=== 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: e709bc9b963055c87db2d7fb03b6a1948d4b7f58
################# 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 2023.05.001, 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=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2023.05.001
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.2
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/01
job id: 48425992
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/02
job id: 48425993
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/03
job id: 48425994
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/04
job id: 48425995
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/05
job id: 48425996
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/06
job id: 48425997
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/07
job id: 48425998
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/08
job id: 48425999
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/09
job id: 48426000
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/10
job id: 48426001
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/11
job id: 48426002
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/12
job id: 48426003
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/13
job id: 48426004
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/14
job id: 48426005
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/15
job id: 48426006
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/16
job id: 48426007
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/17
job id: 48426008
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/18
job id: 48426009
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/19
job id: 48426010
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/20
job id: 48426011
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/21
job id: 48426012
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/22
job id: 48426013
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/23
job id: 48426014
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/24
job id: 48426016
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/25
job id: 48426017
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/26
job id: 48426018
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/27
job id: 48426019
--- 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 0.021 0.034 133.253 133.254
farming_run 1 2.0 132.690 132.695 133.224 133.228
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.488286E+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 0.006 0.024 114.740 114.741
qs_energies 1 2.0 0.000 0.000 114.542 114.544
mp2_main 1 3.0 0.000 0.000 112.465 112.468
mp2_gpw_main 1 4.0 0.028 0.034 111.456 111.459
mp2_ri_gpw_compute_in 1 5.0 0.172 0.175 92.490 92.818
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 54.586 54.915
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.165 40.992 46.042
get_2c_integrals 1 6.0 0.008 0.009 36.986 37.729
integrate_v_rspace 273 8.0 0.441 0.456 24.588 29.498
pw_transfer 6555 10.6 0.375 0.393 26.836 27.302
fft_wrap_pw1pw2 5465 11.4 0.044 0.048 25.442 26.031
grid_integrate_task_list 273 9.0 20.495 25.867 20.495 25.867
fft_wrap_pw1pw2_100 2178 12.4 1.138 1.297 23.002 23.575
compute_2c_integrals 1 7.0 0.003 0.003 19.405 19.406
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.661 19.137
mp2_eri_2c_integrate_gpw 1 9.0 2.366 2.421 18.658 19.136
rpa_ri_compute_en 1 5.0 0.020 0.022 18.848 19.000
cp_fm_cholesky_decompose 12 8.2 17.523 18.252 17.523 18.252
cholesky_decomp 1 7.0 0.000 0.000 16.428 17.153
fft3d_s 5443 13.4 16.139 16.498 16.163 16.520
ao_to_mo_and_store_B_mult_1 272 7.0 10.762 15.306 10.762 15.306
calculate_wavefunction 272 8.0 5.406 5.540 12.312 13.030
rpa_num_int 1 6.0 0.000 0.002 10.635 10.645
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.616 10.620
calc_mat_Q 8 8.0 0.000 0.000 9.473 9.550
contract_S_to_Q 8 9.0 0.000 0.000 8.894 8.973
calc_potential_gpw 544 9.5 0.005 0.006 8.256 8.736
parallel_gemm_fm 14 9.1 0.000 0.000 8.478 8.581
parallel_gemm_fm_cosma 14 10.1 8.478 8.581 8.478 8.581
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.117 8.416
potential_pw2rs 545 10.0 0.108 0.110 7.498 8.067
create_integ_mat 1 6.0 0.013 0.028 7.792 7.802
collocate_single_gaussian 272 10.0 0.039 0.041 7.368 7.663
array2fm 1 7.0 0.000 0.000 6.753 7.268
pw_scatter_s 2720 13.7 4.345 4.462 4.345 4.462
pw_gather_s 2722 13.2 3.423 3.730 3.423 3.730
array2fm_buffer_send 1 8.0 3.015 3.127 3.015 3.127
pw_poisson_solve 545 10.5 1.119 1.196 2.220 2.458
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=111.456439, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2805.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 0.032 0.048 420.107 420.108
farming_run 1 2.0 419.262 419.268 420.063 420.072
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.238204E+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 14116942.
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 0.034 217.235 217.236
qs_energies 1 2.0 0.001 0.003 216.954 216.963
scf_env_do_scf 1 3.0 0.000 0.000 114.981 114.981
qs_ks_update_qs_env 5 5.0 0.000 0.000 114.093 114.100
rebuild_ks_matrix 4 6.0 0.000 0.000 114.092 114.099
qs_ks_build_kohn_sham_matrix 4 7.0 0.055 0.062 114.092 114.099
hfx_ks_matrix 4 8.0 0.001 0.001 113.745 113.749
integrate_four_center 4 9.0 0.153 0.465 113.744 113.748
integrate_four_center_main 4 10.0 0.162 0.605 102.120 105.208
integrate_four_center_bin 266 11.0 101.958 104.887 101.958 104.887
mp2_main 1 3.0 0.001 0.013 101.683 101.693
mp2_gpw_main 1 4.0 0.049 0.073 100.834 100.848
init_scf_loop 1 4.0 0.000 0.000 96.433 96.433
mp2_ri_gpw_compute_in 1 5.0 0.064 0.067 73.944 74.963
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.003 53.789 54.807
mp2_eri_3c_integrate_gpw 91 7.0 0.144 0.161 41.696 46.496
integrate_v_rspace 95 8.0 0.395 0.564 28.162 32.781
pw_transfer 2240 10.6 0.145 0.176 29.505 29.892
fft_wrap_pw1pw2 1868 11.4 0.017 0.021 28.511 28.937
mp2_ri_gpw_compute_en 1 5.0 0.061 0.072 26.722 28.719
grid_integrate_task_list 95 9.0 23.586 28.407 23.586 28.407
ao_to_mo_and_store_B_mult_1 91 7.0 10.414 27.586 10.414 27.586
fft_wrap_pw1pw2_100 730 12.4 1.289 1.498 26.262 26.673
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.817 1.889 25.024 25.034
get_2c_integrals 1 6.0 0.002 0.019 20.063 20.091
compute_2c_integrals 1 7.0 0.003 0.004 19.046 19.047
compute_2c_integrals_loop_lm 1 8.0 0.001 0.001 18.628 18.915
mp2_eri_2c_integrate_gpw 1 9.0 1.736 1.867 18.627 18.914
fft3d_s 1823 13.4 18.505 18.850 18.519 18.864
scf_env_do_scf_inner_loop 4 4.0 0.000 0.001 18.546 18.546
calculate_wavefunction 91 8.0 2.007 2.033 9.622 9.818
mp2_ri_gpw_compute_en_expansio 172 7.0 0.558 0.579 8.814 9.393
potential_pw2rs 186 10.0 0.034 0.035 8.423 8.918
local_gemm 172 8.0 8.256 8.830 8.256 8.830
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.157 8.570
mp2_ri_gpw_compute_en_comm 22 7.0 0.501 0.514 8.007 8.434
calc_potential_gpw 182 9.5 0.002 0.002 7.905 8.076
collocate_single_gaussian 91 10.0 0.017 0.029 7.750 8.019
mp_sync 37 10.5 3.599 7.234 3.599 7.234
integrate_four_center_load 4 10.0 0.000 0.000 6.758 6.762
hfx_load_balance 1 11.0 0.000 0.000 6.758 6.762
mp_sendrecv_dm3 2068 8.0 6.037 6.440 6.037 6.440
mp2_ri_gpw_compute_en_ener 172 7.0 6.342 6.411 6.342 6.411
pw_gather_s 912 13.2 4.464 5.019 4.464 5.019
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=100.828497, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1498.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 452.026368E+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 503918.
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 0.027 0.096 54.865 54.870
qs_mol_dyn_low 1 2.0 0.004 0.006 54.529 54.536
qs_forces 11 3.9 0.002 0.003 54.468 54.469
qs_energies 11 4.9 0.023 0.098 52.989 53.014
scf_env_do_scf 11 5.9 0.001 0.001 46.648 46.649
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 44.638 44.639
dbcsr_multiply_generic 2286 12.5 0.094 0.099 34.631 35.063
qs_scf_new_mos 108 7.5 0.000 0.001 34.773 35.048
qs_scf_loop_do_ot 108 8.5 0.000 0.001 34.773 35.048
ot_scf_mini 108 9.5 0.002 0.002 33.093 33.293
multiply_cannon 2286 13.5 0.189 0.195 26.847 28.498
multiply_cannon_loop 2286 14.5 1.827 1.938 26.203 27.878
velocity_verlet 10 3.0 0.001 0.002 26.153 26.154
ot_mini 108 10.5 0.001 0.001 20.124 20.368
qs_ot_get_derivative 108 11.5 0.001 0.001 17.062 17.246
mp_waitall_1 245248 16.5 8.822 15.131 8.822 15.131
multiply_cannon_metrocomm3 54864 15.5 0.073 0.079 6.376 13.341
multiply_cannon_multrec 54864 15.5 3.649 5.714 7.769 11.220
qs_ot_get_p 119 10.4 0.002 0.005 8.190 8.492
rebuild_ks_matrix 119 8.3 0.000 0.000 7.723 7.848
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.012 7.723 7.848
mp_sum_l 7287 12.8 5.973 7.654 5.973 7.654
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.799 6.918
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.708 6.165
multiply_cannon_sync_h2d 54864 15.5 5.123 6.087 5.123 6.087
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.805 5.943
dbcsr_mm_accdrv_process 76910 16.1 1.813 2.870 4.034 5.779
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.309 5.382
init_scf_run 11 5.9 0.000 0.001 4.955 4.956
scf_env_initial_rho_setup 11 6.9 0.000 0.001 4.955 4.956
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.462 4.462
sum_up_and_integrate 119 10.3 0.012 0.014 4.348 4.357
integrate_v_rspace 119 11.3 0.002 0.003 4.336 4.345
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.280 4.281
cp_fm_redistribute_end 50 14.0 2.184 4.258 2.189 4.260
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.085 4.230
calculate_rho_elec 119 8.7 0.011 0.017 4.084 4.230
cp_fm_diag_elpa_base 50 14.0 2.066 4.154 2.070 4.162
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.949 3.145
apply_single 119 13.6 0.000 0.000 2.949 3.145
calculate_dm_sparse 119 9.5 0.000 0.000 2.953 3.133
multiply_cannon_metrocomm1 54864 15.5 0.058 0.064 1.863 3.094
acc_transpose_blocks 54864 15.5 0.231 0.262 2.290 2.850
jit_kernel_multiply 13 15.8 2.156 2.822 2.156 2.822
ot_diis_step 108 11.5 0.006 0.006 2.773 2.774
calculate_first_density_matrix 1 7.0 0.001 0.004 2.737 2.741
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.568 2.622
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.426 2.428
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.128 2.182
density_rs2pw 119 9.7 0.004 0.004 2.066 2.145
wfi_extrapolate 11 7.9 0.001 0.001 2.139 2.139
grid_integrate_task_list 119 12.3 2.001 2.103 2.001 2.103
mp_sum_d 4135 12.0 1.443 2.023 1.443 2.023
init_scf_loop 11 6.9 0.001 0.003 1.995 1.997
potential_pw2rs 119 12.3 0.004 0.004 1.738 1.749
pw_transfer 1439 11.6 0.051 0.057 1.582 1.653
make_m2s 4572 13.5 0.054 0.056 1.563 1.614
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 1.507 1.579
make_images 4572 14.5 0.133 0.139 1.480 1.531
mp_alltoall_d11v 2130 13.8 1.316 1.481 1.316 1.481
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.001 1.441 1.476
acc_transpose_blocks_sync 164592 16.5 1.197 1.432 1.197 1.432
transfer_rs2pw 487 10.6 0.005 0.006 1.327 1.430
grid_collocate_task_list 119 9.7 1.355 1.422 1.355 1.422
mp_waitany 12084 13.8 1.265 1.370 1.265 1.370
fft3d_ps 1201 14.6 0.369 0.475 1.221 1.291
transfer_pw2rs 487 13.2 0.006 0.007 1.274 1.280
fft_wrap_pw1pw2_140 487 13.2 0.141 0.155 1.169 1.241
dbcsr_dot_sd 1205 11.9 0.048 0.059 0.808 1.191
acc_transpose_blocks_kernels 54864 16.5 0.243 0.377 0.816 1.116
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=54.870000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=431.272727, yerr=1.052349
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 487.698432E+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 1798723.
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 0.015 0.035 38.989 38.990
qs_mol_dyn_low 1 2.0 0.003 0.004 38.770 38.778
qs_forces 11 3.9 0.003 0.008 38.708 38.709
qs_energies 11 4.9 0.001 0.004 37.008 37.013
scf_env_do_scf 11 5.9 0.001 0.001 31.478 31.479
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 29.062 29.064
dbcsr_multiply_generic 2286 12.5 0.100 0.106 22.301 22.664
qs_scf_new_mos 108 7.5 0.001 0.001 20.558 20.791
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.557 20.790
ot_scf_mini 108 9.5 0.002 0.003 19.635 19.808
multiply_cannon 2286 13.5 0.210 0.219 17.258 18.809
velocity_verlet 10 3.0 0.002 0.002 18.174 18.175
multiply_cannon_loop 2286 14.5 1.196 1.258 16.112 17.574
ot_mini 108 10.5 0.001 0.001 12.165 12.400
mp_waitall_1 200699 16.5 5.626 10.897 5.626 10.897
qs_ot_get_derivative 108 11.5 0.001 0.001 9.812 9.982
multiply_cannon_metrocomm3 27432 15.5 0.071 0.075 4.166 9.686
multiply_cannon_multrec 27432 15.5 1.821 4.235 6.307 9.240
rebuild_ks_matrix 119 8.3 0.000 0.000 6.742 6.884
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.014 6.742 6.884
dbcsr_mm_accdrv_process 47894 16.0 3.716 5.884 4.405 6.539
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.900 6.030
qs_ot_get_p 119 10.4 0.002 0.009 4.586 4.808
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.768 4.616
init_scf_run 11 5.9 0.000 0.001 4.196 4.196
scf_env_initial_rho_setup 11 6.9 0.000 0.000 4.196 4.196
mp_sum_l 7287 12.8 2.013 4.068 2.013 4.068
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.994 4.048
apply_single 119 13.6 0.000 0.000 2.994 4.048
sum_up_and_integrate 119 10.3 0.025 0.028 3.675 3.689
integrate_v_rspace 119 11.3 0.002 0.004 3.651 3.666
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.544 3.577
calculate_rho_elec 119 8.7 0.021 0.024 3.544 3.576
qs_ot_p2m_diag 50 11.0 0.009 0.014 2.941 2.959
make_m2s 4572 13.5 0.053 0.055 2.578 2.917
make_images 4572 14.5 0.205 0.244 2.489 2.831
calculate_first_density_matrix 1 7.0 0.000 0.002 2.705 2.706
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.544 2.544
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.318 2.410
init_scf_loop 11 6.9 0.001 0.007 2.397 2.400
ot_diis_step 108 11.5 0.011 0.011 2.306 2.306
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.238 2.238
calculate_dm_sparse 119 9.5 0.001 0.001 2.164 2.237
cp_fm_redistribute_end 50 14.0 1.135 2.215 1.138 2.218
multiply_cannon_sync_h2d 27432 15.5 1.674 2.175 1.674 2.175
cp_fm_diag_elpa_base 50 14.0 1.049 2.135 1.076 2.168
acc_transpose_blocks 27432 15.5 0.115 0.120 1.726 2.119
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.088 2.090
grid_integrate_task_list 119 12.3 1.841 1.946 1.841 1.946
density_rs2pw 119 9.7 0.004 0.004 1.848 1.941
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 1.895 1.934
pw_transfer 1439 11.6 0.065 0.068 1.826 1.861
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.735 1.772
make_images_data 4572 15.5 0.047 0.054 1.230 1.581
prepare_preconditioner 11 7.9 0.000 0.000 1.496 1.521
make_preconditioner 11 8.9 0.000 0.001 1.495 1.520
jit_kernel_multiply 7 16.5 0.630 1.459 0.630 1.459
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.398 1.455
hybrid_alltoall_any 4725 16.4 0.054 0.116 1.086 1.453
wfi_extrapolate 11 7.9 0.001 0.001 1.434 1.434
fft_wrap_pw1pw2_140 487 13.2 0.159 0.168 1.395 1.430
potential_pw2rs 119 12.3 0.006 0.006 1.413 1.426
fft3d_ps 1201 14.6 0.516 0.570 1.372 1.404
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.349 1.357
grid_collocate_task_list 119 9.7 1.284 1.337 1.284 1.337
mp_alltoall_d11v 2130 13.8 1.178 1.286 1.178 1.286
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.193 1.239
transfer_rs2pw 487 10.6 0.005 0.005 1.070 1.179
mp_sum_d 4135 12.0 0.605 1.105 0.605 1.105
qs_energies_init_hamiltonians 11 5.9 0.013 0.047 1.066 1.067
acc_transpose_blocks_kernels 27432 16.5 0.188 0.276 0.762 1.028
mp_allgather_i34 2286 14.5 0.567 0.963 0.567 0.963
acc_transpose_blocks_sync 82296 16.5 0.822 0.945 0.822 0.945
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.901 0.914
transfer_pw2rs 487 13.2 0.005 0.005 0.864 0.868
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=38.990000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.090909, yerr=1.564059
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 519.639040E+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 1607811.
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 0.023 0.040 31.920 31.922
qs_mol_dyn_low 1 2.0 0.003 0.004 31.607 31.615
qs_forces 11 3.9 0.004 0.008 31.544 31.550
qs_energies 11 4.9 0.003 0.006 29.994 30.001
scf_env_do_scf 11 5.9 0.001 0.002 25.232 25.232
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 22.826 22.826
dbcsr_multiply_generic 2286 12.5 0.097 0.101 16.852 16.943
qs_scf_new_mos 108 7.5 0.001 0.001 15.273 15.287
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.272 15.287
velocity_verlet 10 3.0 0.002 0.002 14.983 14.984
ot_scf_mini 108 9.5 0.002 0.003 14.514 14.527
multiply_cannon 2286 13.5 0.194 0.201 13.656 14.401
multiply_cannon_loop 2286 14.5 0.871 0.926 12.902 13.655
ot_mini 108 10.5 0.001 0.001 9.005 9.019
qs_ot_get_derivative 108 11.5 0.001 0.001 7.510 7.522
multiply_cannon_multrec 18288 15.5 1.867 2.917 6.939 7.239
rebuild_ks_matrix 119 8.3 0.000 0.000 5.925 5.941
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.015 5.925 5.941
dbcsr_mm_accdrv_process 38222 16.0 4.891 5.755 4.980 5.820
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.224 5.240
mp_waitall_1 158411 16.6 2.544 3.622 2.544 3.622
init_scf_run 11 5.9 0.000 0.001 3.533 3.533
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.532 3.533
sum_up_and_integrate 119 10.3 0.030 0.031 3.510 3.515
integrate_v_rspace 119 11.3 0.003 0.004 3.479 3.488
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.827 3.420
qs_ot_get_p 119 10.4 0.001 0.001 3.280 3.309
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.194 3.200
calculate_rho_elec 119 8.7 0.031 0.031 3.193 3.199
multiply_cannon_metrocomm3 18288 15.5 0.047 0.049 1.434 2.397
init_scf_loop 11 6.9 0.002 0.010 2.389 2.390
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.003 2.350
apply_single 119 13.6 0.000 0.000 2.003 2.350
calculate_first_density_matrix 1 7.0 0.001 0.002 2.322 2.324
qs_ot_p2m_diag 50 11.0 0.017 0.027 2.193 2.199
grid_integrate_task_list 119 12.3 1.805 1.916 1.805 1.916
make_m2s 4572 13.5 0.046 0.048 1.779 1.911
cp_dbcsr_syevd 50 12.0 0.003 0.005 1.892 1.898
density_rs2pw 119 9.7 0.004 0.005 1.764 1.869
calculate_dm_sparse 119 9.5 0.001 0.001 1.844 1.857
make_images 4572 14.5 0.192 0.205 1.693 1.823
pw_transfer 1439 11.6 0.065 0.068 1.800 1.808
acc_transpose_blocks 18288 15.5 0.081 0.083 1.679 1.804
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.726 1.729
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.708 1.718
qs_ot_get_derivative_diag 49 12.0 0.001 0.002 1.703 1.712
cp_fm_diag_elpa 50 13.0 0.000 0.001 1.645 1.655
cp_fm_diag_elpa_base 50 14.0 1.620 1.633 1.642 1.652
prepare_preconditioner 11 7.9 0.000 0.000 1.613 1.616
make_preconditioner 11 8.9 0.000 0.001 1.613 1.616
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 1.559 1.567
make_full_inverse_cholesky 11 9.9 0.000 0.001 1.473 1.554
ot_diis_step 108 11.5 0.011 0.011 1.479 1.479
mp_sum_l 7287 12.8 1.033 1.457 1.033 1.457
fft_wrap_pw1pw2_140 487 13.2 0.212 0.219 1.388 1.400
grid_collocate_task_list 119 9.7 1.233 1.301 1.233 1.301
potential_pw2rs 119 12.3 0.007 0.009 1.291 1.297
fft3d_ps 1201 14.6 0.527 0.543 1.276 1.284
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.182 1.186
wfi_extrapolate 11 7.9 0.001 0.001 1.151 1.151
multiply_cannon_sync_h2d 18288 15.5 1.025 1.144 1.025 1.144
transfer_rs2pw 487 10.6 0.005 0.005 0.964 1.074
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 0.996 1.002
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.939 0.957
make_images_data 4572 15.5 0.048 0.052 0.754 0.904
acc_transpose_blocks_sync 54864 16.5 0.772 0.886 0.772 0.886
hybrid_alltoall_any 4725 16.4 0.058 0.116 0.656 0.826
acc_transpose_blocks_kernels 18288 16.5 0.217 0.224 0.806 0.819
transfer_pw2rs 487 13.2 0.004 0.004 0.779 0.781
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.778 0.780
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.680 0.750
mp_alltoall_d11v 2130 13.8 0.652 0.749 0.652 0.749
jit_kernel_multiply 4 15.0 0.029 0.747 0.029 0.747
cp_fm_cholesky_invert 11 10.9 0.684 0.687 0.684 0.687
mp_alltoall_z22v 1201 16.6 0.600 0.668 0.600 0.668
mp_waitany 9880 13.7 0.546 0.657 0.546 0.657
transfer_rs2pw_140 130 11.5 0.118 0.122 0.537 0.652
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=31.922000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=495.636364, yerr=0.771389
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 559.165440E+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 0.014 0.031 35.917 35.919
qs_mol_dyn_low 1 2.0 0.003 0.004 35.732 35.740
qs_forces 11 3.9 0.002 0.003 35.673 35.674
qs_energies 11 4.9 0.001 0.004 33.974 33.978
scf_env_do_scf 11 5.9 0.001 0.001 28.823 28.824
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 25.501 25.502
dbcsr_multiply_generic 2286 12.5 0.099 0.104 19.704 19.844
velocity_verlet 10 3.0 0.002 0.002 18.174 18.175
qs_scf_new_mos 108 7.5 0.001 0.001 17.559 17.612
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.558 17.612
ot_scf_mini 108 9.5 0.002 0.003 16.533 16.589
multiply_cannon 2286 13.5 0.218 0.232 16.067 16.448
multiply_cannon_loop 2286 14.5 1.523 1.629 15.128 15.618
ot_mini 108 10.5 0.001 0.001 10.200 10.267
multiply_cannon_multrec 27432 15.5 2.449 3.139 8.996 9.291
qs_ot_get_derivative 108 11.5 0.001 0.001 8.341 8.397
dbcsr_mm_accdrv_process 47916 15.9 6.004 7.692 6.445 7.782
rebuild_ks_matrix 119 8.3 0.000 0.000 6.257 6.303
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.257 6.303
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.542 5.583
init_scf_run 11 5.9 0.000 0.001 3.781 3.782
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.781 3.781
qs_ot_get_p 119 10.4 0.001 0.001 3.510 3.583
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.105 3.501
sum_up_and_integrate 119 10.3 0.035 0.037 3.450 3.459
integrate_v_rspace 119 11.3 0.003 0.003 3.415 3.424
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.268 3.304
calculate_rho_elec 119 8.7 0.040 0.046 3.268 3.303
init_scf_loop 11 6.9 0.001 0.004 3.303 3.303
acc_transpose_blocks 27432 15.5 0.120 0.125 2.386 2.524
prepare_preconditioner 11 7.9 0.000 0.002 2.476 2.485
make_preconditioner 11 8.9 0.000 0.002 2.476 2.485
calculate_first_density_matrix 1 7.0 0.000 0.002 2.412 2.414
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.085 2.406
make_m2s 4572 13.5 0.055 0.057 2.239 2.345
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.302 2.332
mp_waitall_1 137007 16.6 1.714 2.271 1.714 2.271
make_images 4572 14.5 0.274 0.335 2.129 2.234
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.103 2.230
apply_single 119 13.6 0.000 0.000 2.102 2.230
calculate_dm_sparse 119 9.5 0.000 0.000 2.167 2.219
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.126 2.138
grid_integrate_task_list 119 12.3 1.820 1.922 1.820 1.922
pw_transfer 1439 11.6 0.064 0.069 1.882 1.915
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.791 1.827
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.821 1.823
ot_diis_step 108 11.5 0.012 0.012 1.819 1.820
density_rs2pw 119 9.7 0.004 0.004 1.726 1.816
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.727 1.727
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 1.652 1.672
acc_transpose_blocks_sync 82296 16.5 1.417 1.552 1.417 1.552
fft_wrap_pw1pw2_140 487 13.2 0.246 0.259 1.485 1.523
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.498 1.507
cp_fm_diag_elpa_base 50 14.0 1.464 1.481 1.496 1.505
grid_collocate_task_list 119 9.7 1.249 1.336 1.249 1.336
fft3d_ps 1201 14.6 0.553 0.605 1.310 1.335
multiply_cannon_metrocomm3 27432 15.5 0.040 0.042 0.759 1.332
mp_sum_l 7287 12.8 0.970 1.320 0.970 1.320
wfi_extrapolate 11 7.9 0.001 0.001 1.318 1.318
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.276 1.298
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.268 1.278
potential_pw2rs 119 12.3 0.008 0.010 1.238 1.241
cp_fm_upper_to_full 72 14.2 0.802 1.139 0.802 1.139
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.108 1.109
jit_kernel_multiply 6 16.6 0.375 1.069 0.375 1.069
dbcsr_complete_redistribute 329 12.2 0.120 0.151 0.747 1.015
transfer_rs2pw 487 10.6 0.005 0.005 0.890 0.985
make_images_data 4572 15.5 0.048 0.052 0.834 0.939
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.802 0.878
mp_alltoall_d11v 2130 13.8 0.728 0.874 0.728 0.874
hybrid_alltoall_any 4725 16.4 0.065 0.155 0.716 0.864
acc_transpose_blocks_kernels 27432 16.5 0.269 0.277 0.819 0.826
copy_fm_to_dbcsr 176 11.2 0.001 0.002 0.558 0.818
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.804 0.808
cp_fm_cholesky_invert 11 10.9 0.718 0.721 0.718 0.721
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=35.919000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=531.909091, yerr=3.342687
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 620.916736E+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 0.023 0.048 28.651 28.659
qs_mol_dyn_low 1 2.0 0.003 0.003 28.474 28.482
qs_forces 11 3.9 0.002 0.002 28.407 28.408
qs_energies 11 4.9 0.001 0.001 26.690 26.692
scf_env_do_scf 11 5.9 0.001 0.002 21.969 21.969
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 19.501 19.502
velocity_verlet 10 3.0 0.002 0.002 14.733 14.736
dbcsr_multiply_generic 2286 12.5 0.093 0.098 13.099 13.167
qs_scf_new_mos 108 7.5 0.001 0.001 11.896 11.920
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.895 11.919
ot_scf_mini 108 9.5 0.002 0.002 11.187 11.211
multiply_cannon 2286 13.5 0.223 0.229 10.419 10.840
multiply_cannon_loop 2286 14.5 0.647 0.668 9.484 9.647
ot_mini 108 10.5 0.001 0.001 6.383 6.412
multiply_cannon_multrec 9144 15.5 1.652 1.926 5.916 6.163
rebuild_ks_matrix 119 8.3 0.000 0.000 5.801 5.827
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.800 5.827
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.152 5.176
qs_ot_get_derivative 108 11.5 0.001 0.001 5.097 5.120
dbcsr_mm_accdrv_process 12550 15.8 3.630 4.173 4.154 4.216
sum_up_and_integrate 119 10.3 0.037 0.041 3.470 3.474
integrate_v_rspace 119 11.3 0.003 0.003 3.432 3.437
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.295 3.300
calculate_rho_elec 119 8.7 0.060 0.061 3.294 3.300
init_scf_run 11 5.9 0.000 0.001 3.280 3.280
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.280 3.280
qs_ot_get_p 119 10.4 0.001 0.001 2.905 2.940
init_scf_loop 11 6.9 0.000 0.000 2.447 2.448
calculate_first_density_matrix 1 7.0 0.000 0.000 2.114 2.115
make_m2s 4572 13.5 0.035 0.036 1.846 2.011
pw_transfer 1439 11.6 0.065 0.068 1.975 1.987
grid_integrate_task_list 119 12.3 1.866 1.941 1.866 1.941
make_images 4572 14.5 0.270 0.300 1.756 1.918
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.882 1.896
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.881 1.883
mp_waitall_1 115863 16.7 1.410 1.855 1.410 1.855
density_rs2pw 119 9.7 0.004 0.004 1.701 1.787
calculate_dm_sparse 119 9.5 0.000 0.000 1.747 1.764
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.730 1.755
prepare_preconditioner 11 7.9 0.000 0.000 1.716 1.722
make_preconditioner 11 8.9 0.000 0.000 1.716 1.722
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.606 1.637
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.633 1.634
fft_wrap_pw1pw2_140 487 13.2 0.323 0.331 1.550 1.563
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.537 1.538
acc_transpose_blocks 9144 15.5 0.042 0.043 1.431 1.464
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.372 1.383
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.372 1.382
grid_collocate_task_list 119 9.7 1.298 1.369 1.298 1.369
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.349 1.356
cp_fm_diag_elpa_base 50 14.0 1.321 1.338 1.347 1.355
fft3d_ps 1201 14.6 0.559 0.571 1.296 1.305
ot_diis_step 108 11.5 0.013 0.013 1.273 1.273
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.225 1.272
apply_single 119 13.6 0.000 0.000 1.224 1.272
potential_pw2rs 119 12.3 0.010 0.011 1.247 1.250
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.223 1.224
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.139 1.143
wfi_extrapolate 11 7.9 0.001 0.001 1.114 1.114
jit_kernel_multiply 6 15.8 0.485 1.059 0.485 1.059
hybrid_alltoall_any 4725 16.4 0.065 0.176 0.782 1.027
make_images_data 4572 15.5 0.042 0.046 0.795 0.990
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.876 0.928
transfer_rs2pw 487 10.6 0.004 0.004 0.786 0.877
mp_alltoall_d11v 2130 13.8 0.774 0.863 0.774 0.863
cp_fm_cholesky_invert 11 10.9 0.852 0.855 0.852 0.855
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.833 0.841
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.770 0.773
acc_transpose_blocks_sync 27432 16.5 0.724 0.753 0.724 0.753
multiply_cannon_metrocomm3 9144 15.5 0.020 0.021 0.360 0.747
qs_env_update_s_mstruct 11 6.9 0.001 0.001 0.673 0.724
mp_allgather_i34 2286 14.5 0.246 0.678 0.246 0.678
transfer_pw2rs 487 13.2 0.003 0.004 0.671 0.673
acc_transpose_blocks_kernels 9144 16.5 0.117 0.119 0.650 0.653
mp_alltoall_z22v 1201 16.6 0.610 0.643 0.610 0.643
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=28.659000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=585.363636, yerr=7.176638
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 774.414336E+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 0.014 0.033 42.389 42.390
qs_mol_dyn_low 1 2.0 0.018 0.029 42.155 42.162
qs_forces 11 3.9 0.011 0.043 42.079 42.088
qs_energies 11 4.9 0.001 0.001 40.088 40.109
scf_env_do_scf 11 5.9 0.001 0.002 34.274 34.275
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 26.455 26.457
velocity_verlet 10 3.0 0.002 0.002 23.833 23.838
dbcsr_multiply_generic 2286 12.5 0.100 0.103 18.823 19.062
qs_scf_new_mos 108 7.5 0.001 0.001 17.108 17.210
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.107 17.210
ot_scf_mini 108 9.5 0.002 0.002 15.951 16.052
multiply_cannon 2286 13.5 0.292 0.296 14.913 15.855
multiply_cannon_loop 2286 14.5 0.858 0.879 13.664 14.636
ot_mini 108 10.5 0.001 0.001 9.685 9.800
multiply_cannon_multrec 9144 15.5 3.380 4.618 8.738 8.845
init_scf_loop 11 6.9 0.000 0.000 7.793 7.797
qs_ot_get_derivative 108 11.5 0.001 0.001 7.654 7.757
rebuild_ks_matrix 119 8.3 0.000 0.000 7.096 7.237
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.096 7.236
prepare_preconditioner 11 7.9 0.000 0.000 6.806 6.819
make_preconditioner 11 8.9 0.000 0.000 6.806 6.819
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.363 6.701
dbcsr_mm_accdrv_process 12550 15.8 4.540 6.457 5.222 6.527
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.397 6.524
cp_fm_upper_to_full 72 14.2 3.219 4.608 3.219 4.608
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.994 4.007
calculate_rho_elec 119 8.7 0.118 0.121 3.993 4.006
sum_up_and_integrate 119 10.3 0.064 0.065 3.806 3.812
integrate_v_rspace 119 11.3 0.004 0.004 3.741 3.748
init_scf_run 11 5.9 0.000 0.001 3.681 3.681
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.681 3.681
qs_ot_get_p 119 10.4 0.001 0.001 3.375 3.511
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.809 3.208
mp_waitall_1 94719 16.7 2.194 3.197 2.194 3.197
dbcsr_complete_redistribute 329 12.2 0.327 0.355 2.042 2.904
pw_transfer 1439 11.6 0.068 0.069 2.572 2.577
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.733 2.574
make_m2s 4572 13.5 0.039 0.039 2.380 2.524
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.177 2.494
apply_single 119 13.6 0.000 0.000 2.177 2.494
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.474 2.478
make_images 4572 14.5 0.356 0.390 2.259 2.403
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.437 2.269
calculate_dm_sparse 119 9.5 0.000 0.000 2.230 2.244
calculate_first_density_matrix 1 7.0 0.000 0.000 2.223 2.224
mp_alltoall_i22 627 13.8 1.413 2.214 1.413 2.214
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.287 2.205
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.048 2.098
fft_wrap_pw1pw2_140 487 13.2 0.576 0.581 2.086 2.091
grid_integrate_task_list 119 12.3 2.077 2.091 2.077 2.091
density_rs2pw 119 9.7 0.004 0.004 2.025 2.043
ot_diis_step 108 11.5 0.014 0.014 2.006 2.006
qs_ot_p2m_diag 50 11.0 0.043 0.044 1.961 1.963
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.928 1.928
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.853 1.870
acc_transpose_blocks 9144 15.5 0.044 0.045 1.806 1.838
mp_sum_l 7287 12.8 0.983 1.715 0.983 1.715
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.612 1.654
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.606 1.607
fft3d_ps 1201 14.6 0.594 0.605 1.570 1.572
grid_collocate_task_list 119 9.7 1.520 1.547 1.520 1.547
cp_fm_cholesky_invert 11 10.9 1.396 1.400 1.396 1.400
wfi_extrapolate 11 7.9 0.001 0.001 1.396 1.396
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.310 1.330
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.329 1.329
cp_fm_diag_elpa_base 50 14.0 1.182 1.237 1.327 1.328
potential_pw2rs 119 12.3 0.014 0.015 1.272 1.273
hybrid_alltoall_any 4725 16.4 0.090 0.150 1.037 1.246
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.237 1.244
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.184 1.210
mp_alltoall_d11v 2130 13.8 1.139 1.160 1.139 1.160
make_images_data 4572 15.5 0.046 0.049 0.974 1.155
acc_transpose_blocks_sync 27432 16.5 1.095 1.125 1.095 1.125
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.991 1.029
qs_create_task_list 11 7.9 0.044 0.044 0.972 0.983
generate_qs_task_list 11 8.9 0.367 0.386 0.928 0.940
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.911 0.924
mp_alltoall_z22v 1201 16.6 0.838 0.859 0.838 0.859
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=42.390000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=722.000000, yerr=17.246871
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 501.874688E+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 2077478.
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 0.029 0.079 82.181 82.184
qs_mol_dyn_low 1 2.0 0.014 0.030 81.762 81.772
qs_forces 11 3.9 0.004 0.012 81.662 81.671
qs_energies 11 4.9 0.001 0.005 78.813 78.831
scf_env_do_scf 11 5.9 0.000 0.001 69.720 69.722
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 64.297 64.297
dbcsr_multiply_generic 2055 12.4 0.107 0.116 51.364 51.621
qs_scf_new_mos 99 7.5 0.000 0.001 47.422 47.543
qs_scf_loop_do_ot 99 8.5 0.001 0.001 47.422 47.543
ot_scf_mini 99 9.5 0.002 0.003 45.017 45.120
multiply_cannon 2055 13.4 0.185 0.189 42.823 43.678
velocity_verlet 10 3.0 0.001 0.002 43.394 43.395
multiply_cannon_loop 2055 14.4 1.806 1.854 41.883 42.695
ot_mini 99 10.5 0.001 0.001 26.639 26.760
qs_ot_get_derivative 99 11.5 0.001 0.001 19.800 19.909
multiply_cannon_multrec 49320 15.4 11.367 11.979 17.358 17.880
rebuild_ks_matrix 110 8.3 0.000 0.000 14.480 14.562
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.027 14.480 14.561
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.708 12.782
mp_waitall_1 220248 16.4 11.045 11.760 11.045 11.760
multiply_cannon_sync_h2d 49320 15.4 9.561 10.102 9.561 10.102
qs_ot_get_p 110 10.4 0.001 0.001 9.738 9.881
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.678 8.217
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.234 7.819
apply_single 110 13.6 0.000 0.000 7.234 7.819
multiply_cannon_metrocomm3 49320 15.4 0.083 0.087 6.819 7.729
sum_up_and_integrate 110 10.3 0.036 0.043 6.968 6.981
integrate_v_rspace 110 11.3 0.003 0.004 6.932 6.953
init_scf_run 11 5.9 0.000 0.001 6.851 6.851
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.851 6.851
ot_diis_step 99 11.5 0.005 0.006 6.594 6.594
qs_ot_p2m_diag 48 11.0 0.013 0.019 6.399 6.423
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.218 6.347
calculate_rho_elec 110 8.6 0.021 0.024 6.218 6.347
dbcsr_mm_accdrv_process 87628 16.1 3.043 3.130 5.861 6.131
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.587 5.588
init_scf_loop 11 6.9 0.001 0.002 5.397 5.398
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.282 5.356
mp_sum_l 6594 12.7 4.313 5.237 4.313 5.237
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.032 5.060
cp_fm_diag_elpa_base 48 14.0 5.019 5.048 5.030 5.058
wfi_extrapolate 11 7.9 0.001 0.001 4.004 4.004
make_m2s 4110 13.4 0.062 0.067 3.822 3.922
calculate_dm_sparse 110 9.5 0.001 0.001 3.774 3.881
make_images 4110 14.4 0.177 0.188 3.726 3.829
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 3.625 3.630
multiply_cannon_metrocomm1 49320 15.4 0.068 0.070 2.427 3.432
density_rs2pw 110 9.6 0.004 0.005 3.266 3.410
grid_integrate_task_list 110 12.3 3.261 3.408 3.261 3.408
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.344 3.388
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 3.216 3.261
prepare_preconditioner 11 7.9 0.000 0.000 3.183 3.209
make_preconditioner 11 8.9 0.000 0.001 3.182 3.209
pw_transfer 1331 11.6 0.055 0.066 3.015 3.088
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.965 3.006
fft_wrap_pw1pw2 1111 12.6 0.007 0.008 2.927 3.002
calculate_first_density_matrix 1 7.0 0.001 0.006 2.722 2.731
acc_transpose_blocks 49320 15.4 0.218 0.225 2.572 2.635
fft_wrap_pw1pw2_140 451 13.1 0.390 0.428 2.508 2.587
jit_kernel_multiply 13 15.9 2.538 2.560 2.538 2.560
potential_pw2rs 110 12.3 0.006 0.006 2.528 2.548
mp_alltoall_d11v 2046 13.8 2.055 2.475 2.055 2.475
grid_collocate_task_list 110 9.6 2.161 2.278 2.161 2.278
fft3d_ps 1111 14.6 0.792 0.880 2.222 2.271
mp_sum_d 3889 11.9 1.506 2.167 1.506 2.167
mp_waitany 14300 13.8 1.867 2.082 1.867 2.082
transfer_rs2pw 451 10.6 0.005 0.006 1.895 2.032
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.903 1.916
make_images_data 4110 15.4 0.043 0.046 1.752 1.889
hybrid_alltoall_any 4261 16.3 0.085 0.486 1.519 1.759
cp_fm_cholesky_invert 11 10.9 1.740 1.744 1.740 1.744
transfer_pw2rs 451 13.1 0.006 0.007 1.723 1.732
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.649 1.675
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=82.184000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=476.636364, yerr=3.082877
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 594.092032E+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 1880216.
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 0.023 0.052 68.566 68.568
qs_mol_dyn_low 1 2.0 0.003 0.004 68.262 68.279
qs_forces 11 3.9 0.011 0.039 68.195 68.196
qs_energies 11 4.9 0.002 0.006 64.836 64.840
scf_env_do_scf 11 5.9 0.001 0.002 56.295 56.298
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 48.833 48.834
dbcsr_multiply_generic 2055 12.4 0.115 0.120 37.771 38.009
velocity_verlet 10 3.0 0.002 0.009 35.753 35.754
qs_scf_new_mos 99 7.5 0.001 0.001 33.071 33.195
qs_scf_loop_do_ot 99 8.5 0.001 0.001 33.070 33.194
multiply_cannon 2055 13.4 0.224 0.245 31.240 32.175
ot_scf_mini 99 9.5 0.003 0.004 31.406 31.526
multiply_cannon_loop 2055 14.4 1.167 1.193 30.118 31.053
ot_mini 99 10.5 0.001 0.001 18.290 18.405
multiply_cannon_multrec 24660 15.4 6.969 9.148 14.015 16.089
rebuild_ks_matrix 110 8.3 0.000 0.000 13.496 13.672
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.016 13.496 13.672
qs_ot_get_derivative 99 11.5 0.001 0.001 12.531 12.653
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.877 12.038
mp_waitall_1 176588 16.5 7.516 9.867 7.516 9.867
multiply_cannon_metrocomm3 24660 15.4 0.072 0.074 5.174 7.798
multiply_cannon_sync_h2d 24660 15.4 6.422 7.773 6.422 7.773
init_scf_loop 11 6.9 0.001 0.006 7.427 7.428
dbcsr_mm_accdrv_process 52282 16.1 5.518 6.450 6.876 7.170
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.436 7.081
apply_single 110 13.6 0.000 0.001 6.436 7.081
qs_ot_get_p 110 10.4 0.001 0.001 6.465 6.595
sum_up_and_integrate 110 10.3 0.053 0.059 6.342 6.353
integrate_v_rspace 110 11.3 0.003 0.004 6.289 6.301
init_scf_run 11 5.9 0.000 0.001 6.093 6.094
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.093 6.094
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.846 5.855
calculate_rho_elec 110 8.6 0.039 0.047 5.845 5.855
ot_diis_step 99 11.5 0.010 0.010 5.713 5.713
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.717 5.452
prepare_preconditioner 11 7.9 0.000 0.000 5.395 5.422
make_preconditioner 11 8.9 0.001 0.002 5.395 5.422
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.973 5.142
qs_ot_p2m_diag 48 11.0 0.029 0.045 4.583 4.602
make_m2s 4110 13.4 0.058 0.061 4.180 4.487
make_images 4110 14.4 0.407 0.466 4.071 4.374
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.118 4.119
pw_transfer 1331 11.6 0.066 0.074 3.521 3.659
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.545 3.558
cp_fm_diag_elpa_base 48 14.0 3.499 3.514 3.542 3.555
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.414 3.554
wfi_extrapolate 11 7.9 0.001 0.001 3.484 3.484
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.268 3.325
grid_integrate_task_list 110 12.3 3.161 3.325 3.161 3.325
density_rs2pw 110 9.6 0.005 0.006 3.105 3.312
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.247 3.249
fft_wrap_pw1pw2_140 451 13.1 0.458 0.471 2.927 3.063
calculate_dm_sparse 110 9.5 0.001 0.001 2.968 2.993
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.902 2.965
fft3d_ps 1111 14.6 1.111 1.338 2.512 2.664
make_images_data 4110 15.4 0.049 0.053 2.269 2.601
hybrid_alltoall_any 4261 16.3 0.107 0.458 1.962 2.574
cp_fm_cholesky_invert 11 10.9 2.507 2.514 2.507 2.514
calculate_first_density_matrix 1 7.0 0.001 0.003 2.504 2.506
mp_sum_l 6594 12.7 1.726 2.361 1.726 2.361
grid_collocate_task_list 110 9.6 2.167 2.358 2.167 2.358
potential_pw2rs 110 12.3 0.008 0.009 2.201 2.210
jit_kernel_multiply 11 16.3 0.996 2.077 0.996 2.077
acc_transpose_blocks 24660 15.4 0.117 0.119 1.982 2.019
mp_alltoall_d11v 2046 13.8 1.719 2.017 1.719 2.017
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.963 1.988
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.885 1.885
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.777 1.794
multiply_cannon_metrocomm4 22605 15.4 0.078 0.083 0.766 1.765
transfer_rs2pw 451 10.6 0.007 0.007 1.428 1.615
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.538 1.549
mp_irecv_dv 57340 16.2 0.636 1.533 0.636 1.533
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.394 1.491
mp_allgather_i34 2055 14.4 0.431 1.430 0.431 1.430
dbcsr_complete_redistribute 325 12.2 0.242 0.300 1.129 1.394
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=68.568000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=560.545455, yerr=6.155022
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 666.406912E+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 0.030 0.048 59.188 59.192
qs_mol_dyn_low 1 2.0 0.006 0.006 58.905 58.920
qs_forces 11 3.9 0.003 0.003 58.836 58.837
qs_energies 11 4.9 0.001 0.001 55.658 55.662
scf_env_do_scf 11 5.9 0.001 0.001 47.779 47.779
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 39.408 39.409
velocity_verlet 10 3.0 0.002 0.002 32.152 32.154
dbcsr_multiply_generic 2055 12.4 0.105 0.110 28.473 28.784
qs_scf_new_mos 99 7.5 0.001 0.001 24.956 25.078
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.956 25.078
ot_scf_mini 99 9.5 0.002 0.002 23.711 23.846
multiply_cannon 2055 13.4 0.212 0.220 22.342 23.229
multiply_cannon_loop 2055 14.4 0.814 0.851 21.307 22.265
ot_mini 99 10.5 0.001 0.001 13.574 13.712
rebuild_ks_matrix 110 8.3 0.000 0.000 12.024 12.189
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 12.023 12.189
multiply_cannon_multrec 16440 15.4 3.624 4.745 9.789 10.866
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.580 10.724
mp_waitall_1 139946 16.5 6.817 9.737 6.817 9.737
qs_ot_get_derivative 99 11.5 0.001 0.001 9.165 9.300
init_scf_loop 11 6.9 0.000 0.000 8.338 8.339
multiply_cannon_metrocomm3 16440 15.4 0.046 0.047 4.390 7.239
prepare_preconditioner 11 7.9 0.000 0.000 6.597 6.613
make_preconditioner 11 8.9 0.000 0.001 6.597 6.613
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.937 6.284
sum_up_and_integrate 110 10.3 0.061 0.062 6.231 6.255
integrate_v_rspace 110 11.3 0.003 0.003 6.170 6.194
dbcsr_mm_accdrv_process 34862 16.1 5.344 5.735 6.008 6.143
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.666 5.680
calculate_rho_elec 110 8.6 0.058 0.059 5.665 5.679
qs_ot_get_p 110 10.4 0.001 0.001 5.347 5.514
init_scf_run 11 5.9 0.000 0.001 5.464 5.464
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.463 5.464
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.926 5.383
apply_single 110 13.6 0.000 0.000 4.926 5.383
ot_diis_step 99 11.5 0.010 0.011 4.379 4.379
make_m2s 4110 13.4 0.051 0.053 4.043 4.307
make_images 4110 14.4 0.399 0.523 3.928 4.193
multiply_cannon_sync_h2d 16440 15.4 3.232 3.876 3.232 3.876
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.119 3.762
qs_ot_p2m_diag 48 11.0 0.041 0.044 3.724 3.728
pw_transfer 1331 11.6 0.065 0.071 3.510 3.520
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.403 3.415
grid_integrate_task_list 110 12.3 3.205 3.372 3.205 3.372
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.357 3.357
density_rs2pw 110 9.6 0.004 0.005 2.907 3.053
fft_wrap_pw1pw2_140 451 13.1 0.579 0.586 2.958 2.972
wfi_extrapolate 11 7.9 0.001 0.001 2.949 2.949
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.831 2.840
cp_fm_diag_elpa_base 48 14.0 2.766 2.795 2.829 2.838
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.771 2.773
cp_fm_cholesky_invert 11 10.9 2.592 2.599 2.592 2.599
make_images_data 4110 15.4 0.045 0.049 2.244 2.598
hybrid_alltoall_any 4261 16.3 0.109 0.384 2.009 2.578
calculate_dm_sparse 110 9.5 0.001 0.001 2.544 2.574
multiply_cannon_metrocomm4 14385 15.4 0.049 0.053 0.893 2.506
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.436 2.495
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.404 2.482
calculate_first_density_matrix 1 7.0 0.000 0.001 2.425 2.427
grid_collocate_task_list 110 9.6 2.222 2.399 2.222 2.399
mp_irecv_dv 48980 15.7 0.817 2.367 0.817 2.367
fft3d_ps 1111 14.6 1.091 1.101 2.345 2.364
mp_sum_l 6594 12.7 1.467 2.077 1.467 2.077
potential_pw2rs 110 12.3 0.011 0.011 2.064 2.068
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.951 1.952
mp_alltoall_d11v 2046 13.8 1.703 1.887 1.703 1.887
dbcsr_complete_redistribute 325 12.2 0.329 0.355 1.416 1.850
cp_fm_upper_to_full 70 14.2 1.357 1.734 1.357 1.734
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.636 1.657
acc_transpose_blocks 16440 15.4 0.078 0.080 1.583 1.653
cp_fm_cholesky_decompose 22 10.9 1.536 1.557 1.536 1.557
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.486 1.500
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.375 1.490
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.951 1.381
transfer_rs2pw 451 10.6 0.005 0.006 1.234 1.380
mp_allgather_i34 2055 14.4 0.342 1.327 0.342 1.327
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.275 1.281
mp_waitany 17072 13.8 1.113 1.269 1.113 1.269
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=59.192000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=631.545455, yerr=7.203076
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 737.710080E+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 0.039 0.125 65.379 65.381
qs_mol_dyn_low 1 2.0 0.003 0.003 65.009 65.019
qs_forces 11 3.9 0.003 0.003 64.945 64.946
qs_energies 11 4.9 0.002 0.006 61.562 61.564
scf_env_do_scf 11 5.9 0.001 0.002 53.219 53.222
scf_env_do_scf_inner_loop 99 6.5 0.005 0.009 41.720 41.720
velocity_verlet 10 3.0 0.002 0.002 37.023 37.025
dbcsr_multiply_generic 2055 12.4 0.113 0.120 30.548 30.749
qs_scf_new_mos 99 7.5 0.001 0.001 27.197 27.314
qs_scf_loop_do_ot 99 8.5 0.002 0.005 27.197 27.313
ot_scf_mini 99 9.5 0.003 0.003 25.512 25.616
multiply_cannon 2055 13.4 0.239 0.261 23.665 24.675
multiply_cannon_loop 2055 14.4 1.427 1.488 22.141 22.766
ot_mini 99 10.5 0.001 0.001 14.698 14.832
multiply_cannon_multrec 24660 15.4 4.092 6.761 13.009 14.235
rebuild_ks_matrix 110 8.3 0.000 0.000 11.936 12.035
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.936 12.035
init_scf_loop 11 6.9 0.001 0.004 11.460 11.461
qs_ot_get_derivative 99 11.5 0.001 0.001 10.531 10.640
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.548 10.634
dbcsr_mm_accdrv_process 52304 16.0 7.706 9.021 8.757 9.749
prepare_preconditioner 11 7.9 0.000 0.000 9.707 9.721
make_preconditioner 11 8.9 0.000 0.002 9.707 9.721
make_full_inverse_cholesky 11 9.9 0.000 0.001 7.957 9.392
sum_up_and_integrate 110 10.3 0.067 0.071 6.218 6.231
integrate_v_rspace 110 11.3 0.003 0.004 6.151 6.167
mp_waitall_1 121746 16.5 3.989 5.896 3.989 5.896
qs_ot_get_p 110 10.4 0.001 0.001 5.690 5.849
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.793 5.806
calculate_rho_elec 110 8.6 0.078 0.081 5.792 5.805
make_m2s 4110 13.4 0.061 0.063 5.284 5.755
make_images 4110 14.4 0.579 0.697 5.141 5.607
init_scf_run 11 5.9 0.000 0.001 5.552 5.553
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.552 5.552
cp_fm_upper_to_full 70 14.2 3.319 4.795 3.319 4.795
ot_diis_step 99 11.5 0.011 0.011 4.131 4.131
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.043 4.113
apply_single 110 13.6 0.000 0.000 4.043 4.113
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.827 3.842
dbcsr_complete_redistribute 325 12.2 0.428 0.476 2.614 3.738
pw_transfer 1331 11.6 0.065 0.074 3.626 3.656
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.520 3.553
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.487 3.540
grid_integrate_task_list 110 12.3 3.296 3.409 3.296 3.409
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.338 3.401
cp_dbcsr_syevd 48 12.0 0.003 0.004 3.259 3.260
copy_fm_to_dbcsr 174 11.2 0.001 0.002 2.102 3.212
fft_wrap_pw1pw2_140 451 13.1 0.609 0.630 3.037 3.073
density_rs2pw 110 9.6 0.004 0.006 2.897 3.056
hybrid_alltoall_any 4261 16.3 0.123 0.460 2.220 3.047
make_images_data 4110 15.4 0.049 0.053 2.571 3.023
calculate_dm_sparse 110 9.5 0.001 0.001 2.980 3.014
wfi_extrapolate 11 7.9 0.001 0.001 2.980 2.980
multiply_cannon_metrocomm3 24660 15.4 0.038 0.039 1.318 2.909
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.741 2.831
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.818 2.821
mp_alltoall_i22 605 13.7 1.625 2.766 1.625 2.766
cp_fm_diag_elpa 48 13.0 0.000 0.001 2.740 2.748
cp_fm_diag_elpa_base 48 14.0 2.591 2.650 2.737 2.744
cp_fm_cholesky_invert 11 10.9 2.613 2.621 2.613 2.621
acc_transpose_blocks 24660 15.4 0.114 0.118 2.443 2.533
multiply_cannon_sync_h2d 24660 15.4 2.394 2.529 2.394 2.529
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 2.449 2.479
calculate_first_density_matrix 1 7.0 0.001 0.002 2.471 2.474
fft3d_ps 1111 14.6 1.088 1.115 2.402 2.416
grid_collocate_task_list 110 9.6 2.262 2.399 2.262 2.399
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 2.285 2.286
potential_pw2rs 110 12.3 0.013 0.013 1.980 1.989
mp_alltoall_d11v 2046 13.8 1.770 1.942 1.770 1.942
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.794 1.830
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.007 1.617 1.712
cp_fm_cholesky_decompose 22 10.9 1.617 1.665 1.617 1.665
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.633 1.645
mp_allgather_i34 2055 14.4 0.614 1.584 0.614 1.584
mp_sum_l 6594 12.7 0.933 1.556 0.933 1.556
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.538 1.555
acc_transpose_blocks_sync 73980 16.4 1.433 1.510 1.433 1.510
multiply_cannon_metrocomm4 20550 15.4 0.063 0.067 0.848 1.485
mp_irecv_dv 62702 16.1 0.742 1.402 0.742 1.402
transfer_rs2pw 451 10.6 0.005 0.006 1.187 1.374
qs_env_update_s_mstruct 11 6.9 0.002 0.005 1.218 1.314
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=65.381000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=697.000000, yerr=9.448425
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 857.165824E+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 0.029 54.676 54.677
qs_mol_dyn_low 1 2.0 0.003 0.003 54.457 54.465
qs_forces 11 3.9 0.003 0.003 54.391 54.392
qs_energies 11 4.9 0.001 0.001 50.737 50.742
scf_env_do_scf 11 5.9 0.000 0.001 42.390 42.391
scf_env_do_scf_inner_loop 99 6.5 0.003 0.006 34.848 34.848
velocity_verlet 10 3.0 0.002 0.002 30.807 30.811
dbcsr_multiply_generic 2055 12.4 0.104 0.109 23.322 23.475
qs_scf_new_mos 99 7.5 0.001 0.001 20.461 20.539
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.461 20.539
ot_scf_mini 99 9.5 0.002 0.002 19.219 19.283
multiply_cannon 2055 13.4 0.239 0.248 17.894 19.195
multiply_cannon_loop 2055 14.4 0.603 0.632 16.668 16.962
rebuild_ks_matrix 110 8.3 0.000 0.000 11.530 11.592
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 11.529 11.591
ot_mini 99 10.5 0.001 0.001 10.707 10.775
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.256 10.316
multiply_cannon_multrec 8220 15.4 3.198 4.568 7.805 9.020
init_scf_loop 11 6.9 0.000 0.000 7.497 7.498
mp_waitall_1 103326 16.6 5.786 7.399 5.786 7.399
qs_ot_get_derivative 99 11.5 0.001 0.001 6.985 7.049
sum_up_and_integrate 110 10.3 0.080 0.082 6.178 6.191
integrate_v_rspace 110 11.3 0.003 0.004 6.098 6.108
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.876 5.890
calculate_rho_elec 110 8.6 0.114 0.115 5.876 5.890
prepare_preconditioner 11 7.9 0.000 0.000 5.866 5.878
make_preconditioner 11 8.9 0.000 0.000 5.866 5.878
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.456 5.528
dbcsr_mm_accdrv_process 17442 15.9 3.169 4.092 4.467 5.447
init_scf_run 11 5.9 0.000 0.001 5.237 5.237
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.236 5.237
qs_ot_get_p 110 10.4 0.001 0.001 4.576 4.655
multiply_cannon_metrocomm3 8220 15.4 0.019 0.019 3.090 4.624
make_m2s 4110 13.4 0.039 0.041 4.127 4.385
make_images 4110 14.4 0.651 0.715 3.995 4.251
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.757 3.795
apply_single 110 13.6 0.000 0.000 3.757 3.795
pw_transfer 1331 11.6 0.065 0.070 3.766 3.780
ot_diis_step 99 11.5 0.012 0.012 3.701 3.701
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.659 3.677
grid_integrate_task_list 110 12.3 3.380 3.485 3.380 3.485
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.250 3.254
fft_wrap_pw1pw2_140 451 13.1 0.776 0.788 3.182 3.203
density_rs2pw 110 9.6 0.004 0.004 2.855 3.008
cp_dbcsr_syevd 48 12.0 0.003 0.003 2.915 2.915
cp_fm_cholesky_invert 11 10.9 2.814 2.817 2.814 2.817
wfi_extrapolate 11 7.9 0.001 0.001 2.682 2.682
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 2.658 2.659
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.640 2.641
hybrid_alltoall_any 4261 16.3 0.200 0.854 2.219 2.636
calculate_dm_sparse 110 9.5 0.001 0.001 2.538 2.576
multiply_cannon_sync_h2d 8220 15.4 2.344 2.551 2.344 2.551
make_images_data 4110 15.4 0.041 0.046 2.194 2.517
grid_collocate_task_list 110 9.6 2.364 2.499 2.364 2.499
calculate_first_density_matrix 1 7.0 0.000 0.000 2.461 2.462
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.353 2.360
cp_fm_diag_elpa_base 48 14.0 2.298 2.323 2.352 2.359
fft3d_ps 1111 14.6 1.132 1.150 2.326 2.338
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.274 2.287
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.028 2.076
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.793 2.023
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.905 1.939
potential_pw2rs 110 12.3 0.016 0.016 1.924 1.928
mp_alltoall_d11v 2046 13.8 1.718 1.838 1.718 1.838
cp_fm_cholesky_decompose 22 10.9 1.636 1.652 1.636 1.652
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.602 1.605
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.488 1.599
jit_kernel_multiply 8 15.6 0.986 1.558 0.986 1.558
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.497 1.509
dbcsr_complete_redistribute 325 12.2 0.565 0.602 1.402 1.485
acc_transpose_blocks 8220 15.4 0.040 0.040 1.435 1.484
mp_allgather_i34 2055 14.4 0.399 1.440 0.399 1.440
multiply_cannon_metrocomm4 6165 15.4 0.019 0.020 0.479 1.411
mp_irecv_dv 24056 15.7 0.451 1.360 0.451 1.360
qs_create_task_list 11 7.9 0.000 0.000 1.209 1.308
generate_qs_task_list 11 8.9 0.372 0.439 1.209 1.308
transfer_rs2pw 451 10.6 0.005 0.005 1.091 1.249
mp_waitany 9240 13.8 1.043 1.221 1.043 1.221
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.179 1.214
multiply_cannon_metrocomm1 8220 15.4 0.022 0.023 0.697 1.162
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=54.677000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=808.181818, yerr=8.726326
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 1.385361E+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 0.019 0.035 86.137 86.138
qs_mol_dyn_low 1 2.0 0.003 0.003 85.844 85.857
qs_forces 11 3.9 0.003 0.003 85.542 85.543
qs_energies 11 4.9 0.001 0.001 81.467 81.469
scf_env_do_scf 11 5.9 0.000 0.001 71.498 71.499
velocity_verlet 10 3.0 0.002 0.002 55.035 55.045
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 43.601 43.602
dbcsr_multiply_generic 2055 12.4 0.119 0.124 29.193 29.293
init_scf_loop 11 6.9 0.000 0.000 27.828 27.830
qs_scf_new_mos 99 7.5 0.001 0.001 26.385 26.423
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.384 26.422
prepare_preconditioner 11 7.9 0.000 0.000 25.832 25.840
make_preconditioner 11 8.9 0.000 0.000 25.832 25.840
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.165 25.313
ot_scf_mini 99 9.5 0.002 0.002 24.594 24.616
multiply_cannon 2055 13.4 0.334 0.357 22.263 23.034
multiply_cannon_loop 2055 14.4 0.827 0.847 20.496 20.787
cp_fm_upper_to_full 70 14.2 12.612 17.997 12.612 17.997
ot_mini 99 10.5 0.001 0.001 13.912 13.938
rebuild_ks_matrix 110 8.3 0.000 0.000 13.551 13.582
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 13.551 13.581
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.295 12.322
dbcsr_complete_redistribute 325 12.2 1.022 1.050 7.205 10.415
multiply_cannon_multrec 8220 15.4 4.050 4.186 9.716 9.831
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.205 9.408
qs_ot_get_derivative 99 11.5 0.001 0.001 9.360 9.384
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.651 8.830
mp_alltoall_i22 605 13.7 5.265 8.451 5.265 8.451
mp_waitall_1 84994 16.7 7.297 8.104 7.297 8.104
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.037 7.078
calculate_rho_elec 110 8.6 0.225 0.225 7.037 7.077
sum_up_and_integrate 110 10.3 0.150 0.151 6.753 6.768
integrate_v_rspace 110 11.3 0.004 0.004 6.603 6.617
dbcsr_mm_accdrv_process 11614 15.7 3.929 4.364 5.517 5.746
make_m2s 4110 13.4 0.045 0.046 5.242 5.725
init_scf_run 11 5.9 0.000 0.001 5.721 5.721
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.721 5.721
make_images 4110 14.4 0.888 0.941 5.052 5.535
qs_ot_get_p 110 10.4 0.001 0.001 5.437 5.468
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.698 5.156
apply_single 110 13.6 0.000 0.000 4.698 5.156
cp_fm_cholesky_invert 11 10.9 5.113 5.117 5.113 5.117
multiply_cannon_metrocomm3 8220 15.4 0.019 0.020 4.748 5.053
pw_transfer 1331 11.6 0.074 0.075 4.828 4.836
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 4.711 4.719
ot_diis_step 99 11.5 0.015 0.015 4.538 4.538
fft_wrap_pw1pw2_140 451 13.1 1.280 1.291 4.131 4.144
qs_ot_p2m_diag 48 11.0 0.151 0.155 3.891 3.898
grid_integrate_task_list 110 12.3 3.674 3.742 3.674 3.742
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.641 3.642
density_rs2pw 110 9.6 0.004 0.004 3.509 3.523
hybrid_alltoall_any 4261 16.3 0.265 0.581 2.757 3.471
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.984 3.432
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.377 3.377
make_images_data 4110 15.4 0.045 0.049 2.731 3.327
wfi_extrapolate 11 7.9 0.001 0.001 3.272 3.272
calculate_dm_sparse 110 9.5 0.001 0.001 3.158 3.186
multiply_cannon_sync_h2d 8220 15.4 3.120 3.144 3.120 3.144
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.846 2.847
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.803 2.804
cp_fm_diag_elpa_base 48 14.0 2.267 2.452 2.802 2.802
fft3d_ps 1111 14.6 1.301 1.316 2.763 2.775
grid_collocate_task_list 110 9.6 2.664 2.680 2.664 2.680
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.612 2.624
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.355 2.366
calculate_first_density_matrix 1 7.0 0.000 0.000 2.337 2.339
qs_env_update_s_mstruct 11 6.9 0.001 0.001 2.161 2.224
potential_pw2rs 110 12.3 0.021 0.022 2.182 2.185
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.093 2.155
mp_alltoall_d11v 2046 13.8 1.971 2.027 1.971 2.027
cp_fm_cholesky_decompose 22 10.9 1.934 1.960 1.934 1.960
qs_create_task_list 11 7.9 0.001 0.001 1.881 1.927
generate_qs_task_list 11 8.9 0.729 0.782 1.880 1.926
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.858 1.862
acc_transpose_blocks 8220 15.4 0.041 0.041 1.723 1.754
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.714 1.744
jit_kernel_multiply 10 15.2 1.392 1.723 1.392 1.723
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=86.138000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1247.363636, yerr=52.378470
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430458527744 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 11613083000832 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 94.228655E+12 0.0% 0.0% 100.0%
flops max/rank 1.103326E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806314816 0.0% 0.0% 100.0%
number of processed stacks 12044928 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.1
marketing flops 145.647559E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 630.968320E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10431456
MPI messages size (bytes):
total size 4.526012E+12
min size 0.000000E+00
max size 4.537280E+06
average size 433.881094E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3605352 96403587072
32768 < size <= 131072 1305088 74666999808
131072 < size <= 4194304 5190114 3200148065288
4194304 < size <= 16777216 263934 1154745839544
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4030 57365.
MP_Allreduce 11154 793.
MP_Sync 87
MP_Alltoall 2242 3063321.
MP_SendRecv 24510 18752.
MP_ISendRecv 24510 18752.
MP_Wait 42802
MP_ISend 16140 108019.
MP_IRecv 16140 108019.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.031 207.683 207.684
qs_mol_dyn_low 1 2.0 0.004 0.004 207.249 207.262
qs_forces 11 3.9 0.005 0.005 207.155 207.156
qs_energies 11 4.9 0.001 0.001 201.547 201.574
scf_env_do_scf 11 5.9 0.001 0.001 185.212 185.215
scf_env_do_scf_inner_loop 118 6.6 0.006 0.017 165.005 165.007
dbcsr_multiply_generic 2527 12.6 0.180 0.184 127.365 127.907
qs_scf_new_mos 118 7.6 0.001 0.001 125.422 125.724
qs_scf_loop_do_ot 118 8.6 0.001 0.001 125.421 125.724
velocity_verlet 10 3.0 0.001 0.002 124.151 124.152
ot_scf_mini 118 9.6 0.003 0.003 118.570 118.815
multiply_cannon 2527 13.6 0.241 0.250 102.795 104.224
multiply_cannon_loop 2527 14.6 2.447 2.522 100.691 102.072
ot_mini 118 10.6 0.001 0.001 67.006 67.279
multiply_cannon_multrec 60648 15.6 32.086 34.742 41.886 43.831
qs_ot_get_derivative 118 11.6 0.001 0.001 42.036 42.314
rebuild_ks_matrix 129 8.3 0.001 0.001 33.749 34.088
qs_ks_build_kohn_sham_matrix 129 9.3 0.015 0.017 33.748 34.087
mp_waitall_1 269254 16.5 29.434 32.775 29.434 32.775
qs_ks_update_qs_env 129 7.6 0.001 0.001 30.316 30.623
qs_ot_get_p 129 10.4 0.001 0.001 28.939 29.356
multiply_cannon_sync_h2d 60648 15.6 26.555 28.467 26.555 28.467
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 24.547 25.072
apply_single 129 13.6 0.001 0.001 24.547 25.072
ot_diis_step 118 11.6 0.008 0.008 24.729 24.730
qs_ot_p2m_diag 83 11.4 0.079 0.092 22.312 22.375
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.081 20.304
init_scf_loop 11 6.9 0.000 0.001 20.134 20.135
cp_dbcsr_syevd 83 12.4 0.005 0.005 19.416 19.417
multiply_cannon_metrocomm3 60648 15.6 0.120 0.124 16.600 18.780
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.502 16.541
cp_fm_diag_elpa_base 83 14.4 16.436 16.477 16.498 16.538
prepare_preconditioner 11 7.9 0.000 0.000 15.539 15.585
make_preconditioner 11 8.9 0.000 0.000 15.539 15.585
make_full_inverse_cholesky 11 9.9 0.000 0.000 14.788 15.001
make_m2s 5054 13.6 0.107 0.115 13.928 14.289
make_images 5054 14.6 0.401 0.420 13.743 14.112
sum_up_and_integrate 129 10.3 0.089 0.107 13.968 13.986
integrate_v_rspace 129 11.3 0.004 0.005 13.878 13.898
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.087 13.213
calculate_rho_elec 129 8.7 0.045 0.064 13.086 13.212
init_scf_run 11 5.9 0.000 0.001 12.192 12.193
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.192 12.193
mp_sum_l 8010 12.9 9.369 10.503 9.369 10.503
dbcsr_mm_accdrv_process 125468 16.2 4.767 4.901 9.365 9.877
wfi_extrapolate 11 7.9 0.001 0.001 9.041 9.041
cp_fm_cholesky_invert 11 10.9 8.812 8.820 8.812 8.820
calculate_dm_sparse 129 9.5 0.001 0.001 8.651 8.752
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 8.466 8.599
qs_ot_get_orbitals 118 10.6 0.001 0.001 8.323 8.439
multiply_cannon_metrocomm1 60648 15.6 0.100 0.104 6.413 8.239
make_images_data 5054 15.6 0.068 0.074 6.815 7.807
pw_transfer 1559 11.6 0.074 0.092 7.580 7.750
fft_wrap_pw1pw2 1301 12.7 0.010 0.012 7.378 7.549
grid_integrate_task_list 129 12.3 7.089 7.546 7.089 7.546
hybrid_alltoall_any 5240 16.5 0.298 2.274 5.984 7.215
density_rs2pw 129 9.7 0.006 0.007 6.655 7.046
fft_wrap_pw1pw2_140 527 13.2 1.147 1.194 6.519 6.699
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.634 6.646
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.778 5.867
fft3d_ps 1301 14.7 2.204 2.856 5.354 5.678
mp_alltoall_d11v 2423 14.1 4.295 5.605 4.295 5.605
grid_collocate_task_list 129 9.7 4.880 5.306 4.880 5.306
cp_fm_cholesky_decompose 22 10.9 4.559 4.574 4.559 4.574
potential_pw2rs 129 12.3 0.009 0.011 4.410 4.429
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=207.684000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=597.272727, yerr=6.607034
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/16/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 2.183246E+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 5975232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.7
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 843.640832E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100942E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703955E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70860 2317615104
32768 < size <= 131072 722992 55511613440
131072 < size <= 4194304 1375664 1398181724160
4194304 < size <= 16777216 154704 1463834332048
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58357.
MP_Allreduce 11058 960.
MP_Sync 87
MP_Alltoall 1969 5057574.
MP_SendRecv 12032 47072.
MP_ISendRecv 12032 47072.
MP_Wait 25916
MP_ISend 11748 212467.
MP_IRecv 11748 212467.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.033 0.060 187.925 187.935
qs_mol_dyn_low 1 2.0 0.003 0.004 187.374 187.397
qs_forces 11 3.9 0.005 0.006 187.295 187.296
qs_energies 11 4.9 0.001 0.001 180.537 180.547
scf_env_do_scf 11 5.9 0.001 0.002 164.338 164.348
scf_env_do_scf_inner_loop 117 6.6 0.009 0.018 131.833 131.835
velocity_verlet 10 3.0 0.001 0.002 118.371 118.372
dbcsr_multiply_generic 2507 12.6 0.188 0.195 97.276 98.591
qs_scf_new_mos 117 7.6 0.001 0.001 93.492 94.167
qs_scf_loop_do_ot 117 8.6 0.001 0.001 93.491 94.166
ot_scf_mini 117 9.6 0.004 0.004 88.685 89.322
multiply_cannon 2507 13.6 0.504 0.562 77.227 81.732
multiply_cannon_loop 2507 14.6 1.577 1.646 73.882 76.822
ot_mini 117 10.6 0.001 0.001 50.075 50.804
mp_waitall_1 214728 16.6 23.982 39.838 23.982 39.838
multiply_cannon_multrec 30084 15.6 20.804 25.528 31.353 36.348
rebuild_ks_matrix 128 8.3 0.001 0.001 32.188 33.017
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.019 32.188 33.017
init_scf_loop 11 6.9 0.000 0.000 32.415 32.416
multiply_cannon_metrocomm3 30084 15.6 0.097 0.101 15.891 29.724
qs_ks_update_qs_env 128 7.6 0.001 0.001 28.943 29.698
qs_ot_get_derivative 117 11.6 0.001 0.002 28.146 28.794
prepare_preconditioner 11 7.9 0.000 0.000 28.070 28.154
make_preconditioner 11 8.9 0.000 0.000 28.070 28.154
make_full_inverse_cholesky 11 9.9 0.000 0.000 26.793 27.331
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.038 23.197
apply_single 128 13.6 0.001 0.001 22.038 23.196
ot_diis_step 117 11.6 0.014 0.015 21.749 21.751
qs_ot_get_p 128 10.4 0.001 0.001 20.703 21.587
multiply_cannon_sync_h2d 30084 15.6 17.930 20.809 17.930 20.809
cp_fm_cholesky_invert 11 10.9 16.363 16.375 16.363 16.375
qs_ot_p2m_diag 83 11.4 0.188 0.216 15.667 15.700
make_m2s 5014 13.6 0.091 0.098 13.908 15.591
make_images 5014 14.6 1.168 1.367 13.698 15.382
cp_dbcsr_syevd 83 12.4 0.005 0.006 14.473 14.474
sum_up_and_integrate 128 10.3 0.116 0.134 13.946 13.974
integrate_v_rspace 128 11.3 0.003 0.004 13.830 13.863
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.238 13.272
calculate_rho_elec 128 8.7 0.087 0.104 13.237 13.272
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 11.171 11.630
init_scf_run 11 5.9 0.000 0.001 11.425 11.426
scf_env_initial_rho_setup 11 6.9 0.000 0.001 11.425 11.426
cp_fm_diag_elpa 83 13.4 0.000 0.001 11.368 11.399
cp_fm_diag_elpa_base 83 14.4 11.117 11.214 11.362 11.389
multiply_cannon_metrocomm4 27577 15.6 0.106 0.122 3.904 10.884
dbcsr_mm_accdrv_process 62242 16.2 5.415 6.232 9.996 10.589
mp_irecv_dv 69486 16.3 3.698 10.470 3.698 10.470
make_images_data 5014 15.6 0.067 0.075 8.164 10.206
hybrid_alltoall_any 5200 16.5 0.352 1.538 6.944 9.717
pw_transfer 1547 11.6 0.084 0.093 8.491 8.550
wfi_extrapolate 11 7.9 0.001 0.001 8.346 8.346
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.268 8.328
grid_integrate_task_list 128 12.3 7.189 7.534 7.189 7.534
fft_wrap_pw1pw2_140 523 13.2 1.224 1.245 7.355 7.428
density_rs2pw 128 9.7 0.006 0.007 6.955 7.297
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.251 7.084
cp_fm_cholesky_decompose 22 10.9 6.820 6.902 6.820 6.902
mp_sum_l 7950 12.9 4.342 6.621 4.342 6.621
calculate_dm_sparse 128 9.5 0.001 0.001 6.458 6.619
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.100 6.109
fft3d_ps 1291 14.7 2.836 2.994 5.816 5.871
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.198 5.459
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.352 5.430
grid_collocate_task_list 128 9.7 5.005 5.345 5.005 5.345
mp_allgather_i34 2507 14.6 1.875 4.832 1.875 4.832
mp_alltoall_d11v 2415 14.1 4.042 4.469 4.042 4.469
potential_pw2rs 128 12.3 0.015 0.018 4.380 4.397
mp_sum_d 4465 12.1 2.622 4.108 2.622 4.108
dbcsr_complete_redistribute 395 12.7 0.771 0.872 3.105 3.961
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=187.935000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=801.545455, yerr=2.387986
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 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 11444707676160 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 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 2.906045E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 3951168 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1697.1
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 956.596224E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1033760
MPI messages size (bytes):
total size 2.695213E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.607194E+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 279168 36591108096
131072 < size <= 4194304 654272 987691483136
4194304 < size <= 16777216 65184 925172905552
16777216 < size 28448 745747251200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3995 58307.
MP_Allreduce 11065 1000.
MP_Sync 86
MP_Alltoall 1700 9383497.
MP_SendRecv 7874 75008.
MP_ISendRecv 7874 75008.
MP_Wait 21654
MP_ISend 11660 275234.
MP_IRecv 11660 275234.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.040 0.094 172.737 172.739
qs_mol_dyn_low 1 2.0 0.017 0.024 172.161 172.177
qs_forces 11 3.9 0.013 0.018 172.054 172.069
qs_energies 11 4.9 0.004 0.013 165.439 165.469
scf_env_do_scf 11 5.9 0.001 0.002 149.622 149.623
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 115.221 115.223
velocity_verlet 10 3.0 0.002 0.002 110.200 110.202
dbcsr_multiply_generic 2485 12.5 0.178 0.183 81.261 82.512
qs_scf_new_mos 116 7.6 0.001 0.001 78.769 79.162
qs_scf_loop_do_ot 116 8.6 0.001 0.001 78.768 79.161
ot_scf_mini 116 9.6 0.003 0.004 74.546 74.974
multiply_cannon 2485 13.5 0.507 0.529 61.975 66.476
multiply_cannon_loop 2485 14.5 1.119 1.189 58.862 61.094
ot_mini 116 10.6 0.001 0.001 42.188 42.601
init_scf_loop 11 6.9 0.001 0.003 34.303 34.306
mp_waitall_1 169034 16.6 24.333 33.774 24.333 33.774
rebuild_ks_matrix 127 8.3 0.001 0.001 29.935 30.439
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.018 29.935 30.438
prepare_preconditioner 11 7.9 0.000 0.001 30.299 30.362
make_preconditioner 11 8.9 0.000 0.001 30.299 30.362
make_full_inverse_cholesky 11 9.9 0.000 0.000 27.931 29.310
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.890 27.358
multiply_cannon_multrec 19880 15.5 12.892 15.847 22.162 25.082
multiply_cannon_metrocomm3 19880 15.5 0.062 0.065 15.362 24.449
qs_ot_get_derivative 116 11.6 0.001 0.002 22.554 22.973
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.749 20.981
apply_single 127 13.6 0.001 0.001 19.749 20.981
ot_diis_step 116 11.6 0.017 0.020 19.523 19.523
qs_ot_get_p 127 10.4 0.001 0.002 17.785 18.231
make_m2s 4970 13.5 0.080 0.086 14.342 15.757
make_images 4970 14.5 1.139 1.230 14.111 15.525
multiply_cannon_sync_h2d 19880 15.5 13.550 15.007 13.550 15.007
cp_fm_cholesky_invert 11 10.9 14.089 14.097 14.089 14.097
sum_up_and_integrate 127 10.3 0.131 0.144 13.861 13.898
integrate_v_rspace 127 11.3 0.004 0.004 13.730 13.771
qs_ot_p2m_diag 82 11.4 0.262 0.269 13.692 13.701
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.254 13.290
calculate_rho_elec 127 8.7 0.129 0.144 13.254 13.289
cp_dbcsr_syevd 82 12.4 0.005 0.005 12.655 12.655
init_scf_run 11 5.9 0.000 0.001 10.682 10.683
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.682 10.683
make_images_data 4970 15.5 0.063 0.072 8.526 10.246
cp_fm_diag_elpa 82 13.4 0.000 0.001 9.584 9.600
cp_fm_diag_elpa_base 82 14.4 9.189 9.331 9.581 9.597
hybrid_alltoall_any 5155 16.4 0.449 2.039 7.442 9.596
multiply_cannon_metrocomm4 17395 15.5 0.067 0.076 3.510 9.507
mp_irecv_dv 49801 16.2 3.381 9.237 3.381 9.237
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 8.831 9.143
dbcsr_mm_accdrv_process 41158 16.2 5.603 5.937 8.726 8.882
pw_transfer 1535 11.6 0.084 0.101 8.564 8.670
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 8.343 8.452
grid_integrate_task_list 127 12.3 7.256 7.695 7.256 7.695
fft_wrap_pw1pw2_140 519 13.2 1.303 1.343 7.458 7.580
wfi_extrapolate 11 7.9 0.001 0.001 7.406 7.406
cp_fm_cholesky_decompose 22 10.9 7.128 7.215 7.128 7.215
cp_fm_upper_to_full 104 14.8 5.608 7.103 5.608 7.103
density_rs2pw 127 9.7 0.006 0.006 6.703 7.077
dbcsr_complete_redistribute 393 12.7 1.178 1.207 4.573 6.310
calculate_dm_sparse 127 9.5 0.001 0.001 5.825 5.921
fft3d_ps 1281 14.7 2.734 2.949 5.729 5.806
grid_collocate_task_list 127 9.7 5.122 5.575 5.122 5.575
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.491 5.497
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.553 5.256
copy_fm_to_dbcsr 208 11.6 0.002 0.003 3.425 5.128
mp_alltoall_d11v 2401 14.1 4.196 4.846 4.196 4.846
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.582 4.745
mp_allgather_i34 2485 14.5 1.627 4.557 1.627 4.557
mp_sum_l 7884 12.9 3.152 4.436 3.152 4.436
potential_pw2rs 127 12.3 0.020 0.023 4.324 4.348
transfer_fm_to_dbcsr 11 9.9 0.020 0.025 2.348 4.037
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.933 3.971
qs_energies_init_hamiltonians 11 5.9 0.001 0.006 3.801 3.827
mp_alltoall_i22 712 14.1 1.927 3.768 1.927 3.768
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.542 3.588
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=172.739000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=908.545455, yerr=14.086510
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430458527744 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 11613083000832 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 94.228655E+12 0.0% 0.0% 100.0%
flops max/rank 4.387043E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806314816 0.0% 0.0% 100.0%
number of processed stacks 6024576 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1129.8
marketing flops 145.647559E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 1.197462E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1152312
MPI messages size (bytes):
total size 2.039354E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.769793E+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 321600 36333158400
131072 < size <= 4194304 721560 792041881600
4194304 < size <= 16777216 70800 669921130560
16777216 < size 30960 541065216000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4030 57809.
MP_Allreduce 11152 1080.
MP_Sync 87
MP_Alltoall 1724 12509439.
MP_SendRecv 5934 75008.
MP_ISendRecv 5934 75008.
MP_Wait 22612
MP_ISend 15064 244788.
MP_IRecv 15064 244788.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.072 189.280 189.281
qs_mol_dyn_low 1 2.0 0.003 0.004 188.815 188.827
qs_forces 11 3.9 0.005 0.010 188.153 188.163
qs_energies 11 4.9 0.013 0.074 181.003 181.028
scf_env_do_scf 11 5.9 0.003 0.021 164.143 164.156
velocity_verlet 10 3.0 0.001 0.002 125.824 125.849
scf_env_do_scf_inner_loop 118 6.6 0.023 0.039 117.943 117.946
dbcsr_multiply_generic 2527 12.6 0.189 0.195 81.270 82.059
qs_scf_new_mos 118 7.6 0.001 0.001 81.378 81.710
qs_scf_loop_do_ot 118 8.6 0.001 0.003 81.377 81.709
ot_scf_mini 118 9.6 0.004 0.006 76.780 77.133
multiply_cannon 2527 13.6 0.557 0.586 56.140 58.871
multiply_cannon_loop 2527 14.6 1.834 1.944 52.392 54.474
init_scf_loop 11 6.9 0.007 0.049 46.075 46.076
ot_mini 118 10.6 0.001 0.001 43.371 43.730
prepare_preconditioner 11 7.9 0.000 0.000 41.877 41.904
make_preconditioner 11 8.9 0.000 0.002 41.877 41.904
make_full_inverse_cholesky 11 9.9 0.013 0.023 35.574 40.549
multiply_cannon_multrec 30324 15.6 13.750 19.246 26.708 31.839
rebuild_ks_matrix 129 8.3 0.001 0.001 29.533 29.869
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.020 29.533 29.868
qs_ks_update_qs_env 129 7.6 0.001 0.001 26.641 26.935
mp_waitall_1 149060 16.7 17.239 26.214 17.239 26.214
qs_ot_get_derivative 118 11.6 0.002 0.002 23.528 23.887
make_m2s 5054 13.6 0.098 0.102 20.594 21.726
make_images 5054 14.6 1.976 2.290 20.279 21.411
qs_ot_get_p 129 10.4 0.001 0.001 19.465 19.867
ot_diis_step 118 11.6 0.018 0.020 19.719 19.720
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.264 19.720
apply_single 129 13.6 0.001 0.001 19.264 19.720
cp_fm_upper_to_full 105 14.8 11.409 16.797 11.409 16.797
cp_fm_cholesky_invert 11 10.9 15.966 15.974 15.966 15.974
qs_ot_p2m_diag 83 11.4 0.342 0.390 15.112 15.168
multiply_cannon_metrocomm3 30324 15.6 0.050 0.053 6.246 14.943
sum_up_and_integrate 129 10.3 0.141 0.152 14.128 14.155
integrate_v_rspace 129 11.3 0.004 0.004 13.987 14.019
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.790 13.822
calculate_rho_elec 129 8.7 0.174 0.191 13.790 13.821
cp_dbcsr_syevd 83 12.4 0.005 0.006 13.709 13.710
make_images_data 5054 15.6 0.068 0.074 11.054 13.020
dbcsr_mm_accdrv_process 62756 16.2 8.511 9.549 12.524 13.003
dbcsr_complete_redistribute 395 12.7 1.516 1.657 9.114 12.895
hybrid_alltoall_any 5240 16.5 0.535 2.224 9.850 12.115
copy_fm_to_dbcsr 209 11.7 0.002 0.002 7.760 11.532
multiply_cannon_sync_h2d 30324 15.6 10.521 11.136 10.521 11.136
init_scf_run 11 5.9 0.000 0.001 10.819 10.821
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.819 10.821
cp_fm_diag_elpa 83 13.4 0.000 0.001 10.487 10.500
cp_fm_diag_elpa_base 83 14.4 9.530 9.863 10.480 10.492
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.834 10.096
transfer_fm_to_dbcsr 11 9.9 0.002 0.008 6.281 9.973
mp_alltoall_i22 718 14.1 5.635 9.390 5.635 9.390
pw_transfer 1559 11.6 0.086 0.099 9.007 9.104
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 8.779 8.881
fft_wrap_pw1pw2_140 527 13.2 1.467 1.502 7.834 7.968
grid_integrate_task_list 129 12.3 7.601 7.911 7.601 7.911
wfi_extrapolate 11 7.9 0.001 0.001 7.705 7.705
cp_fm_cholesky_decompose 22 10.9 7.434 7.541 7.434 7.541
density_rs2pw 129 9.7 0.006 0.006 6.776 7.134
multiply_cannon_metrocomm4 25270 15.6 0.086 0.098 2.856 6.841
mp_irecv_dv 76685 16.2 2.697 6.558 2.697 6.558
calculate_dm_sparse 129 9.5 0.001 0.001 6.295 6.383
fft3d_ps 1301 14.7 2.871 2.962 5.922 5.975
grid_collocate_task_list 129 9.7 5.325 5.719 5.325 5.719
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.529 5.571
mp_alltoall_d11v 2423 14.1 4.954 5.451 4.954 5.451
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 4.602 4.700
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 4.399 4.487
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.477 4.478
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.275 4.346
potential_pw2rs 129 12.3 0.024 0.024 4.240 4.257
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=189.281000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1106.545455, yerr=28.917952
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 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 11528903135232 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 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 5.865089E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 1960712 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3445.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.550598E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 240672
MPI messages size (bytes):
total size 1.331455E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.532238E+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 113904 59718500352
4194304 < size <= 16777216 104976 550376570880
16777216 < size 20208 721350232272
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8931 51.
MP_Alltoall 9654 799394.
MP_ISend 40068 2102573.
MP_IRecv 40068 2101676.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58203.
MP_Allreduce 11082 1166.
MP_Sync 87
MP_Alltoall 1712 18838222.
MP_SendRecv 3840 122880.
MP_ISendRecv 3840 122880.
MP_Wait 16122
MP_ISend 10680 423556.
MP_IRecv 10680 423556.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.053 168.239 168.249
qs_mol_dyn_low 1 2.0 0.003 0.003 167.833 167.864
qs_forces 11 3.9 0.004 0.004 167.740 167.744
qs_energies 11 4.9 0.001 0.001 160.352 160.357
scf_env_do_scf 11 5.9 0.002 0.023 143.147 143.160
velocity_verlet 10 3.0 0.002 0.002 110.806 110.810
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 107.580 107.581
dbcsr_multiply_generic 2507 12.6 0.180 0.192 71.617 72.011
qs_scf_new_mos 117 7.6 0.001 0.001 71.001 71.094
qs_scf_loop_do_ot 117 8.6 0.001 0.001 71.000 71.093
ot_scf_mini 117 9.6 0.003 0.004 66.595 66.667
multiply_cannon 2507 13.6 0.562 0.601 53.133 56.415
multiply_cannon_loop 2507 14.6 0.806 0.850 50.237 51.075
ot_mini 117 10.6 0.001 0.001 37.284 37.349
init_scf_loop 11 6.9 0.000 0.000 35.395 35.398
prepare_preconditioner 11 7.9 0.000 0.000 31.556 31.580
make_preconditioner 11 8.9 0.000 0.000 31.555 31.580
mp_waitall_1 125778 16.7 24.553 30.690 24.553 30.690
make_full_inverse_cholesky 11 9.9 0.017 0.027 29.297 29.579
rebuild_ks_matrix 128 8.3 0.001 0.001 28.966 29.075
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.018 28.966 29.074
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.420 26.524
multiply_cannon_multrec 10028 15.6 10.421 14.489 17.952 21.135
qs_ot_get_derivative 117 11.6 0.001 0.002 20.321 20.399
multiply_cannon_metrocomm3 10028 15.6 0.025 0.027 12.528 19.863
cp_fm_cholesky_invert 11 10.9 17.846 17.852 17.846 17.852
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 16.971 17.206
apply_single 128 13.6 0.001 0.001 16.971 17.206
ot_diis_step 117 11.6 0.020 0.028 16.893 16.893
qs_ot_get_p 128 10.4 0.001 0.001 16.637 16.755
make_m2s 5014 13.6 0.069 0.072 14.683 15.484
make_images 5014 14.6 2.173 2.595 14.374 15.171
sum_up_and_integrate 128 10.3 0.182 0.193 14.107 14.156
integrate_v_rspace 128 11.3 0.004 0.004 13.925 13.978
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.885 13.914
calculate_rho_elec 128 8.7 0.256 0.267 13.885 13.913
qs_ot_p2m_diag 83 11.4 0.495 0.501 12.929 12.947
cp_dbcsr_syevd 83 12.4 0.005 0.006 11.771 11.772
multiply_cannon_sync_h2d 10028 15.6 10.889 11.302 10.889 11.302
init_scf_run 11 5.9 0.000 0.001 10.290 10.290
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.289 10.290
hybrid_alltoall_any 5200 16.5 0.843 3.778 8.277 9.707
make_images_data 5014 15.6 0.056 0.065 8.445 9.662
pw_transfer 1547 11.6 0.085 0.093 9.180 9.206
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.958 8.985
cp_fm_diag_elpa 83 13.4 0.000 0.000 8.926 8.936
cp_fm_diag_elpa_base 83 14.4 8.688 8.761 8.923 8.932
grid_integrate_task_list 128 12.3 7.791 8.093 7.791 8.093
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 7.961 8.012
fft_wrap_pw1pw2_140 523 13.2 1.819 1.864 7.969 7.999
cp_fm_cholesky_decompose 22 10.9 7.843 7.959 7.843 7.959
dbcsr_mm_accdrv_process 20762 16.1 3.220 4.517 7.180 7.882
wfi_extrapolate 11 7.9 0.001 0.001 7.297 7.297
multiply_cannon_metrocomm1 10028 15.6 0.030 0.031 4.372 7.135
density_rs2pw 128 9.7 0.006 0.006 6.601 6.866
calculate_dm_sparse 128 9.5 0.001 0.001 6.054 6.128
grid_collocate_task_list 128 9.7 5.568 5.811 5.568 5.811
dbcsr_complete_redistribute 395 12.7 2.134 2.185 5.341 5.768
fft3d_ps 1291 14.7 2.760 2.832 5.670 5.716
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.304 5.304
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.113 5.121
mp_alltoall_d11v 2415 14.1 4.629 5.021 4.629 5.021
mp_allgather_i34 2507 14.6 1.094 4.528 1.094 4.528
potential_pw2rs 128 12.3 0.027 0.027 4.160 4.172
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.030 4.079
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.597 3.896
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.523 3.851
multiply_cannon_metrocomm4 7521 15.6 0.027 0.030 1.732 3.736
mp_irecv_dv 28860 15.9 1.691 3.657 1.691 3.657
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.568 3.642
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.559 3.594
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.413 3.476
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.450 3.457
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=168.249000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1432.090909, yerr=45.452182
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456868864 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1967095021568 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986252263424 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992004743168 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753954734080 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 11613074706432 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239170400256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239170400256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911117062144 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.237285E+12 0.0% 0.0% 100.0%
flops max/rank 11.786603E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806448352 0.0% 0.0% 100.0%
number of processed stacks 1981024 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3435.8
marketing flops 145.655226E+12
-------------------------------------------------------------------------------
# multiplications 2531
max memory usage/rank 3.148083E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101240
MPI messages size (bytes):
total size 1.145036E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.310119E+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 45728 35500589056
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592178544
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4080 58046.
MP_Allreduce 11277 1498.
MP_Sync 87
MP_Alltoall 1724 36993574.
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 0.051 0.079 289.992 290.003
qs_mol_dyn_low 1 2.0 0.003 0.003 289.393 289.433
qs_forces 11 3.9 0.005 0.005 289.301 289.303
qs_energies 11 4.9 0.001 0.001 280.581 280.588
scf_env_do_scf 11 5.9 0.001 0.002 258.321 258.327
velocity_verlet 10 3.0 0.002 0.002 209.650 209.657
scf_env_do_scf_inner_loop 118 6.6 0.004 0.009 134.359 134.361
init_scf_loop 11 6.9 0.000 0.000 123.692 123.693
prepare_preconditioner 11 7.9 0.000 0.000 118.965 118.986
make_preconditioner 11 8.9 0.000 0.000 118.965 118.985
make_full_inverse_cholesky 11 9.9 0.036 0.039 94.513 116.117
qs_scf_new_mos 118 7.6 0.001 0.001 90.586 90.654
qs_scf_loop_do_ot 118 8.6 0.001 0.001 90.585 90.654
ot_scf_mini 118 9.6 0.004 0.004 85.670 85.689
dbcsr_multiply_generic 2531 12.6 0.214 0.228 82.376 82.885
cp_fm_upper_to_full 105 14.8 52.323 75.787 52.323 75.787
multiply_cannon 2531 13.6 0.687 0.731 58.993 60.204
multiply_cannon_loop 2531 14.6 1.058 1.082 54.958 56.369
ot_mini 118 10.6 0.001 0.001 44.809 44.842
dbcsr_complete_redistribute 395 12.7 4.003 4.041 30.288 43.729
copy_fm_to_dbcsr 209 11.7 0.001 0.002 26.807 40.202
transfer_fm_to_dbcsr 11 9.9 0.030 0.031 24.408 37.684
mp_alltoall_i22 718 14.1 22.173 35.389 22.173 35.389
rebuild_ks_matrix 129 8.3 0.001 0.001 34.226 34.272
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 34.226 34.271
cp_fm_cholesky_invert 11 10.9 33.093 33.100 33.093 33.100
mp_waitall_1 104660 16.8 28.067 32.244 28.067 32.244
qs_ks_update_qs_env 129 7.6 0.001 0.001 31.823 31.871
qs_ot_get_p 129 10.4 0.001 0.001 25.876 25.941
qs_ot_get_derivative 118 11.6 0.002 0.002 24.700 24.720
qs_ot_p2m_diag 83 11.4 0.891 0.974 21.785 21.846
make_m2s 5062 13.6 0.080 0.082 18.784 20.257
multiply_cannon_metrocomm3 10124 15.6 0.025 0.027 18.743 20.102
ot_diis_step 118 11.6 0.022 0.022 20.075 20.076
cp_dbcsr_syevd 83 12.4 0.006 0.006 19.973 19.975
make_images 5062 14.6 3.096 3.281 18.299 19.770
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.377 19.496
apply_single 129 13.6 0.001 0.001 19.377 19.496
multiply_cannon_multrec 10124 15.6 10.353 12.144 18.166 18.248
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.806 16.806
cp_fm_diag_elpa_base 83 14.4 12.392 13.978 16.802 16.802
qs_rho_update_rho_low 129 7.7 0.001 0.001 16.465 16.477
calculate_rho_elec 129 8.7 0.485 0.486 16.464 16.476
sum_up_and_integrate 129 10.3 0.324 0.328 16.340 16.429
integrate_v_rspace 129 11.3 0.005 0.005 16.016 16.101
multiply_cannon_sync_h2d 10124 15.6 14.415 14.445 14.415 14.445
hybrid_alltoall_any 5248 16.5 1.331 3.102 10.803 12.952
make_images_data 5062 15.6 0.066 0.070 10.644 12.821
init_scf_run 11 5.9 0.000 0.001 12.058 12.058
scf_env_initial_rho_setup 11 6.9 0.000 0.000 12.058 12.058
pw_transfer 1559 11.6 0.093 0.094 11.256 11.270
fft_wrap_pw1pw2 1301 12.7 0.011 0.011 11.019 11.032
fft_wrap_pw1pw2_140 527 13.2 3.060 3.081 9.839 9.848
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.702 9.719
dbcsr_mm_accdrv_process 20942 16.1 4.292 6.064 7.560 9.391
cp_fm_cholesky_decompose 22 10.9 8.929 9.017 8.929 9.017
grid_integrate_task_list 129 12.3 8.617 8.805 8.617 8.805
wfi_extrapolate 11 7.9 0.001 0.001 8.799 8.799
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 8.118 8.119
density_rs2pw 129 9.7 0.006 0.006 7.456 7.502
mp_alltoall_d11v 2423 14.1 6.743 6.880 6.743 6.880
calculate_dm_sparse 129 9.5 0.001 0.001 6.773 6.838
grid_collocate_task_list 129 9.7 6.460 6.533 6.460 6.533
fft3d_ps 1301 14.7 2.872 2.881 6.236 6.269
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.229 6.265
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.105 6.175
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=290.003000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2840.545455, yerr=162.216109
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 1.259860E+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 0.024 0.042 84.872 84.873
qs_energies 1 2.0 0.000 0.000 84.380 84.391
ls_scf 1 3.0 0.000 0.000 83.484 83.496
dbcsr_multiply_generic 111 6.7 0.014 0.015 72.524 72.684
multiply_cannon 111 7.7 0.017 0.020 55.825 56.898
multiply_cannon_loop 111 8.7 0.229 0.249 52.436 53.577
ls_scf_main 1 4.0 0.000 0.000 52.286 52.286
density_matrix_trs4 2 5.0 0.002 0.003 46.742 46.820
ls_scf_init_scf 1 4.0 0.000 0.002 28.143 28.144
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.035 27.078
mp_waitall_1 11031 10.9 22.159 25.155 22.159 25.155
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 24.936 24.957
multiply_cannon_multrec 2664 9.7 8.161 8.936 15.554 17.385
multiply_cannon_sync_h2d 2664 9.7 13.725 15.308 13.725 15.308
make_m2s 222 7.7 0.009 0.011 13.054 13.565
make_images 222 8.7 0.098 0.107 13.032 13.546
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.514 11.686
make_images_data 222 9.7 0.004 0.005 7.608 8.200
multiply_cannon_metrocomm3 2664 9.7 0.009 0.011 5.370 8.065
dbcsr_mm_accdrv_process 4760 10.4 0.595 0.714 7.007 8.009
hybrid_alltoall_any 227 10.6 0.216 1.850 6.513 7.788
dbcsr_mm_accdrv_process_sort 4760 11.4 6.211 7.145 6.211 7.145
calculate_norms 4752 9.8 5.550 6.286 5.550 6.286
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.016 5.147
mp_sum_l 887 5.1 3.102 4.451 3.102 4.451
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.065 3.530
mp_irecv_dv 6231 10.9 2.049 3.519 2.049 3.519
make_images_sizes 222 9.7 0.000 0.000 0.779 3.507
mp_alltoall_i44 222 10.7 0.778 3.507 0.778 3.507
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.299 3.315
arnoldi_extremal 4 6.8 0.000 0.000 3.184 3.210
arnoldi_normal_ev 4 7.8 0.001 0.002 3.184 3.210
build_subspace 16 8.4 0.009 0.012 3.087 3.089
ls_scf_post 1 4.0 0.000 0.000 3.055 3.066
ls_scf_store_result 1 5.0 0.000 0.000 2.872 2.913
dbcsr_special_finalize 555 9.7 0.005 0.006 2.315 2.892
dbcsr_merge_single_wm 555 10.7 0.454 0.584 2.306 2.883
make_images_pack 222 9.7 2.207 2.616 2.209 2.618
dbcsr_sort_data 658 11.4 2.111 2.607 2.111 2.607
dbcsr_matrix_vector_mult 304 9.0 0.006 0.014 2.311 2.544
dbcsr_matrix_vector_mult_local 304 10.0 2.069 2.453 2.071 2.455
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.308 2.402
buffer_matrices_ensure_size 222 8.7 1.741 2.075 1.741 2.075
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.770 1.772
rebuild_ks_matrix 3 7.3 0.000 0.000 1.761 1.762
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.761 1.762
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.873000, 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/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 2.172572E+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 0.027 0.042 89.636 89.637
qs_energies 1 2.0 0.000 0.000 89.143 89.147
ls_scf 1 3.0 0.000 0.000 87.815 87.819
dbcsr_multiply_generic 111 6.7 0.022 0.034 74.003 74.419
multiply_cannon 111 7.7 0.027 0.038 52.429 57.199
ls_scf_main 1 4.0 0.000 0.000 54.357 54.357
multiply_cannon_loop 111 8.7 0.136 0.146 49.549 53.231
density_matrix_trs4 2 5.0 0.002 0.003 48.714 48.929
mp_waitall_1 9105 10.9 20.295 30.248 20.295 30.248
ls_scf_init_scf 1 4.0 0.000 0.001 29.789 29.790
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.593 28.700
multiply_cannon_multrec 1332 9.7 13.392 16.285 22.663 26.619
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.125 26.138
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.065 20.594
make_m2s 222 7.7 0.006 0.008 15.135 15.737
make_images 222 8.7 1.366 1.690 15.104 15.707
dbcsr_mm_accdrv_process 4041 10.4 0.355 0.532 8.863 10.435
dbcsr_mm_accdrv_process_sort 4041 11.4 8.380 9.903 8.380 9.903
make_images_data 222 9.7 0.004 0.004 8.754 9.622
hybrid_alltoall_any 227 10.6 0.544 2.549 8.069 9.562
mp_sum_l 887 5.1 4.944 7.837 4.944 7.837
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.188 7.775
mp_irecv_dv 3311 11.0 3.168 7.719 3.168 7.719
calculate_norms 2376 9.8 6.076 6.659 6.076 6.659
multiply_cannon_sync_h2d 1332 9.7 4.938 6.378 4.938 6.378
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.739 6.257
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.001 5.216
arnoldi_extremal 4 6.8 0.000 0.000 4.602 4.620
arnoldi_normal_ev 4 7.8 0.001 0.005 4.602 4.620
build_subspace 16 8.4 0.014 0.021 4.353 4.355
ls_scf_post 1 4.0 0.000 0.000 3.670 3.673
ls_scf_store_result 1 5.0 0.000 0.000 3.388 3.492
dbcsr_matrix_vector_mult 304 9.0 0.010 0.023 3.148 3.369
dbcsr_matrix_vector_mult_local 304 10.0 2.762 3.237 2.765 3.239
mp_allgather_i34 111 8.7 0.875 3.195 0.875 3.195
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.125 2.997
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.591 2.691
dbcsr_data_new 4174 10.1 2.111 2.414 2.111 2.414
make_images_pack 222 9.7 1.846 2.134 1.849 2.136
dbcsr_sort_data 436 11.2 1.874 2.109 1.874 2.109
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.848 1.850
rebuild_ks_matrix 3 7.3 0.000 0.000 1.835 1.837
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.835 1.837
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=89.637000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1804.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 2.918748E+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 0.059 0.183 93.335 93.336
qs_energies 1 2.0 0.001 0.007 92.449 92.460
ls_scf 1 3.0 0.003 0.027 90.497 90.507
dbcsr_multiply_generic 111 6.7 0.016 0.021 75.188 75.501
ls_scf_main 1 4.0 0.001 0.014 56.490 56.495
multiply_cannon 111 7.7 0.033 0.081 51.978 55.841
multiply_cannon_loop 111 8.7 0.118 0.140 49.223 53.230
density_matrix_trs4 2 5.0 0.003 0.026 50.571 50.802
mp_waitall_1 7281 11.0 23.567 34.123 23.567 34.123
ls_scf_init_scf 1 4.0 0.001 0.006 30.254 30.257
ls_scf_init_matrix_S 1 5.0 0.000 0.001 29.028 29.113
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.670 26.686
multiply_cannon_multrec 888 9.7 12.660 15.361 21.215 24.529
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 10.947 23.567
make_m2s 222 7.7 0.006 0.008 16.397 17.138
make_images 222 8.7 1.580 1.854 16.359 17.099
make_images_data 222 9.7 0.004 0.005 9.529 10.621
hybrid_alltoall_any 227 10.6 0.641 2.950 9.078 10.234
dbcsr_mm_accdrv_process 3754 10.4 0.299 0.502 8.077 9.332
dbcsr_mm_accdrv_process_sort 3754 11.4 7.642 8.830 7.642 8.830
mp_sum_l 887 5.1 5.099 8.448 5.099 8.448
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.473 7.399
mp_irecv_dv 2335 11.1 2.458 7.338 2.458 7.338
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.735 7.290
multiply_cannon_sync_h2d 888 9.7 5.968 6.780 5.968 6.780
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.883 6.749
arnoldi_extremal 4 6.8 0.000 0.000 5.047 5.069
arnoldi_normal_ev 4 7.8 0.001 0.004 5.047 5.069
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.799 5.002
calculate_norms 1584 9.8 4.387 4.820 4.387 4.820
build_subspace 16 8.4 0.014 0.020 4.750 4.755
mp_allgather_i34 111 8.7 0.880 3.802 0.880 3.802
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.436 3.766
ls_scf_post 1 4.0 0.001 0.005 3.749 3.755
dbcsr_matrix_vector_mult_local 304 10.0 3.038 3.607 3.041 3.610
ls_scf_store_result 1 5.0 0.000 0.000 3.486 3.578
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.839 2.952
dbcsr_data_new 4116 9.9 2.106 2.441 2.106 2.441
make_images_sizes 222 9.7 0.000 0.000 1.033 2.228
mp_alltoall_i44 222 10.7 1.033 2.228 1.033 2.228
dbcsr_sort_data 325 11.1 1.883 2.160 1.883 2.160
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 1.943 1.943
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.885 1.887
make_images_pack 222 9.7 1.629 1.874 1.632 1.877
rebuild_ks_matrix 3 7.3 0.000 0.000 1.867 1.869
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.867 1.869
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=93.336000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2232.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 3.349975E+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 0.030 0.042 97.203 97.204
qs_energies 1 2.0 0.000 0.000 96.624 96.629
ls_scf 1 3.0 0.000 0.000 94.967 94.973
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.896 79.163
ls_scf_main 1 4.0 0.000 0.000 59.199 59.199
multiply_cannon 111 7.7 0.040 0.068 51.687 56.187
density_matrix_trs4 2 5.0 0.002 0.003 53.093 53.219
multiply_cannon_loop 111 8.7 0.154 0.167 46.688 50.464
ls_scf_init_scf 1 4.0 0.000 0.001 32.506 32.509
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.305 31.374
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.853 28.865
mp_waitall_1 6369 11.0 22.729 28.269 22.729 28.269
multiply_cannon_multrec 1332 9.7 14.165 17.297 22.075 25.558
make_m2s 222 7.7 0.007 0.008 21.194 22.588
make_images 222 8.7 3.140 3.598 21.143 22.541
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.025 17.013
make_images_data 222 9.7 0.004 0.004 11.781 13.277
hybrid_alltoall_any 227 10.6 0.797 3.797 11.178 12.827
dbcsr_mm_accdrv_process 3641 10.4 0.296 0.489 7.537 9.052
mp_sum_l 887 5.1 4.444 8.800 4.444 8.800
dbcsr_mm_accdrv_process_sort 3641 11.4 7.096 8.548 7.096 8.548
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.541 7.199
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.098 6.034
mp_irecv_dv 3229 10.9 2.072 5.956 2.072 5.956
multiply_cannon_sync_h2d 1332 9.7 5.454 5.947 5.454 5.947
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.755 5.895
arnoldi_extremal 4 6.8 0.000 0.000 5.181 5.192
arnoldi_normal_ev 4 7.8 0.001 0.005 5.180 5.192
build_subspace 16 8.4 0.014 0.021 4.850 4.857
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.535 4.708
calculate_norms 2376 9.8 4.178 4.545 4.178 4.545
mp_allgather_i34 111 8.7 2.123 4.438 2.123 4.438
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 3.592 3.886
dbcsr_matrix_vector_mult_local 304 10.0 3.207 3.708 3.209 3.710
dbcsr_sort_data 658 11.4 3.109 3.436 3.109 3.436
ls_scf_post 1 4.0 0.000 0.000 3.262 3.267
dbcsr_special_finalize 555 9.7 0.006 0.007 2.871 3.266
dbcsr_merge_single_wm 555 10.7 0.542 0.669 2.863 3.258
ls_scf_store_result 1 5.0 0.000 0.000 2.989 3.069
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.990 3.031
dbcsr_data_release 10477 10.7 1.587 2.394 1.587 2.394
dbcsr_finalize 304 7.8 0.049 0.061 1.787 1.964
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.204000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2728.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 4.751757E+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 0.039 0.053 91.349 91.350
qs_energies 1 2.0 0.000 0.000 90.611 90.616
ls_scf 1 3.0 0.000 0.000 88.630 88.635
dbcsr_multiply_generic 111 6.7 0.018 0.019 70.198 70.399
ls_scf_main 1 4.0 0.000 0.000 55.951 55.951
multiply_cannon 111 7.7 0.076 0.169 52.201 54.964
multiply_cannon_loop 111 8.7 0.088 0.096 49.670 50.959
density_matrix_trs4 2 5.0 0.002 0.003 49.008 49.082
ls_scf_init_scf 1 4.0 0.000 0.001 29.242 29.243
mp_waitall_1 5436 11.0 23.986 28.497 23.986 28.497
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.997 28.034
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.900 25.909
multiply_cannon_multrec 444 9.7 13.690 16.508 20.738 22.057
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 10.784 15.244
make_m2s 222 7.7 0.005 0.005 13.362 14.294
make_images 222 8.7 2.039 2.484 13.294 14.223
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 5.917 13.475
make_images_data 222 9.7 0.003 0.004 8.100 9.620
hybrid_alltoall_any 227 10.6 0.801 3.817 8.015 9.394
dbcsr_mm_accdrv_process 3003 10.4 0.350 0.399 6.741 7.863
multiply_cannon_sync_h2d 444 9.7 6.762 7.786 6.762 7.786
dbcsr_mm_accdrv_process_sort 3003 11.4 6.380 7.464 6.380 7.464
arnoldi_extremal 4 6.8 0.000 0.000 5.830 5.839
arnoldi_normal_ev 4 7.8 0.001 0.005 5.830 5.839
build_subspace 16 8.4 0.015 0.020 5.445 5.457
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.404 4.682
mp_sum_l 887 5.1 2.630 4.514 2.630 4.514
dbcsr_matrix_vector_mult 304 9.0 0.011 0.022 4.195 4.400
dbcsr_matrix_vector_mult_local 304 10.0 3.733 4.213 3.736 4.215
calculate_norms 792 9.8 3.634 3.752 3.634 3.752
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.501 3.697
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.825 3.691
mp_irecv_dv 1241 11.2 1.489 3.660 1.489 3.660
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.514 3.588
mp_allgather_i34 111 8.7 1.116 3.483 1.116 3.483
ls_scf_post 1 4.0 0.000 0.000 3.437 3.442
ls_scf_store_result 1 5.0 0.000 0.000 3.217 3.262
make_images_sizes 222 9.7 0.000 0.000 0.874 3.262
mp_alltoall_i44 222 10.7 0.874 3.261 0.874 3.261
dbcsr_data_new 4608 9.7 1.799 2.318 1.799 2.318
dbcsr_finalize 304 7.8 0.062 0.077 2.198 2.309
dbcsr_merge_all 275 8.9 0.479 0.528 2.056 2.141
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.018 2.019
rebuild_ks_matrix 3 7.3 0.000 0.000 1.986 1.987
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.986 1.987
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 1.965 1.965
dbcsr_add_d 103 6.2 0.000 0.000 1.744 1.829
dbcsr_add_anytype 103 7.2 0.835 0.868 1.743 1.829
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=91.350000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3774.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 8.882426E+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 0.078 0.097 106.802 106.803
qs_energies 1 2.0 0.000 0.000 105.465 105.473
ls_scf 1 3.0 0.000 0.000 102.547 102.554
dbcsr_multiply_generic 111 6.7 0.024 0.028 76.253 76.384
ls_scf_main 1 4.0 0.000 0.000 63.504 63.504
density_matrix_trs4 2 5.0 0.002 0.003 54.596 54.653
multiply_cannon 111 7.7 0.128 0.216 49.856 51.873
multiply_cannon_loop 111 8.7 0.104 0.155 46.899 47.552
ls_scf_init_scf 1 4.0 0.000 0.001 35.257 35.258
ls_scf_init_matrix_S 1 5.0 0.000 0.000 33.746 33.769
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 31.068 31.075
mp_waitall_1 4527 11.1 21.812 25.205 21.812 25.205
make_m2s 222 7.7 0.005 0.005 22.704 23.778
make_images 222 8.7 3.578 3.892 22.596 23.669
multiply_cannon_multrec 444 9.7 17.844 18.489 22.482 23.145
hybrid_alltoall_any 227 10.6 1.654 3.615 12.845 15.832
make_images_data 222 9.7 0.003 0.004 13.097 15.213
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.123 12.258
multiply_cannon_sync_h2d 444 9.7 8.797 8.867 8.797 8.867
arnoldi_extremal 4 6.8 0.000 0.000 7.406 7.416
arnoldi_normal_ev 4 7.8 0.002 0.008 7.406 7.416
build_subspace 16 8.4 0.026 0.036 6.861 6.874
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.494 5.664
dbcsr_matrix_vector_mult_local 304 10.0 5.090 5.401 5.092 5.404
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.062 5.155
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.815 5.068
dbcsr_mm_accdrv_process 1814 10.4 0.267 0.357 4.449 4.575
dbcsr_mm_accdrv_process_sort 1814 11.4 4.113 4.246 4.113 4.246
ls_scf_post 1 4.0 0.000 0.000 3.787 3.795
make_images_sizes 222 9.7 0.000 0.000 1.472 3.613
mp_alltoall_i44 222 10.7 1.472 3.613 1.472 3.613
ls_scf_store_result 1 5.0 0.000 0.000 3.507 3.546
mp_allgather_i34 111 8.7 1.091 3.499 1.091 3.499
calculate_norms 792 9.8 3.227 3.271 3.227 3.271
dbcsr_finalize 304 7.8 0.082 0.089 3.078 3.180
dbcsr_merge_all 275 8.9 0.885 0.916 2.863 2.960
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.888 2.888
dbcsr_complete_redistribute 5 7.6 1.429 1.469 2.738 2.852
dbcsr_sort_data 325 11.1 2.440 2.512 2.440 2.512
matrix_ls_to_qs 2 6.0 0.000 0.000 2.377 2.497
compute_matrix_preconditioner 1 6.0 0.002 0.002 2.467 2.475
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.414 2.417
dbcsr_data_new 6591 9.6 1.864 2.354 1.864 2.354
rebuild_ks_matrix 3 7.3 0.000 0.000 2.348 2.351
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.348 2.351
dbcsr_new_transposed 4 7.5 0.241 0.253 2.302 2.326
dbcsr_frobenius_norm 74 6.6 2.058 2.139 2.216 2.250
dbcsr_add_d 103 6.2 0.000 0.000 2.125 2.202
dbcsr_add_anytype 103 7.2 0.859 0.891 2.124 2.201
dbcsr_data_release 12724 10.6 1.983 2.181 1.983 2.181
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=106.803000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6966.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e709bc9b963055c87db2d7fb03b6a1948d4b7f58_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 592.834560E+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 4807712.
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 0.085 0.268 239.641 239.655
qs_mol_dyn_low 1 2.0 0.004 0.052 238.260 238.288
qs_forces 5 3.8 0.005 0.011 238.126 238.128
qs_energies 5 4.8 0.002 0.032 234.841 234.882
scf_env_do_scf 5 5.8 0.000 0.002 221.491 221.494
scf_env_do_scf_inner_loop 105 6.6 0.004 0.025 193.868 193.886
qs_scf_new_mos 105 7.6 0.000 0.001 151.375 151.522
qs_scf_loop_do_ot 105 8.6 0.001 0.001 151.375 151.521
ot_scf_mini 105 9.6 0.003 0.004 141.376 141.494
dbcsr_multiply_generic 1445 12.2 0.124 0.131 132.989 133.452
multiply_cannon 1445 13.2 0.275 0.286 113.155 115.427
multiply_cannon_loop 1445 14.2 2.861 3.021 111.319 112.644
velocity_verlet 4 3.0 0.003 0.012 109.341 109.342
ot_mini 105 10.6 0.001 0.001 60.746 60.849
qs_ot_get_p 112 10.4 0.001 0.001 50.380 50.706
multiply_cannon_multrec 69360 15.2 29.577 35.216 39.395 44.841
mp_waitall_1 488190 16.1 35.547 41.801 35.547 41.801
qs_ot_p2m_diag 40 11.0 0.020 0.030 38.958 39.098
qs_ot_get_derivative 55 11.6 0.001 0.001 38.876 39.003
cp_dbcsr_syevd 40 12.0 0.002 0.003 35.540 35.541
multiply_cannon_sync_h2d 69360 15.2 29.162 33.770 29.162 33.770
multiply_cannon_metrocomm3 69360 15.2 0.199 0.212 25.860 33.345
rebuild_ks_matrix 110 8.4 0.000 0.000 30.285 30.449
qs_ks_build_kohn_sham_matrix 110 9.4 0.012 0.013 30.284 30.449
cp_fm_syevd 40 13.0 0.000 0.002 29.991 30.156
qs_ks_update_qs_env 112 7.6 0.001 0.001 27.698 27.847
init_scf_loop 7 6.6 0.000 0.006 27.576 27.578
cp_fm_redistribute_end 40 14.0 12.571 25.078 12.576 25.080
cp_fm_syevd_base 40 14.0 12.495 25.005 12.495 25.005
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 23.145 23.432
apply_single 62 13.6 0.000 0.000 23.145 23.432
prepare_preconditioner 7 7.6 0.000 0.000 22.442 22.472
make_preconditioner 7 8.6 0.000 0.003 22.442 22.472
ot_new_cg_direction 55 11.6 0.001 0.001 21.116 21.117
qs_rho_update_rho_low 110 7.6 0.001 0.001 18.283 18.656
calculate_rho_elec 110 8.6 0.030 0.032 18.283 18.656
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.987 15.130
make_full_inverse_cholesky 7 9.6 0.000 0.001 15.052 15.112
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 14.270 14.379
mp_sum_l 4764 12.2 12.621 13.373 12.621 13.373
pw_transfer 1645 12.4 0.079 0.102 12.654 12.925
density_rs2pw 110 9.6 0.005 0.007 12.485 12.908
fft_wrap_pw1pw2 1425 13.5 0.012 0.015 12.516 12.791
calculate_dm_sparse 110 9.5 0.000 0.001 11.546 11.744
fft_wrap_pw1pw2_240 915 15.0 1.130 1.225 11.050 11.269
qs_vxc_create 110 10.4 0.004 0.007 11.030 11.077
init_scf_run 5 5.8 0.001 0.023 10.606 10.607
scf_env_initial_rho_setup 5 6.8 0.000 0.001 10.605 10.607
dbcsr_mm_accdrv_process 154766 15.8 6.221 6.431 9.686 10.479
cp_fm_cholesky_invert 7 10.6 10.333 10.341 10.333 10.341
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 10.255 10.336
check_diag 80 13.5 8.602 8.885 9.826 9.997
fft3d_pb 915 16.0 2.376 2.647 9.284 9.549
transfer_rs2pw 445 10.6 0.007 0.008 8.935 9.333
sum_up_and_integrate 60 10.3 0.028 0.031 9.147 9.159
integrate_v_rspace 60 11.3 0.002 0.015 9.119 9.132
xc_rho_set_and_dset_create 110 12.4 0.077 0.103 7.670 7.935
make_m2s 2890 13.2 0.079 0.087 6.887 7.491
xc_vxc_pw_create 60 11.3 0.038 0.049 7.440 7.488
make_images 2890 14.2 0.242 0.262 6.781 7.386
multiply_cannon_metrocomm1 69360 15.2 0.102 0.108 4.584 7.144
make_full_single_inverse 7 9.6 0.001 0.002 7.101 7.130
xc_pw_derive 510 13.4 0.005 0.007 6.728 6.828
acc_transpose_blocks 69360 15.2 0.360 0.376 6.321 6.698
mp_alltoall_z22v 2340 17.7 6.372 6.656 6.372 6.656
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.001 6.133 6.150
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.001 5.604 5.662
calculate_first_density_matrix 1 7.0 0.000 0.003 5.617 5.633
potential_pw2rs 60 12.3 0.003 0.003 5.238 5.275
mp_waitany 7680 13.5 4.540 5.087 4.540 5.087
multiply_cannon_metrocomm4 67915 15.2 0.189 0.208 2.026 5.063
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=239.655000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=562.600000, yerr=3.322650
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: e709bc9b963055c87db2d7fb03b6a1948d4b7f58
Summary: empty
Status: OK