455 9.117 5.455 9.117
density_rs2pw 128 9.7 0.006 0.007 7.127 8.349
fft_wrap_pw1pw2_140 523 13.2 1.565 1.594 8.002 8.115
grid_integrate_task_list 128 12.3 7.518 7.886 7.518 7.886
wfi_extrapolate 11 7.9 0.001 0.001 7.813 7.813
cp_fm_cholesky_decompose 22 10.9 7.489 7.588 7.489 7.588
multiply_cannon_metrocomm4 25070 15.6 0.082 0.094 2.894 7.142
mp_irecv_dv 76098 16.2 2.739 6.864 2.739 6.864
calculate_dm_sparse 128 9.5 0.001 0.001 6.271 6.361
rs_pw_transfer 1046 11.9 0.013 0.014 4.784 6.080
fft3d_ps 1291 14.7 2.844 2.931 5.949 6.018
grid_collocate_task_list 128 9.7 5.157 5.855 5.157 5.855
mp_alltoall_d11v 2415 14.1 4.931 5.604 4.931 5.604
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.520 5.578
potential_pw2rs 128 12.3 0.023 0.024 4.601 4.622
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.509 4.611
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.399 4.528
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 4.463 4.463
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.222 4.286
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=191.829000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1076.636364, yerr=17.342920
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022121472 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444702699520 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796573E+12 0.0% 0.0% 100.0%
flops max/rank 5.820059E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499744 0.0% 0.0% 100.0%
number of processed stacks 1944496 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3448.5
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.547350E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 238560
MPI messages size (bytes):
total size 1.321104E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.537828E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 112800 59139686400
4194304 < size <= 16777216 104112 545846722560
16777216 < size 20064 716108580288
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8852 52.
MP_Alltoall 9584 804353.
MP_ISend 39716 2104723.
MP_IRecv 39716 2103824.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4003 58187.
MP_Allreduce 11085 1167.
MP_Sync 86
MP_Alltoall 1700 18828189.
MP_SendRecv 3810 122880.
MP_ISendRecv 3810 122880.
MP_Wait 16000
MP_ISend 10600 423612.
MP_IRecv 10600 423612.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.032 172.444 172.445
qs_mol_dyn_low 1 2.0 0.003 0.003 172.042 172.062
qs_forces 11 3.9 0.004 0.004 171.937 171.940
qs_energies 11 4.9 0.002 0.002 164.322 164.330
scf_env_do_scf 11 5.9 0.001 0.001 146.948 146.950
velocity_verlet 10 3.0 0.002 0.002 112.310 112.314
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 111.214 111.215
qs_scf_new_mos 116 7.6 0.001 0.001 72.688 72.784
qs_scf_loop_do_ot 116 8.6 0.001 0.001 72.687 72.784
dbcsr_multiply_generic 2485 12.5 0.178 0.191 71.437 71.703
ot_scf_mini 116 9.6 0.003 0.004 68.276 68.331
multiply_cannon 2485 13.5 0.557 0.586 52.706 56.005
multiply_cannon_loop 2485 14.5 0.795 0.819 49.736 50.376
ot_mini 116 10.6 0.001 0.001 37.113 37.165
init_scf_loop 11 6.9 0.000 0.000 35.586 35.587
prepare_preconditioner 11 7.9 0.000 0.000 31.663 31.692
make_preconditioner 11 8.9 0.000 0.000 31.663 31.692
mp_waitall_1 124680 16.7 24.452 30.615 24.452 30.615
rebuild_ks_matrix 127 8.3 0.001 0.001 29.864 29.927
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.018 29.864 29.927
make_full_inverse_cholesky 11 9.9 0.016 0.028 29.637 29.898
qs_ks_update_qs_env 127 7.6 0.001 0.001 27.079 27.134
multiply_cannon_multrec 9940 15.5 10.335 14.640 17.821 21.247
qs_ot_get_derivative 116 11.6 0.001 0.002 20.322 20.382
multiply_cannon_metrocomm3 9940 15.5 0.025 0.027 12.344 19.531
qs_ot_get_p 127 10.4 0.001 0.001 18.555 18.623
cp_fm_cholesky_invert 11 10.9 18.104 18.110 18.104 18.110
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 16.812 16.902
apply_single 127 13.6 0.001 0.001 16.811 16.902
ot_diis_step 116 11.6 0.020 0.020 16.720 16.720
make_m2s 4970 13.5 0.066 0.069 14.885 15.748
qs_rho_update_rho_low 127 7.7 0.001 0.001 15.424 15.467
calculate_rho_elec 127 8.7 0.257 0.267 15.424 15.466
make_images 4970 14.5 2.157 2.578 14.583 15.443
sum_up_and_integrate 127 10.3 0.180 0.190 15.022 15.076
integrate_v_rspace 127 11.3 0.004 0.004 14.842 14.905
qs_ot_p2m_diag 82 11.4 0.489 0.495 14.738 14.754
cp_dbcsr_syevd 82 12.4 0.005 0.006 13.574 13.575
multiply_cannon_sync_h2d 9940 15.5 10.821 11.309 10.821 11.309
cp_fm_diag_elpa 82 13.4 0.000 0.000 10.647 10.658
cp_fm_diag_elpa_base 82 14.4 10.405 10.485 10.642 10.653
init_scf_run 11 5.9 0.000 0.001 10.569 10.569
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.569 10.569
make_images_data 4970 15.5 0.056 0.064 8.549 9.922
hybrid_alltoall_any 5155 16.4 0.844 3.793 8.360 9.862
pw_transfer 1535 11.6 0.086 0.096 9.529 9.549
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 9.307 9.333
fft_wrap_pw1pw2_140 519 13.2 1.912 1.943 8.210 8.237
grid_integrate_task_list 127 12.3 7.694 8.233 7.694 8.233
cp_fm_cholesky_decompose 22 10.9 7.924 8.036 7.924 8.036
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 7.944 7.995
density_rs2pw 127 9.7 0.006 0.006 6.961 7.855
dbcsr_mm_accdrv_process 20590 16.1 3.270 4.184 7.135 7.839
wfi_extrapolate 11 7.9 0.001 0.001 7.462 7.462
multiply_cannon_metrocomm1 9940 15.5 0.029 0.030 4.300 7.324
calculate_dm_sparse 127 9.5 0.001 0.001 6.058 6.129
grid_collocate_task_list 127 9.7 5.431 6.126 5.431 6.126
mp_alltoall_d11v 2401 14.1 4.888 6.019 4.888 6.019
fft3d_ps 1281 14.7 2.741 2.821 5.791 5.818
dbcsr_complete_redistribute 393 12.7 2.105 2.182 5.132 5.536
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.258 5.259
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.224 5.233
rs_pw_transfer 1038 11.9 0.013 0.013 4.186 5.095
mp_allgather_i34 2485 14.5 1.172 4.640 1.172 4.640
potential_pw2rs 127 12.3 0.026 0.026 4.458 4.479
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.143 4.191
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.588 3.882
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.682 3.690
multiply_cannon_metrocomm4 7455 15.5 0.025 0.029 1.705 3.663
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.600 3.661
copy_fm_to_dbcsr 208 11.6 0.001 0.002 3.282 3.607
mp_irecv_dv 28618 15.9 1.669 3.596 1.669 3.596
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.537 3.559
copy_dbcsr_to_fm 185 11.7 0.004 0.004 3.469 3.550
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=172.445000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1462.545455, yerr=28.541401
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 11.786061E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 1980288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3437.1
marketing flops 145.650931E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 3.207709E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101160
MPI messages size (bytes):
total size 1.144970E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.318403E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 44 2883584
131072 < size <= 4194304 45648 35433480192
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592466352
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 58558.
MP_Allreduce 11184 1500.
MP_Sync 88
MP_Alltoall 1724 36993632.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.042 305.159 305.160
qs_mol_dyn_low 1 2.0 0.003 0.004 304.598 304.617
qs_forces 11 3.9 0.004 0.004 304.504 304.506
qs_energies 11 4.9 0.050 0.051 294.978 294.987
scf_env_do_scf 11 5.9 0.001 0.001 272.000 272.009
velocity_verlet 10 3.0 0.002 0.002 217.748 217.756
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 146.240 146.241
init_scf_loop 11 6.9 0.000 0.000 125.470 125.472
prepare_preconditioner 11 7.9 0.000 0.000 120.399 120.418
make_preconditioner 11 8.9 0.000 0.000 120.399 120.418
make_full_inverse_cholesky 11 9.9 0.038 0.039 96.517 117.555
qs_scf_new_mos 118 7.6 0.001 0.001 95.677 95.753
qs_scf_loop_do_ot 118 8.6 0.001 0.001 95.676 95.752
ot_scf_mini 118 9.6 0.004 0.004 90.687 90.722
dbcsr_multiply_generic 2529 12.6 0.213 0.226 85.730 86.146
cp_fm_upper_to_full 106 14.8 53.593 76.465 53.593 76.465
multiply_cannon 2529 13.6 0.681 0.728 60.328 61.887
multiply_cannon_loop 2529 14.6 1.085 1.108 56.274 57.593
ot_mini 118 10.6 0.001 0.001 47.422 47.460
dbcsr_complete_redistribute 397 12.7 4.001 4.057 29.952 42.491
copy_fm_to_dbcsr 210 11.7 0.001 0.002 26.388 38.924
rebuild_ks_matrix 129 8.3 0.001 0.001 37.990 38.030
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.017 37.990 38.029
transfer_fm_to_dbcsr 11 9.9 0.030 0.030 23.840 36.302
qs_ks_update_qs_env 129 7.6 0.001 0.001 34.912 34.939
mp_alltoall_i22 720 14.1 21.615 34.299 21.615 34.299
cp_fm_cholesky_invert 11 10.9 33.743 33.749 33.743 33.749
mp_waitall_1 104580 16.8 28.898 33.345 28.898 33.345
qs_ot_get_p 129 10.4 0.001 0.001 27.953 28.009
qs_ot_get_derivative 118 11.6 0.002 0.002 27.464 27.499
qs_ot_p2m_diag 84 11.4 0.891 0.897 23.722 23.753
cp_dbcsr_syevd 84 12.4 0.006 0.006 21.857 21.859
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 21.086 21.476
apply_single 129 13.6 0.001 0.001 21.086 21.475
qs_rho_update_rho_low 129 7.7 0.001 0.001 20.813 20.833
calculate_rho_elec 129 8.7 0.486 0.487 20.812 20.833
multiply_cannon_metrocomm3 10116 15.6 0.025 0.027 19.657 20.801
multiply_cannon_multrec 10116 15.6 10.411 12.314 18.517 20.349
make_m2s 5058 13.6 0.076 0.079 18.751 20.200
ot_diis_step 118 11.6 0.022 0.023 19.925 19.925
sum_up_and_integrate 129 10.3 0.323 0.324 19.771 19.858
make_images 5058 14.6 3.084 3.284 18.268 19.721
integrate_v_rspace 129 11.3 0.004 0.004 19.448 19.535
cp_fm_diag_elpa 84 13.4 0.000 0.000 18.522 18.523
cp_fm_diag_elpa_base 84 14.4 14.015 15.592 18.516 18.516
multiply_cannon_sync_h2d 10116 15.6 14.374 14.455 14.374 14.455
hybrid_alltoall_any 5245 16.5 1.327 3.079 10.606 12.746
make_images_data 5058 15.6 0.064 0.070 10.474 12.667
init_scf_run 11 5.9 0.000 0.001 12.501 12.502
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.501 12.502
pw_transfer 1559 11.6 0.093 0.095 11.738 11.744
fft_wrap_pw1pw2 1301 12.7 0.011 0.012 11.501 11.508
fft_wrap_pw1pw2_140 527 13.2 3.124 3.194 10.208 10.218
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 10.083 10.117
mp_alltoall_d11v 2429 14.1 8.248 9.693 8.248 9.693
dbcsr_mm_accdrv_process 20934 16.1 4.630 6.183 7.856 9.458
wfi_extrapolate 11 7.9 0.001 0.001 9.309 9.309
cp_fm_cholesky_decompose 22 10.9 9.122 9.137 9.122 9.137
grid_integrate_task_list 129 12.3 8.644 8.866 8.644 8.866
density_rs2pw 129 9.7 0.006 0.006 8.191 8.354
qs_energies_init_hamiltonians 11 5.9 0.007 0.012 8.196 8.196
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.899 7.171
calculate_dm_sparse 129 9.5 0.001 0.001 6.878 6.965
grid_collocate_task_list 129 9.7 6.482 6.578 6.482 6.578
fft3d_ps 1301 14.7 2.874 2.890 6.509 6.565
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.431 6.468
copy_dbcsr_to_fm 187 11.8 0.004 0.004 6.244 6.341
rs_scatter_matrices 140 9.7 3.927 4.858 6.121 6.340
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=305.160000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2909.090909, yerr=155.267192
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/21/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 2.766000E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 419739 0.0% 0.0% 100.0%
average stack size 0.0 0.0 22952.9
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 1.259569E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255646.
MP_Allreduce 3139 6114.
MP_Sync 4
MP_Alltoall 54
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.038 85.346 85.347
qs_energies 1 2.0 0.000 0.001 84.819 84.833
ls_scf 1 3.0 0.000 0.002 83.911 83.924
dbcsr_multiply_generic 111 6.7 0.016 0.017 72.807 72.956
multiply_cannon 111 7.7 0.018 0.021 55.948 57.120
multiply_cannon_loop 111 8.7 0.228 0.244 52.544 53.929
ls_scf_main 1 4.0 0.000 0.003 52.569 52.570
density_matrix_trs4 2 5.0 0.002 0.003 47.002 47.074
ls_scf_init_scf 1 4.0 0.000 0.001 28.298 28.299
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.149 27.203
mp_waitall_1 11031 10.9 22.640 26.789 22.640 26.789
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 25.052 25.071
multiply_cannon_multrec 2664 9.7 8.151 8.823 15.575 17.187
multiply_cannon_sync_h2d 2664 9.7 13.447 14.842 13.447 14.842
make_m2s 222 7.7 0.009 0.012 13.229 13.687
make_images 222 8.7 0.101 0.112 13.207 13.666
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.752 13.002
make_images_data 222 9.7 0.005 0.007 7.740 8.382
multiply_cannon_metrocomm3 2664 9.7 0.010 0.011 5.483 8.261
hybrid_alltoall_any 227 10.6 0.218 1.851 6.648 8.197
dbcsr_mm_accdrv_process 4760 10.4 0.591 0.705 7.043 8.026
dbcsr_mm_accdrv_process_sort 4760 11.4 6.247 7.132 6.247 7.132
calculate_norms 4752 9.8 5.513 6.152 5.513 6.152
mp_sum_l 887 5.1 3.096 5.435 3.096 5.435
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.011 5.107
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.316 4.243
make_images_sizes 222 9.7 0.000 0.000 0.810 3.869
mp_alltoall_i44 222 10.7 0.810 3.869 0.810 3.869
multiply_cannon_metrocomm4 2442 9.7 0.012 0.016 2.088 3.676
mp_irecv_dv 6231 10.9 2.071 3.642 2.071 3.642
arnoldi_extremal 4 6.8 0.000 0.000 3.261 3.284
arnoldi_normal_ev 4 7.8 0.001 0.002 3.261 3.284
build_subspace 16 8.4 0.009 0.012 3.169 3.172
ls_scf_post 1 4.0 0.000 0.001 3.043 3.057
ls_scf_store_result 1 5.0 0.000 0.000 2.869 2.904
dbcsr_special_finalize 555 9.7 0.006 0.006 2.313 2.779
dbcsr_merge_single_wm 555 10.7 0.453 0.562 2.304 2.769
make_images_pack 222 9.7 2.216 2.644 2.218 2.645
dbcsr_matrix_vector_mult 304 9.0 0.006 0.014 2.332 2.569
dbcsr_sort_data 658 11.4 2.109 2.546 2.109 2.546
dbcsr_matrix_vector_mult_local 304 10.0 2.068 2.455 2.071 2.457
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.343 2.434
buffer_matrices_ensure_size 222 8.7 1.751 2.080 1.751 2.080
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.795 1.797
rebuild_ks_matrix 3 7.3 0.000 0.000 1.785 1.788
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.009 1.785 1.788
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=85.347000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1131.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/22/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 5.588524E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 368848 0.0% 0.0% 100.0%
average stack size 0.0 0.0 26119.8
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 2.163479E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 10075.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.033 0.049 91.052 91.053
qs_energies 1 2.0 0.000 0.000 90.483 90.487
ls_scf 1 3.0 0.000 0.000 89.100 89.104
dbcsr_multiply_generic 111 6.7 0.016 0.017 75.051 75.360
multiply_cannon 111 7.7 0.028 0.045 52.980 58.037
ls_scf_main 1 4.0 0.000 0.000 55.385 55.390
multiply_cannon_loop 111 8.7 0.135 0.145 50.120 53.634
density_matrix_trs4 2 5.0 0.002 0.003 49.685 49.925
mp_waitall_1 9105 10.9 21.041 30.678 21.041 30.678
ls_scf_init_scf 1 4.0 0.000 0.002 30.065 30.067
ls_scf_init_matrix_S 1 5.0 0.000 0.001 28.847 28.941
multiply_cannon_multrec 1332 9.7 13.369 17.017 22.705 27.448
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.465 26.478
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.588 20.942
make_m2s 222 7.7 0.007 0.009 15.310 15.919
make_images 222 8.7 1.386 1.724 15.279 15.890
dbcsr_mm_accdrv_process 4041 10.4 0.327 0.539 8.933 10.519
dbcsr_mm_accdrv_process_sort 4041 11.4 8.450 9.980 8.450 9.980
make_images_data 222 9.7 0.005 0.005 8.939 9.886
hybrid_alltoall_any 227 10.6 0.545 2.542 8.281 9.744
mp_sum_l 887 5.1 5.281 8.534 5.281 8.534
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.182 7.711
mp_irecv_dv 3311 11.0 3.162 7.656 3.162 7.656
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.090 6.830
calculate_norms 2376 9.8 6.066 6.689 6.066 6.689
multiply_cannon_sync_h2d 1332 9.7 4.905 6.091 4.905 6.091
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.935 5.183
arnoldi_extremal 4 6.8 0.000 0.000 4.731 4.760
arnoldi_normal_ev 4 7.8 0.001 0.004 4.730 4.760
build_subspace 16 8.4 0.014 0.021 4.470 4.474
ls_scf_post 1 4.0 0.000 0.000 3.650 3.655
ls_scf_store_result 1 5.0 0.000 0.000 3.377 3.471
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.199 3.432
dbcsr_matrix_vector_mult_local 304 10.0 2.781 3.267 2.783 3.269
mp_allgather_i34 111 8.7 0.842 3.152 0.842 3.152
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.641 2.729
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.168 2.688
dbcsr_data_new 4174 10.1 2.116 2.416 2.116 2.416
make_images_pack 222 9.7 1.844 2.140 1.847 2.143
dbcsr_sort_data 436 11.2 1.837 2.067 1.837 2.067
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.882 1.884
rebuild_ks_matrix 3 7.3 0.000 0.000 1.869 1.872
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.869 1.872
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=91.053000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1806.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/23/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 8.404608E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 353133 0.0% 0.0% 100.0%
average stack size 0.0 0.0 27282.1
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 2.883617E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265448.
MP_Allreduce 3138 10896.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.044 93.326 93.327
qs_energies 1 2.0 0.000 0.000 92.783 92.786
ls_scf 1 3.0 0.000 0.000 91.366 91.369
dbcsr_multiply_generic 111 6.7 0.016 0.017 75.845 76.201
ls_scf_main 1 4.0 0.000 0.000 57.206 57.211
multiply_cannon 111 7.7 0.035 0.074 52.274 56.784
multiply_cannon_loop 111 8.7 0.117 0.130 49.483 53.767
density_matrix_trs4 2 5.0 0.002 0.003 51.258 51.453
mp_waitall_1 7281 11.0 23.873 33.606 23.873 33.606
ls_scf_init_scf 1 4.0 0.000 0.001 30.450 30.453
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.214 29.311
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.855 26.869
multiply_cannon_multrec 888 9.7 12.723 15.075 21.329 24.264
multiply_cannon_metrocomm3 888 9.7 0.004 0.005 11.080 23.063
make_m2s 222 7.7 0.007 0.008 16.610 17.368
make_images 222 8.7 1.597 1.870 16.572 17.327
make_images_data 222 9.7 0.004 0.005 9.717 10.788
hybrid_alltoall_any 227 10.6 0.643 2.955 9.243 10.483
dbcsr_mm_accdrv_process 3754 10.4 0.318 0.493 8.128 9.379
mp_sum_l 887 5.1 5.253 9.191 5.253 9.191
dbcsr_mm_accdrv_process_sort 3754 11.4 7.692 8.886 7.692 8.886
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.475 7.164
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.014 7.123
mp_irecv_dv 2335 11.1 2.459 7.094 2.459 7.094
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.721 7.050
multiply_cannon_sync_h2d 888 9.7 5.988 6.918 5.988 6.918
arnoldi_extremal 4 6.8 0.000 0.000 5.216 5.235
arnoldi_normal_ev 4 7.8 0.001 0.005 5.216 5.235
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.746 4.969
calculate_norms 1584 9.8 4.415 4.944 4.415 4.944
build_subspace 16 8.4 0.014 0.020 4.916 4.924
mp_allgather_i34 111 8.7 0.900 3.871 0.900 3.871
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.517 3.851
ls_scf_post 1 4.0 0.000 0.000 3.711 3.714
dbcsr_matrix_vector_mult_local 304 10.0 3.069 3.656 3.071 3.658
ls_scf_store_result 1 5.0 0.000 0.000 3.464 3.538
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.894 2.992
dbcsr_data_new 4116 9.9 2.105 2.459 2.105 2.459
make_images_sizes 222 9.7 0.000 0.000 1.019 2.200
mp_alltoall_i44 222 10.7 1.019 2.200 1.019 2.200
dbcsr_sort_data 325 11.1 1.893 2.157 1.893 2.157
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.915 1.917
make_images_pack 222 9.7 1.637 1.898 1.640 1.900
rebuild_ks_matrix 3 7.3 0.000 0.000 1.897 1.899
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.897 1.899
dbcsr_finalize 304 7.8 0.026 0.032 1.611 1.877
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=93.327000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2209.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/24/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 10.747127E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 369794 0.0% 0.0% 100.0%
average stack size 0.0 0.0 26053.0
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 3.358323E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 13030.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.035 0.045 97.271 97.273
qs_energies 1 2.0 0.000 0.001 96.625 96.629
ls_scf 1 3.0 0.000 0.001 94.913 94.916
dbcsr_multiply_generic 111 6.7 0.018 0.020 78.619 78.822
ls_scf_main 1 4.0 0.001 0.003 58.938 58.939
multiply_cannon 111 7.7 0.050 0.101 51.822 55.880
density_matrix_trs4 2 5.0 0.004 0.008 52.809 52.921
multiply_cannon_loop 111 8.7 0.152 0.167 46.837 49.627
ls_scf_init_scf 1 4.0 0.001 0.003 32.724 32.727
ls_scf_init_matrix_S 1 5.0 0.000 0.002 31.502 31.566
mp_waitall_1 6369 11.0 22.771 29.954 22.771 29.954
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.026 29.043
multiply_cannon_multrec 1332 9.7 14.174 17.297 22.160 25.381
make_m2s 222 7.7 0.008 0.009 21.266 22.600
make_images 222 8.7 3.175 3.656 21.215 22.551
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 9.215 16.474
make_images_data 222 9.7 0.005 0.006 11.829 13.466
hybrid_alltoall_any 227 10.6 0.800 3.760 11.225 13.005
dbcsr_mm_accdrv_process 3641 10.4 0.288 0.485 7.606 9.131
dbcsr_mm_accdrv_process_sort 3641 11.4 7.164 8.634 7.164 8.634
mp_sum_l 887 5.1 3.951 7.224 3.951 7.224
multiply_cannon_sync_h2d 1332 9.7 5.490 6.144 5.490 6.144
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.112 6.028
mp_irecv_dv 3229 10.9 2.086 5.956 2.086 5.956
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.999 5.708
arnoldi_extremal 4 6.8 0.000 0.000 5.249 5.267
arnoldi_normal_ev 4 7.8 0.001 0.004 5.249 5.267
build_subspace 16 8.4 0.014 0.021 4.911 4.919
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.565 4.724
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.526 4.691
calculate_norms 2376 9.8 4.193 4.552 4.193 4.552
mp_allgather_i34 111 8.7 2.086 4.207 2.086 4.207
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 3.640 3.937
dbcsr_matrix_vector_mult_local 304 10.0 3.253 3.760 3.255 3.762
dbcsr_sort_data 658 11.4 3.094 3.403 3.094 3.403
ls_scf_post 1 4.0 0.000 0.001 3.250 3.254
dbcsr_special_finalize 555 9.7 0.006 0.007 2.854 3.237
dbcsr_merge_single_wm 555 10.7 0.539 0.664 2.846 3.229
ls_scf_dm_to_ks 2 5.0 0.000 0.001 3.033 3.094
ls_scf_store_result 1 5.0 0.000 0.000 2.974 3.044
dbcsr_data_release 10477 10.7 1.603 2.431 1.603 2.431
dbcsr_finalize 304 7.8 0.050 0.062 1.802 1.995
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.273000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2707.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/25/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 15.383312E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 336818 0.0% 0.0% 100.0%
average stack size 0.0 0.0 28603.7
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 4.735197E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265536.
MP_Allreduce 3129 15263.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.042 0.057 92.237 92.238
qs_energies 1 2.0 0.000 0.000 91.487 91.490
ls_scf 1 3.0 0.000 0.000 89.546 89.549
dbcsr_multiply_generic 111 6.7 0.018 0.018 71.002 71.173
ls_scf_main 1 4.0 0.000 0.000 56.843 56.844
multiply_cannon 111 7.7 0.079 0.173 52.978 56.350
multiply_cannon_loop 111 8.7 0.088 0.096 50.385 51.698
density_matrix_trs4 2 5.0 0.002 0.003 49.875 49.943
ls_scf_init_scf 1 4.0 0.000 0.001 29.314 29.316
mp_waitall_1 5436 11.0 24.706 28.258 24.706 28.258
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.036 28.063
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.954 25.967
multiply_cannon_multrec 444 9.7 13.662 16.671 20.761 22.231
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 10.907 15.684
make_m2s 222 7.7 0.005 0.006 13.440 14.388
make_images 222 8.7 2.046 2.487 13.372 14.319
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 6.431 13.862
make_images_data 222 9.7 0.003 0.004 8.187 9.631
hybrid_alltoall_any 227 10.6 0.802 3.803 8.098 9.608
multiply_cannon_sync_h2d 444 9.7 6.769 8.083 6.769 8.083
dbcsr_mm_accdrv_process 3003 10.4 0.354 0.404 6.792 7.946
dbcsr_mm_accdrv_process_sort 3003 11.4 6.426 7.543 6.426 7.543
arnoldi_extremal 4 6.8 0.000 0.000 5.900 5.913
arnoldi_normal_ev 4 7.8 0.001 0.004 5.900 5.913
build_subspace 16 8.4 0.015 0.019 5.507 5.518
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.355 4.557
dbcsr_matrix_vector_mult 304 9.0 0.011 0.022 4.243 4.449
dbcsr_matrix_vector_mult_local 304 10.0 3.783 4.259 3.785 4.261
mp_sum_l 887 5.1 2.565 4.214 2.565 4.214
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.542 4.127
mp_irecv_dv 1241 11.2 1.529 4.078 1.529 4.078
mp_allgather_i34 111 8.7 1.160 3.746 1.160 3.746
calculate_norms 792 9.8 3.633 3.729 3.633 3.729
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.561 3.597
ls_scf_post 1 4.0 0.000 0.000 3.389 3.392
ls_scf_store_result 1 5.0 0.000 0.000 3.173 3.215
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.787 3.194
make_images_sizes 222 9.7 0.000 0.000 0.838 3.105
mp_alltoall_i44 222 10.7 0.837 3.105 0.837 3.105
dbcsr_finalize 304 7.8 0.062 0.077 2.210 2.301
dbcsr_data_new 4608 9.7 1.795 2.220 1.795 2.220
dbcsr_merge_all 275 8.9 0.480 0.524 2.065 2.140
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.067 2.068
rebuild_ks_matrix 3 7.3 0.000 0.000 2.035 2.036
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.035 2.036
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 1.925 1.925
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=92.238000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3730.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/26/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 30.358840E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 339931 0.0% 0.0% 100.0%
average stack size 0.0 0.0 28341.7
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 8.895451E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284089.
MP_Allreduce 3123 21388.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.077 0.108 105.590 105.590
qs_energies 1 2.0 0.000 0.000 104.172 104.178
ls_scf 1 3.0 0.000 0.000 101.240 101.245
dbcsr_multiply_generic 111 6.7 0.024 0.027 74.753 74.893
ls_scf_main 1 4.0 0.000 0.000 63.870 63.870
density_matrix_trs4 2 5.0 0.002 0.003 54.827 54.890
multiply_cannon 111 7.7 0.140 0.197 48.518 50.850
multiply_cannon_loop 111 8.7 0.098 0.100 45.401 46.017
ls_scf_init_scf 1 4.0 0.001 0.001 33.592 33.592
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.020 32.045
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.309 29.315
mp_waitall_1 4527 11.1 21.681 25.615 21.681 25.615
make_m2s 222 7.7 0.006 0.006 22.611 23.785
make_images 222 8.7 3.599 3.888 22.500 23.672
multiply_cannon_multrec 444 9.7 17.907 18.552 22.554 23.220
hybrid_alltoall_any 227 10.6 1.659 3.630 12.758 15.592
make_images_data 222 9.7 0.004 0.004 12.969 15.198
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.129 10.488
multiply_cannon_sync_h2d 444 9.7 8.787 8.843 8.787 8.843
arnoldi_extremal 4 6.8 0.000 0.000 7.498 7.506
arnoldi_normal_ev 4 7.8 0.002 0.008 7.498 7.506
build_subspace 16 8.4 0.026 0.036 6.946 6.958
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.605 5.756
dbcsr_matrix_vector_mult_local 304 10.0 5.194 5.497 5.196 5.500
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.175 5.265
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.865 5.117
dbcsr_mm_accdrv_process 1814 10.4 0.266 0.357 4.474 4.604
dbcsr_mm_accdrv_process_sort 1814 11.4 4.139 4.273 4.139 4.273
mp_allgather_i34 111 8.7 1.218 3.908 1.218 3.908
ls_scf_post 1 4.0 0.000 0.000 3.778 3.784
make_images_sizes 222 9.7 0.000 0.000 1.461 3.561
mp_alltoall_i44 222 10.7 1.461 3.561 1.461 3.561
ls_scf_store_result 1 5.0 0.000 0.000 3.500 3.536
calculate_norms 792 9.8 3.233 3.274 3.233 3.274
dbcsr_finalize 304 7.8 0.082 0.090 3.084 3.171
dbcsr_merge_all 275 8.9 0.886 0.917 2.869 2.949
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.902 2.902
dbcsr_complete_redistribute 5 7.6 1.443 1.476 2.743 2.858
matrix_ls_to_qs 2 6.0 0.000 0.000 2.416 2.547
dbcsr_sort_data 325 11.1 2.445 2.497 2.445 2.497
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.457 2.459
rebuild_ks_matrix 3 7.3 0.000 0.000 2.391 2.393
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.391 2.393
dbcsr_data_new 6591 9.6 1.855 2.362 1.855 2.362
dbcsr_new_transposed 4 7.5 0.247 0.263 2.286 2.295
dbcsr_frobenius_norm 74 6.6 2.055 2.123 2.191 2.224
dbcsr_add_d 103 6.2 0.000 0.000 2.128 2.206
dbcsr_add_anytype 103 7.2 0.859 0.890 2.128 2.206
dbcsr_data_release 12724 10.6 1.966 2.172 1.966 2.172
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=105.590000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=7082.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/18d5c6eecb1583739a723fdc877dd6e49c5a3a43_performance_tests/27/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 7009386627072 0.0% 0.0% 100.0%
flops 9 x 9 x 32 7335108845568 0.0% 0.0% 100.0%
flops 9 x 22 x 32 9866241589248 0.0% 0.0% 100.0%
flops 22 x 9 x 32 9884108906496 0.0% 0.0% 100.0%
flops 22 x 22 x 32 13354440523776 0.0% 0.0% 100.0%
flops 32 x 32 x 9 20607185977344 0.0% 0.0% 100.0%
flops 32 x 32 x 22 25186560638976 0.0% 0.0% 100.0%
flops 9 x 32 x 32 28458319085568 0.0% 0.0% 100.0%
flops 22 x 32 x 32 34782389993472 0.0% 0.0% 100.0%
flops 9 x 32 x 9 42881542373376 0.0% 0.0% 100.0%
flops 22 x 32 x 9 55680402235392 0.0% 0.0% 100.0%
flops 9 x 32 x 22 55680402235392 0.0% 0.0% 100.0%
flops 22 x 32 x 22 72328573419520 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 383.054662E+12 0.0% 0.0% 100.0%
flops max/rank 733.641090E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 26899403712 0.0% 0.0% 100.0%
number of processed stacks 118860288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 226.3
marketing flops 780.439111E+12
-------------------------------------------------------------------------------
# multiplications 1445
max memory usage/rank 588.271616E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 78072.
MP_Allreduce 13232 2081.
MP_Sync 1064
MP_Alltoall 2588 4057468.
MP_SendRecv 168740 11136.
MP_ISendRecv 92040 11136.
MP_Wait 102830
MP_comm_split 40
MP_ISend 26090 85106.
MP_IRecv 37890 59644.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.057 0.258 230.234 230.235
qs_mol_dyn_low 1 2.0 0.007 0.013 229.045 229.073
qs_forces 5 3.8 0.005 0.021 228.942 228.950
qs_energies 5 4.8 0.004 0.045 225.990 226.023
scf_env_do_scf 5 5.8 0.000 0.002 207.903 207.906
scf_env_do_scf_inner_loop 105 6.6 0.002 0.007 181.400 181.402
qs_scf_new_mos 105 7.6 0.000 0.001 141.144 141.301
qs_scf_loop_do_ot 105 8.6 0.001 0.001 141.144 141.301
dbcsr_multiply_generic 1445 12.2 0.124 0.132 136.527 137.005
ot_scf_mini 105 9.6 0.003 0.004 131.219 131.367
multiply_cannon 1445 13.2 0.275 0.285 117.908 120.203
multiply_cannon_loop 1445 14.2 2.746 3.002 116.235 117.657
velocity_verlet 4 3.0 0.005 0.027 106.534 106.535
ot_mini 105 10.6 0.001 0.003 59.804 59.927
multiply_cannon_multrec 69360 15.2 30.102 35.054 39.479 44.266
mp_waitall_1 488190 16.1 35.042 42.470 35.042 42.470
qs_ot_get_p 112 10.4 0.001 0.001 41.502 41.799
qs_ot_get_derivative 55 11.6 0.001 0.001 38.051 38.215
multiply_cannon_metrocomm3 69360 15.2 0.209 0.221 26.072 34.518
multiply_cannon_sync_h2d 69360 15.2 29.637 33.627 29.637 33.627
qs_ot_p2m_diag 40 11.0 0.020 0.031 30.166 30.291
rebuild_ks_matrix 110 8.4 0.000 0.000 28.833 29.024
qs_ks_build_kohn_sham_matrix 110 9.4 0.012 0.020 28.833 29.024
cp_dbcsr_syevd 40 12.0 0.002 0.004 27.066 27.066
qs_ks_update_qs_env 112 7.6 0.001 0.001 26.490 26.656
init_scf_loop 7 6.6 0.000 0.005 26.474 26.475
apply_preconditioner_dbcsr 62 12.6 0.000 0.001 23.026 23.272
apply_single 62 13.6 0.000 0.000 23.026 23.272
cp_fm_syevd 40 13.0 0.000 0.002 21.963 22.117
prepare_preconditioner 7 7.6 0.000 0.001 21.613 21.647
make_preconditioner 7 8.6 0.000 0.005 21.613 21.647
ot_new_cg_direction 55 11.6 0.001 0.002 21.007 21.007
qs_rho_update_rho_low 110 7.6 0.001 0.001 17.143 17.570
calculate_rho_elec 110 8.6 0.030 0.034 17.143 17.570
cp_fm_redistribute_end 40 14.0 8.644 17.248 8.648 17.249
cp_fm_syevd_base 40 14.0 8.592 17.199 8.592 17.199
init_scf_run 5 5.8 0.000 0.001 15.358 15.359
scf_env_initial_rho_setup 5 6.8 0.000 0.002 15.358 15.359
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.726 14.887
make_full_inverse_cholesky 7 9.6 0.000 0.001 14.502 14.579
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 13.716 13.826
mp_sum_l 4764 12.2 11.953 12.908 11.953 12.908
calculate_dm_sparse 110 9.5 0.000 0.001 12.674 12.891
rs_pw_transfer 690 11.5 0.011 0.015 11.360 12.560
density_rs2pw 110 9.6 0.006 0.008 11.226 12.456
pw_transfer 1645 12.4 0.082 0.104 11.830 12.009
fft_wrap_pw1pw2 1425 13.5 0.013 0.016 11.689 11.867
acc_transpose_blocks 69360 15.2 0.336 0.373 10.354 11.366
calculate_first_density_matrix 1 7.0 0.000 0.004 10.520 10.538
dbcsr_mm_accdrv_process 154766 15.8 5.766 6.330 9.243 10.454
qs_vxc_create 110 10.4 0.002 0.004 10.291 10.322
cp_fm_cholesky_invert 7 10.6 10.221 10.228 10.221 10.228
fft_wrap_pw1pw2_240 915 15.0 1.204 1.286 9.928 10.146
qs_ot_get_derivative_diag 18 12.0 0.000 0.000 10.046 10.110
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.002 9.774 9.790
check_diag 80 13.5 8.596 8.889 9.456 9.607
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.001 9.338 9.400
acc_transpose_blocks_kernels 69360 16.2 0.797 0.878 7.788 8.597
sum_up_and_integrate 60 10.3 0.028 0.037 8.542 8.556
integrate_v_rspace 60 11.3 0.002 0.002 8.514 8.526
fft3d_pb 915 16.0 2.377 2.552 7.972 8.156
multiply_cannon_metrocomm1 69360 15.2 0.096 0.104 4.827 7.983
jit_kernel_transpose 5 15.0 6.991 7.798 6.991 7.798
xc_rho_set_and_dset_create 110 12.4 0.077 0.097 7.261 7.507
xc_vxc_pw_create 60 11.3 0.039 0.050 6.889 6.920
make_full_single_inverse 7 9.6 0.001 0.002 6.835 6.869
make_m2s 2890 13.2 0.081 0.091 6.227 6.819
make_images 2890 14.2 0.241 0.262 6.120 6.711
xc_pw_derive 510 13.4 0.005 0.007 6.079 6.164
mp_waitany 7680 13.5 4.340 5.637 4.340 5.637
mp_alltoall_z22v 2340 17.7 5.020 5.316 5.020 5.316
multiply_cannon_metrocomm4 67915 15.2 0.185 0.202 2.017 5.021
potential_pw2rs 60 12.3 0.003 0.003 4.766 4.787
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=230.235000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=560.600000, yerr=1.854724
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 18d5c6eecb1583739a723fdc877dd6e49c5a3a43
Summary: empty
Status: OK