=== 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: a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.6, ELPA 2022.11.001, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.2.2, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.2,
# SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Last update: 21.06.2023
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2022.11.001
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.2
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
HDF5_VER := $(strip $(HDF5_VER))
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(HDF5_VER)/lib
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__HDF5
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/01
job id: 47568976
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/02
job id: 47568977
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/03
job id: 47568978
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/04
job id: 47568979
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/05
job id: 47568980
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/06
job id: 47568981
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/07
job id: 47568982
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/08
job id: 47568983
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/09
job id: 47568984
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/10
job id: 47568985
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/11
job id: 47568986
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/12
job id: 47568987
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/13
job id: 47568988
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/14
job id: 47568989
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/15
job id: 47568990
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/16
job id: 47568991
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/17
job id: 47568992
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/18
job id: 47568993
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/19
job id: 47568994
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/20
job id: 47568995
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/21
job id: 47568996
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/22
job id: 47568997
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/23
job id: 47568999
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/24
job id: 47569000
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/25
job id: 47569001
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/26
job id: 47569002
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/27
job id: 47569003
--- 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 0.033 0.035 135.314 135.315
farming_run 1 2.0 134.626 134.628 135.274 135.276
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.463030E+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 0.024 114.729 114.729
qs_energies 1 2.0 0.000 0.000 114.545 114.548
mp2_main 1 3.0 0.000 0.000 112.457 112.460
mp2_gpw_main 1 4.0 0.019 0.025 111.457 111.460
mp2_ri_gpw_compute_in 1 5.0 0.171 0.173 92.552 92.688
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.004 54.779 54.914
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.166 41.215 46.057
get_2c_integrals 1 6.0 0.008 0.011 36.990 37.603
integrate_v_rspace 273 8.0 0.438 0.452 24.644 29.117
pw_transfer 6555 10.6 0.377 0.392 27.402 28.004
fft_wrap_pw1pw2 5465 11.4 0.048 0.051 26.032 26.471
grid_integrate_task_list 273 9.0 20.467 25.351 20.467 25.351
fft_wrap_pw1pw2_100 2178 12.4 1.215 1.421 23.554 24.021
compute_2c_integrals 1 7.0 0.002 0.002 19.692 19.693
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.934 19.431
mp2_eri_2c_integrate_gpw 1 9.0 2.369 2.425 18.932 19.427
rpa_ri_compute_en 1 5.0 0.019 0.022 18.794 18.907
cp_fm_cholesky_decompose 12 8.2 17.231 17.867 17.231 17.867
cholesky_decomp 1 7.0 0.000 0.000 16.138 16.771
fft3d_s 5443 13.4 16.127 16.389 16.150 16.412
ao_to_mo_and_store_B_mult_1 272 7.0 10.725 15.295 10.725 15.295
calculate_wavefunction 272 8.0 5.389 5.466 12.487 13.177
rpa_num_int 1 6.0 0.000 0.000 10.601 10.610
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.586 10.596
calc_mat_Q 8 8.0 0.000 0.000 9.454 9.542
contract_S_to_Q 8 9.0 0.000 0.000 8.874 8.961
calc_potential_gpw 544 9.5 0.005 0.006 8.261 8.642
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.242 8.557
parallel_gemm_fm 14 9.1 0.000 0.000 8.464 8.557
parallel_gemm_fm_cosma 14 10.1 8.464 8.557 8.464 8.557
potential_pw2rs 545 10.0 0.108 0.110 7.701 8.367
create_integ_mat 1 6.0 0.013 0.028 7.962 7.972
collocate_single_gaussian 272 10.0 0.039 0.042 7.510 7.779
array2fm 1 7.0 0.000 0.000 6.876 7.382
pw_scatter_s 2720 13.7 4.413 4.597 4.413 4.597
pw_gather_s 2722 13.2 3.868 4.174 3.868 4.174
array2fm_buffer_send 1 8.0 2.976 3.152 2.976 3.152
pw_poisson_solve 545 10.5 1.107 1.146 2.182 2.327
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=111.460442, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2809.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 0.029 0.038 399.470 399.472
farming_run 1 2.0 397.966 397.974 399.428 399.432
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 897827128576 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 249788821 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 1.226703E+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 11759824.
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 0.009 0.030 210.609 210.610
qs_energies 1 2.0 0.000 0.000 210.337 210.358
scf_env_do_scf 1 3.0 0.000 0.000 107.723 107.723
qs_ks_update_qs_env 5 5.0 0.000 0.000 106.853 106.869
rebuild_ks_matrix 4 6.0 0.000 0.000 106.515 106.524
qs_ks_build_kohn_sham_matrix 4 7.0 0.056 0.064 106.515 106.524
hfx_ks_matrix 4 8.0 0.001 0.001 106.137 106.142
integrate_four_center 4 9.0 0.143 0.458 106.136 106.141
mp2_main 1 3.0 0.000 0.000 102.323 102.345
mp2_gpw_main 1 4.0 0.034 0.077 101.461 101.483
integrate_four_center_main 4 10.0 0.113 0.536 97.408 100.787
integrate_four_center_bin 264 11.0 97.295 100.785 97.295 100.785
init_scf_loop 1 4.0 0.000 0.000 93.187 93.188
mp2_ri_gpw_compute_in 1 5.0 0.064 0.064 74.536 75.684
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 54.075 55.219
mp2_eri_3c_integrate_gpw 91 7.0 0.143 0.160 41.793 46.802
integrate_v_rspace 95 8.0 0.396 0.565 28.177 32.994
pw_transfer 2240 10.6 0.146 0.190 29.879 30.275
fft_wrap_pw1pw2 1868 11.4 0.019 0.024 28.894 29.320
mp2_ri_gpw_compute_en 1 5.0 0.059 0.078 26.780 29.018
grid_integrate_task_list 95 9.0 23.475 28.501 23.475 28.501
fft_wrap_pw1pw2_100 730 12.4 1.270 1.513 26.610 27.101
ao_to_mo_and_store_B_mult_1 91 7.0 10.610 25.930 10.610 25.930
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.824 1.892 24.964 24.975
get_2c_integrals 1 6.0 0.001 0.008 20.378 20.400
compute_2c_integrals 1 7.0 0.003 0.004 19.364 19.374
compute_2c_integrals_loop_lm 1 8.0 0.002 0.008 18.817 19.234
mp2_eri_2c_integrate_gpw 1 9.0 1.738 1.888 18.815 19.232
fft3d_s 1823 13.4 18.424 18.715 18.438 18.728
scf_env_do_scf_inner_loop 4 4.0 0.007 0.345 14.535 14.535
calculate_wavefunction 91 8.0 2.013 2.038 9.731 9.953
mp2_ri_gpw_compute_en_expansio 172 7.0 0.557 0.599 8.775 9.438
potential_pw2rs 186 10.0 0.034 0.035 8.623 9.145
local_gemm 172 8.0 8.218 8.865 8.218 8.865
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.221 8.573
mp2_ri_gpw_compute_en_comm 22 7.0 0.499 0.515 7.971 8.318
calc_potential_gpw 182 9.5 0.002 0.003 7.897 8.197
collocate_single_gaussian 91 10.0 0.017 0.019 7.865 8.107
mp2_ri_gpw_compute_en_ener 172 7.0 6.350 6.430 6.350 6.430
mp_sendrecv_dm3 2068 8.0 6.005 6.362 6.005 6.362
pw_gather_s 912 13.2 4.891 5.349 4.891 5.349
mp_sync 37 10.5 3.408 5.313 3.408 5.313
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.452606, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1514.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 451.919872E+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
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 0.015 0.027 52.564 52.565
qs_mol_dyn_low 1 2.0 0.003 0.003 52.281 52.288
qs_forces 11 3.9 0.002 0.002 52.195 52.197
qs_energies 11 4.9 0.001 0.001 50.677 50.682
scf_env_do_scf 11 5.9 0.000 0.001 43.355 43.356
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 41.194 41.195
dbcsr_multiply_generic 2286 12.5 0.094 0.100 31.358 31.741
qs_scf_new_mos 108 7.5 0.000 0.001 30.113 30.376
qs_scf_loop_do_ot 108 8.5 0.000 0.001 30.112 30.376
ot_scf_mini 108 9.5 0.002 0.002 28.614 28.792
multiply_cannon 2286 13.5 0.187 0.207 25.892 27.292
multiply_cannon_loop 2286 14.5 1.446 1.524 25.094 26.551
velocity_verlet 10 3.0 0.005 0.017 24.017 24.018
ot_mini 108 10.5 0.001 0.001 16.652 16.904
qs_ot_get_derivative 108 11.5 0.001 0.001 13.923 14.105
mp_waitall_1 245248 16.5 6.882 13.443 6.882 13.443
multiply_cannon_multrec 54864 15.5 4.538 6.908 8.994 12.765
multiply_cannon_metrocomm3 54864 15.5 0.071 0.078 4.894 12.131
rebuild_ks_matrix 119 8.3 0.000 0.000 8.400 8.535
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.399 8.535
qs_ot_get_p 119 10.4 0.001 0.001 7.702 7.965
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.402 7.525
multiply_cannon_sync_h2d 54864 15.5 6.279 6.740 6.279 6.740
init_scf_run 11 5.9 0.000 0.001 5.992 5.992
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.992 5.992
dbcsr_mm_accdrv_process 76910 16.1 1.036 1.666 4.375 5.952
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.295 5.342
sum_up_and_integrate 119 10.3 0.012 0.014 4.954 4.961
mp_sum_l 7287 12.8 3.109 4.951 3.109 4.951
integrate_v_rspace 119 11.3 0.002 0.003 4.942 4.950
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 4.465 4.934
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.800 4.866
calculate_rho_elec 119 8.7 0.012 0.017 4.799 4.865
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.624 4.625
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 4.459 4.562
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.317 4.318
jit_kernel_multiply 13 15.8 3.283 4.283 3.283 4.283
cp_fm_redistribute_end 50 14.0 2.200 4.276 2.212 4.282
cp_fm_diag_elpa_base 50 14.0 2.062 4.146 2.067 4.156
calculate_first_density_matrix 1 7.0 0.000 0.000 3.849 3.851
calculate_dm_sparse 119 9.5 0.000 0.001 3.442 3.569
rs_pw_transfer 974 11.9 0.012 0.013 3.323 3.474
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.985 3.189
apply_single 119 13.6 0.000 0.000 2.985 3.189
density_rs2pw 119 9.7 0.004 0.005 2.809 2.917
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.816 2.818
ot_diis_step 108 11.5 0.006 0.006 2.618 2.618
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.566 2.602
pw_transfer 1439 11.6 0.052 0.060 2.223 2.332
potential_pw2rs 119 12.3 0.004 0.004 2.270 2.307
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 2.147 2.259
acc_transpose_blocks 54864 15.5 0.210 0.231 1.703 2.173
init_scf_loop 11 6.9 0.000 0.000 2.131 2.131
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 2.116 2.125
make_m2s 4572 13.5 0.054 0.057 2.045 2.120
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.039 2.097
grid_integrate_task_list 119 12.3 2.000 2.091 2.000 2.091
wfi_extrapolate 11 7.9 0.001 0.001 2.062 2.062
make_images 4572 14.5 0.133 0.138 1.962 2.036
fft3d_ps 1201 14.6 0.371 0.479 1.789 1.896
fft_wrap_pw1pw2_140 487 13.2 0.186 0.206 1.670 1.779
multiply_cannon_metrocomm1 54864 15.5 0.055 0.060 1.008 1.719
mp_sum_d 4135 12.0 1.016 1.637 1.016 1.637
mp_alltoall_d11v 2130 13.8 1.387 1.559 1.387 1.559
mp_waitany 12084 13.8 1.293 1.481 1.293 1.481
grid_collocate_task_list 119 9.7 1.358 1.424 1.358 1.424
acc_transpose_blocks_kernels 54864 16.5 0.227 0.356 0.799 1.138
mp_alltoall_z22v 1201 16.6 1.015 1.101 1.015 1.101
prepare_preconditioner 11 7.9 0.000 0.000 1.043 1.073
make_preconditioner 11 8.9 0.000 0.000 1.043 1.073
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.048 1.061
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=52.565000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=430.454545, yerr=0.782030
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 485.728256E+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 1550568.
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 0.010 0.027 42.170 42.171
qs_mol_dyn_low 1 2.0 0.003 0.003 41.953 41.962
qs_forces 11 3.9 0.002 0.003 41.881 41.883
qs_energies 11 4.9 0.001 0.002 40.132 40.136
scf_env_do_scf 11 5.9 0.000 0.001 33.691 33.691
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 30.995 30.996
dbcsr_multiply_generic 2286 12.5 0.101 0.104 23.151 23.505
qs_scf_new_mos 108 7.5 0.001 0.001 21.100 21.353
qs_scf_loop_do_ot 108 8.5 0.001 0.001 21.099 21.353
ot_scf_mini 108 9.5 0.002 0.003 20.191 20.367
velocity_verlet 10 3.0 0.001 0.001 19.283 19.284
multiply_cannon 2286 13.5 0.208 0.218 17.629 19.168
multiply_cannon_loop 2286 14.5 0.912 0.980 16.467 18.135
ot_mini 108 10.5 0.001 0.001 12.429 12.668
mp_waitall_1 200699 16.5 6.227 11.580 6.227 11.580
qs_ot_get_derivative 108 11.5 0.001 0.001 9.952 10.133
multiply_cannon_metrocomm3 27432 15.5 0.067 0.071 4.372 9.755
multiply_cannon_multrec 27432 15.5 1.996 4.542 6.791 9.645
rebuild_ks_matrix 119 8.3 0.000 0.000 7.596 7.732
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 7.595 7.732
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.712 6.836
dbcsr_mm_accdrv_process 47894 16.0 3.776 6.186 4.724 6.678
init_scf_run 11 5.9 0.000 0.001 5.092 5.092
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.091 5.092
qs_ot_get_p 119 10.4 0.001 0.001 4.828 5.059
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.810 4.988
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.390 4.573
apply_single 119 13.6 0.000 0.000 3.389 4.573
sum_up_and_integrate 119 10.3 0.025 0.028 4.553 4.559
integrate_v_rspace 119 11.3 0.002 0.003 4.528 4.535
mp_sum_l 7287 12.8 2.394 4.382 2.394 4.382
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.228 4.265
calculate_rho_elec 119 8.7 0.021 0.024 4.227 4.264
rs_pw_transfer 974 11.9 0.010 0.011 3.157 3.579
calculate_first_density_matrix 1 7.0 0.000 0.000 3.468 3.470
qs_ot_p2m_diag 50 11.0 0.009 0.013 3.281 3.299
calculate_dm_sparse 119 9.5 0.000 0.001 2.900 2.978
multiply_cannon_sync_h2d 27432 15.5 2.182 2.914 2.182 2.914
density_rs2pw 119 9.7 0.004 0.004 2.494 2.904
make_m2s 4572 13.5 0.052 0.054 2.676 2.901
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.839 2.839
make_images 4572 14.5 0.201 0.237 2.586 2.809
init_scf_loop 11 6.9 0.000 0.000 2.668 2.669
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.451 2.451
ot_diis_step 108 11.5 0.010 0.011 2.428 2.429
cp_fm_redistribute_end 50 14.0 1.240 2.412 1.249 2.418
cp_fm_diag_elpa_base 50 14.0 1.131 2.288 1.162 2.342
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.192 2.282
potential_pw2rs 119 12.3 0.006 0.006 2.246 2.265
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.209 2.211
pw_transfer 1439 11.6 0.066 0.069 2.078 2.120
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 2.106 2.113
jit_kernel_multiply 9 16.1 0.895 2.078 0.895 2.078
fft_wrap_pw1pw2 1201 12.6 0.008 0.009 1.985 2.030
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.983 2.022
grid_integrate_task_list 119 12.3 1.840 1.935 1.840 1.935
acc_transpose_blocks 27432 15.5 0.112 0.118 1.278 1.917
prepare_preconditioner 11 7.9 0.000 0.000 1.720 1.747
make_preconditioner 11 8.9 0.000 0.000 1.720 1.747
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.612 1.670
make_images_data 4572 15.5 0.045 0.052 1.197 1.633
fft3d_ps 1201 14.6 0.522 0.578 1.535 1.577
hybrid_alltoall_any 4725 16.4 0.051 0.112 1.049 1.568
wfi_extrapolate 11 7.9 0.001 0.001 1.562 1.562
fft_wrap_pw1pw2_140 487 13.2 0.203 0.213 1.507 1.554
grid_collocate_task_list 119 9.7 1.275 1.432 1.275 1.432
mp_alltoall_d11v 2130 13.8 1.295 1.415 1.295 1.415
mp_allgather_i34 2286 14.5 0.602 1.372 0.602 1.372
acc_transpose_blocks_kernels 27432 16.5 0.182 0.270 0.727 1.277
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.203 1.249
mp_sum_d 4135 12.0 0.588 1.038 0.588 1.038
jit_kernel_transpose 5 15.5 0.545 1.007 0.545 1.007
mp_waitany 5720 13.7 0.535 0.978 0.535 0.978
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.943 0.957
rs_pw_transfer_RS2PW_140 130 11.5 0.139 0.147 0.557 0.954
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 0.945 0.947
make_images_sizes 4572 15.5 0.005 0.005 0.655 0.934
rs_pw_transfer_PW2RS_50 119 14.3 0.585 0.604 0.860 0.932
mp_alltoall_i44 4572 16.5 0.650 0.929 0.650 0.929
mp_alltoall_z22v 1201 16.6 0.771 0.855 0.771 0.855
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=42.171000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=462.545455, yerr=1.924183
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 518.660096E+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 1054328.
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 0.012 0.030 35.992 35.993
qs_mol_dyn_low 1 2.0 0.003 0.003 35.742 35.750
qs_forces 11 3.9 0.002 0.003 35.662 35.665
qs_energies 11 4.9 0.001 0.002 34.043 34.047
scf_env_do_scf 11 5.9 0.000 0.001 28.702 28.702
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 25.614 25.614
dbcsr_multiply_generic 2286 12.5 0.096 0.098 17.473 17.560
qs_scf_new_mos 108 7.5 0.001 0.001 16.725 16.745
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.724 16.744
velocity_verlet 10 3.0 0.001 0.002 16.385 16.386
ot_scf_mini 108 9.5 0.002 0.002 15.990 16.003
multiply_cannon 2286 13.5 0.196 0.203 13.839 14.602
multiply_cannon_loop 2286 14.5 0.643 0.670 12.994 13.794
ot_mini 108 10.5 0.001 0.001 9.486 9.504
qs_ot_get_derivative 108 11.5 0.001 0.001 7.961 7.981
multiply_cannon_multrec 18288 15.5 2.031 3.311 7.156 7.723
rebuild_ks_matrix 119 8.3 0.000 0.000 7.004 7.017
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.012 7.004 7.016
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.214 6.225
dbcsr_mm_accdrv_process 38222 16.0 4.171 5.906 5.039 5.973
mp_waitall_1 158411 16.6 3.302 4.490 3.302 4.490
sum_up_and_integrate 119 10.3 0.031 0.031 4.357 4.363
integrate_v_rspace 119 11.3 0.003 0.003 4.326 4.335
qs_ot_get_p 119 10.4 0.002 0.015 4.315 4.328
init_scf_run 11 5.9 0.000 0.001 4.003 4.003
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.003 4.003
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.860 3.864
calculate_rho_elec 119 8.7 0.031 0.031 3.859 3.863
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.865 3.492
rs_pw_transfer 974 11.9 0.010 0.010 2.807 3.081
qs_ot_p2m_diag 50 11.0 0.012 0.013 3.072 3.077
init_scf_loop 11 6.9 0.000 0.000 3.066 3.067
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.784 2.784
calculate_first_density_matrix 1 7.0 0.000 0.000 2.604 2.604
density_rs2pw 119 9.7 0.004 0.004 2.324 2.578
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.459 2.470
cp_fm_diag_elpa_base 50 14.0 2.432 2.447 2.457 2.468
multiply_cannon_metrocomm3 18288 15.5 0.044 0.046 1.416 2.421
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.051 2.366
apply_single 119 13.6 0.000 0.000 2.050 2.365
make_m2s 4572 13.5 0.044 0.045 2.059 2.199
jit_kernel_multiply 10 15.9 0.815 2.174 0.815 2.174
pw_transfer 1439 11.6 0.066 0.070 2.108 2.121
make_images 4572 14.5 0.191 0.203 1.973 2.111
potential_pw2rs 119 12.3 0.007 0.008 2.075 2.090
calculate_dm_sparse 119 9.5 0.000 0.000 2.026 2.034
fft_wrap_pw1pw2 1201 12.6 0.008 0.009 2.015 2.029
prepare_preconditioner 11 7.9 0.000 0.000 2.012 2.014
make_preconditioner 11 8.9 0.000 0.000 2.012 2.014
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.851 1.944
grid_integrate_task_list 119 12.3 1.802 1.911 1.802 1.911
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.906 1.908
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.694 1.704
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.674 1.681
multiply_cannon_sync_h2d 18288 15.5 1.369 1.578 1.369 1.578
fft_wrap_pw1pw2_140 487 13.2 0.254 0.262 1.537 1.550
fft3d_ps 1201 14.6 0.531 0.551 1.500 1.515
ot_diis_step 108 11.5 0.011 0.011 1.498 1.499
mp_sum_l 7287 12.8 1.186 1.494 1.186 1.494
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.392 1.396
grid_collocate_task_list 119 9.7 1.249 1.382 1.249 1.382
wfi_extrapolate 11 7.9 0.001 0.001 1.330 1.330
acc_transpose_blocks 18288 15.5 0.078 0.080 1.261 1.330
multiply_cannon_metrocomm1 18288 15.5 0.029 0.030 0.738 1.077
make_images_data 4572 15.5 0.045 0.049 0.894 1.068
qs_energies_init_hamiltonians 11 5.9 0.002 0.011 0.970 0.978
hybrid_alltoall_any 4725 16.4 0.055 0.113 0.775 0.970
mp_alltoall_d11v 2130 13.8 0.817 0.939 0.817 0.939
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.907 0.933
cp_fm_cholesky_invert 11 10.9 0.879 0.884 0.879 0.884
mp_alltoall_z22v 1201 16.6 0.799 0.867 0.799 0.867
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.865 0.867
acc_transpose_blocks_kernels 18288 16.5 0.211 0.220 0.804 0.858
mp_waitany 9880 13.7 0.544 0.822 0.544 0.822
rs_pw_transfer_RS2PW_140 130 11.5 0.120 0.123 0.528 0.789
dbcsr_complete_redistribute 329 12.2 0.123 0.141 0.638 0.751
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=35.993000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=494.272727, yerr=1.354515
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 543.690752E+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 0.011 0.028 37.078 37.079
qs_mol_dyn_low 1 2.0 0.003 0.003 36.881 36.919
qs_forces 11 3.9 0.002 0.003 36.811 36.812
qs_energies 11 4.9 0.001 0.002 35.042 35.047
scf_env_do_scf 11 5.9 0.000 0.001 29.250 29.251
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 25.715 25.715
dbcsr_multiply_generic 2286 12.5 0.099 0.103 18.823 18.932
velocity_verlet 10 3.0 0.001 0.002 18.412 18.413
qs_scf_new_mos 108 7.5 0.001 0.001 16.933 16.979
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.932 16.978
ot_scf_mini 108 9.5 0.002 0.003 15.983 16.032
multiply_cannon 2286 13.5 0.225 0.263 15.049 15.444
multiply_cannon_loop 2286 14.5 0.953 0.981 14.056 14.460
ot_mini 108 10.5 0.001 0.001 9.679 9.737
multiply_cannon_multrec 27432 15.5 2.373 3.084 8.712 9.055
qs_ot_get_derivative 108 11.5 0.001 0.001 7.853 7.901
dbcsr_mm_accdrv_process 47916 15.9 5.317 7.316 6.248 7.428
rebuild_ks_matrix 119 8.3 0.000 0.000 6.764 6.812
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.764 6.811
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.990 6.032
init_scf_run 11 5.9 0.000 0.001 4.042 4.042
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.041 4.042
sum_up_and_integrate 119 10.3 0.035 0.038 3.978 3.984
integrate_v_rspace 119 11.3 0.003 0.003 3.943 3.950
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.745 3.780
calculate_rho_elec 119 8.7 0.040 0.046 3.744 3.779
qs_ot_get_p 119 10.4 0.001 0.001 3.664 3.734
init_scf_loop 11 6.9 0.000 0.000 3.513 3.514
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.992 3.386
prepare_preconditioner 11 7.9 0.000 0.000 2.657 2.666
make_preconditioner 11 8.9 0.000 0.000 2.657 2.666
mp_waitall_1 137007 16.6 2.079 2.643 2.079 2.643
calculate_first_density_matrix 1 7.0 0.000 0.000 2.617 2.618
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.266 2.591
rs_pw_transfer 974 11.9 0.009 0.010 2.248 2.449
make_m2s 4572 13.5 0.054 0.056 2.316 2.404
calculate_dm_sparse 119 9.5 0.000 0.000 2.343 2.396
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.332 2.341
density_rs2pw 119 9.7 0.004 0.004 2.112 2.313
make_images 4572 14.5 0.271 0.331 2.208 2.295
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.059 2.183
apply_single 119 13.6 0.000 0.000 2.058 2.182
pw_transfer 1439 11.6 0.066 0.070 2.099 2.134
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.006 2.045
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.981 2.006
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.004 2.005
jit_kernel_multiply 10 15.8 0.869 1.929 0.869 1.929
grid_integrate_task_list 119 12.3 1.832 1.907 1.832 1.907
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.900 1.901
acc_transpose_blocks 27432 15.5 0.114 0.115 1.800 1.884
ot_diis_step 108 11.5 0.012 0.013 1.788 1.788
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.721 1.730
cp_fm_diag_elpa_base 50 14.0 1.684 1.702 1.718 1.727
potential_pw2rs 119 12.3 0.009 0.010 1.711 1.717
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.699 1.711
fft_wrap_pw1pw2_140 487 13.2 0.289 0.303 1.619 1.660
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.511 1.522
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.494 1.495
fft3d_ps 1201 14.6 0.557 0.606 1.443 1.467
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.864 1.452
grid_collocate_task_list 119 9.7 1.251 1.368 1.251 1.368
wfi_extrapolate 11 7.9 0.001 0.001 1.366 1.366
mp_sum_l 7287 12.8 1.030 1.285 1.030 1.285
acc_transpose_blocks_kernels 27432 16.5 0.268 0.274 1.154 1.221
cp_fm_upper_to_full 72 14.2 0.816 1.162 0.816 1.162
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.121 1.140
dbcsr_complete_redistribute 329 12.2 0.131 0.152 0.846 1.117
multiply_cannon_sync_h2d 27432 15.5 0.983 1.039 0.983 1.039
qs_env_update_s_mstruct 11 6.9 0.222 0.250 0.995 1.026
make_images_data 4572 15.5 0.044 0.048 0.915 1.022
mp_alltoall_d11v 2130 13.8 0.815 0.994 0.815 0.994
hybrid_alltoall_any 4725 16.4 0.062 0.151 0.797 0.962
jit_kernel_transpose 5 15.6 0.887 0.953 0.887 0.953
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.609 0.876
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.800 0.875
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.865 0.870
cp_fm_cholesky_invert 11 10.9 0.826 0.830 0.826 0.830
mp_alltoall_z22v 1201 16.6 0.770 0.806 0.770 0.806
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=37.079000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=515.727273, yerr=2.631657
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 604.692480E+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 0.012 0.028 29.549 29.550
qs_mol_dyn_low 1 2.0 0.003 0.003 29.367 29.375
qs_forces 11 3.9 0.002 0.002 29.284 29.286
qs_energies 11 4.9 0.032 0.037 27.539 27.542
scf_env_do_scf 11 5.9 0.000 0.001 22.076 22.077
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.315 19.315
velocity_verlet 10 3.0 0.016 0.017 14.797 14.799
dbcsr_multiply_generic 2286 12.5 0.092 0.098 12.582 12.661
qs_scf_new_mos 108 7.5 0.001 0.001 11.199 11.224
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.198 11.223
ot_scf_mini 108 9.5 0.002 0.002 10.523 10.549
multiply_cannon 2286 13.5 0.230 0.237 9.949 10.416
multiply_cannon_loop 2286 14.5 0.333 0.343 9.016 9.168
multiply_cannon_multrec 9144 15.5 1.595 1.826 6.032 6.332
rebuild_ks_matrix 119 8.3 0.000 0.000 6.259 6.277
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.258 6.277
ot_mini 108 10.5 0.001 0.001 5.723 5.752
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.592 5.608
dbcsr_mm_accdrv_process 12550 15.8 3.511 4.406 4.336 4.451
qs_ot_get_derivative 108 11.5 0.001 0.001 4.408 4.433
init_scf_run 11 5.9 0.000 0.002 3.970 3.970
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.970 3.970
sum_up_and_integrate 119 10.3 0.038 0.042 3.720 3.725
integrate_v_rspace 119 11.3 0.003 0.003 3.683 3.688
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.620 3.627
calculate_rho_elec 119 8.7 0.060 0.061 3.620 3.626
qs_ot_get_p 119 10.4 0.001 0.001 2.984 3.030
calculate_first_density_matrix 1 7.0 0.000 0.000 2.746 2.747
init_scf_loop 11 6.9 0.000 0.000 2.736 2.737
mp_waitall_1 115863 16.7 1.515 2.263 1.515 2.263
calculate_dm_sparse 119 9.5 0.000 0.000 2.137 2.157
pw_transfer 1439 11.6 0.066 0.069 2.105 2.115
density_rs2pw 119 9.7 0.004 0.004 1.921 2.061
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.020 2.022
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.011 2.022
make_m2s 4572 13.5 0.034 0.035 1.825 1.978
grid_integrate_task_list 119 12.3 1.850 1.923 1.850 1.923
make_images 4572 14.5 0.268 0.303 1.735 1.886
rs_pw_transfer 974 11.9 0.008 0.008 1.750 1.882
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.813 1.814
prepare_preconditioner 11 7.9 0.000 0.000 1.778 1.783
make_preconditioner 11 8.9 0.000 0.000 1.778 1.783
jit_kernel_multiply 8 15.6 0.786 1.707 0.786 1.707
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.667 1.693
fft_wrap_pw1pw2_140 487 13.2 0.366 0.373 1.633 1.645
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.575 1.589
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.570 1.571
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.561 1.566
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.503 1.511
cp_fm_diag_elpa_base 50 14.0 1.474 1.493 1.501 1.509
potential_pw2rs 119 12.3 0.010 0.011 1.438 1.441
grid_collocate_task_list 119 9.7 1.296 1.415 1.296 1.415
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.384 1.394
fft3d_ps 1201 14.6 0.560 0.573 1.343 1.351
ot_diis_step 108 11.5 0.012 0.013 1.301 1.301
qs_energies_init_hamiltonians 11 5.9 0.004 0.006 1.231 1.233
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.177 1.202
apply_single 119 13.6 0.000 0.000 1.177 1.201
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.156 1.168
wfi_extrapolate 11 7.9 0.001 0.001 1.166 1.166
hybrid_alltoall_any 4725 16.4 0.063 0.175 0.834 1.069
acc_transpose_blocks 9144 15.5 0.039 0.041 1.029 1.045
make_images_data 4572 15.5 0.039 0.043 0.854 1.033
mp_alltoall_d11v 2130 13.8 0.822 0.942 0.822 0.942
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.874 0.923
cp_fm_cholesky_invert 11 10.9 0.901 0.904 0.901 0.904
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.791 0.793
acc_transpose_blocks_kernels 9144 16.5 0.116 0.119 0.781 0.792
multiply_cannon_sync_h2d 9144 15.5 0.709 0.776 0.709 0.776
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.742 0.750
qs_env_update_s_mstruct 11 6.9 0.001 0.001 0.676 0.727
mp_allgather_i34 2286 14.5 0.265 0.707 0.265 0.707
multiply_cannon_metrocomm3 9144 15.5 0.019 0.019 0.316 0.687
mp_alltoall_z22v 1201 16.6 0.657 0.683 0.657 0.683
jit_kernel_transpose 5 15.6 0.664 0.675 0.664 0.675
multiply_cannon_metrocomm1 9144 15.5 0.022 0.023 0.318 0.655
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=29.550000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=572.272727, yerr=5.925153
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 746.422272E+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 0.034 41.909 41.910
qs_mol_dyn_low 1 2.0 0.003 0.003 41.708 41.717
qs_forces 11 3.9 0.002 0.002 41.635 41.636
qs_energies 11 4.9 0.001 0.001 39.627 39.630
scf_env_do_scf 11 5.9 0.001 0.001 33.395 33.395
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 25.513 25.515
velocity_verlet 10 3.0 0.002 0.002 23.353 23.360
dbcsr_multiply_generic 2286 12.5 0.099 0.102 17.760 17.776
qs_scf_new_mos 108 7.5 0.001 0.001 15.830 15.937
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.829 15.936
ot_scf_mini 108 9.5 0.002 0.002 14.754 14.861
multiply_cannon 2286 13.5 0.302 0.313 14.135 14.779
multiply_cannon_loop 2286 14.5 0.346 0.351 12.896 13.576
multiply_cannon_multrec 9144 15.5 3.465 5.231 8.724 9.274
ot_mini 108 10.5 0.001 0.001 8.927 9.048
init_scf_loop 11 6.9 0.000 0.000 7.854 7.858
rebuild_ks_matrix 119 8.3 0.000 0.001 7.258 7.405
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.258 7.404
qs_ot_get_derivative 108 11.5 0.001 0.001 6.917 7.022
prepare_preconditioner 11 7.9 0.000 0.000 6.870 6.884
make_preconditioner 11 8.9 0.000 0.000 6.870 6.884
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.406 6.766
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.541 6.674
dbcsr_mm_accdrv_process 12550 15.8 4.247 5.650 5.135 6.384
cp_fm_upper_to_full 72 14.2 3.197 4.607 3.197 4.607
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.230 4.233
calculate_rho_elec 119 8.7 0.118 0.121 4.229 4.233
init_scf_run 11 5.9 0.000 0.001 4.193 4.193
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.193 4.193
sum_up_and_integrate 119 10.3 0.065 0.066 3.964 3.970
integrate_v_rspace 119 11.3 0.004 0.004 3.899 3.906
mp_waitall_1 94719 16.7 2.520 3.499 2.520 3.499
qs_ot_get_p 119 10.4 0.001 0.001 3.121 3.261
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.798 3.117
dbcsr_complete_redistribute 329 12.2 0.286 0.291 2.000 2.832
calculate_first_density_matrix 1 7.0 0.000 0.000 2.736 2.738
mp_alltoall_i22 627 13.8 1.676 2.683 1.676 2.683
pw_transfer 1439 11.6 0.069 0.069 2.644 2.649
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.545 2.551
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.697 2.508
multiply_cannon_metrocomm3 9144 15.5 0.020 0.021 1.561 2.459
make_m2s 4572 13.5 0.038 0.038 2.183 2.342
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.072 2.278
apply_single 119 13.6 0.000 0.000 2.072 2.278
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.271 2.272
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.459 2.261
calculate_dm_sparse 119 9.5 0.000 0.000 2.226 2.242
make_images 4572 14.5 0.353 0.383 2.062 2.220
density_rs2pw 119 9.7 0.004 0.004 2.151 2.168
fft_wrap_pw1pw2_140 487 13.2 0.618 0.622 2.153 2.160
grid_integrate_task_list 119 12.3 2.045 2.062 2.045 2.062
ot_diis_step 108 11.5 0.014 0.014 1.985 1.985
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.782 1.980
qs_ot_p2m_diag 50 11.0 0.043 0.044 1.891 1.892
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.763 1.764
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.657 1.710
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.619 1.620
fft3d_ps 1201 14.6 0.595 0.607 1.561 1.564
grid_collocate_task_list 119 9.7 1.497 1.508 1.497 1.508
rs_pw_transfer 974 11.9 0.009 0.009 1.445 1.459
mp_sum_l 7287 12.8 0.898 1.458 0.898 1.458
cp_fm_cholesky_invert 11 10.9 1.447 1.451 1.447 1.451
wfi_extrapolate 11 7.9 0.001 0.001 1.390 1.390
potential_pw2rs 119 12.3 0.014 0.014 1.366 1.368
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.340 1.340
cp_fm_diag_elpa_base 50 14.0 1.192 1.245 1.338 1.338
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.306 1.312
hybrid_alltoall_any 4725 16.4 0.087 0.147 1.060 1.250
make_images_data 4572 15.5 0.043 0.046 0.992 1.191
acc_transpose_blocks 9144 15.5 0.040 0.040 1.062 1.176
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.125 1.144
copy_dbcsr_to_fm 153 11.3 0.002 0.003 0.922 1.141
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.088 1.102
mp_alltoall_d11v 2130 13.8 1.061 1.087 1.061 1.087
jit_kernel_multiply 6 15.5 0.861 1.064 0.861 1.064
multiply_cannon_sync_h2d 9144 15.5 1.038 1.042 1.038 1.042
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.000 1.031
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.929 0.942
qs_create_task_list 11 7.9 0.000 0.001 0.930 0.942
generate_qs_task_list 11 8.9 0.367 0.386 0.929 0.941
acc_transpose_blocks_kernels 9144 16.5 0.117 0.118 0.804 0.915
mp_alltoall_z22v 1201 16.6 0.830 0.846 0.830 0.846
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=41.910000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=703.181818, yerr=13.009215
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 501.035008E+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 1252937.
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 0.013 0.053 82.479 82.480
qs_mol_dyn_low 1 2.0 0.003 0.003 81.730 81.740
qs_forces 11 3.9 0.003 0.003 81.662 81.664
qs_energies 11 4.9 0.001 0.002 78.781 78.797
scf_env_do_scf 11 5.9 0.000 0.001 69.198 69.201
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 63.555 63.556
dbcsr_multiply_generic 2055 12.4 0.106 0.109 50.572 50.993
qs_scf_new_mos 99 7.5 0.000 0.001 46.107 46.237
qs_scf_loop_do_ot 99 8.5 0.001 0.001 46.107 46.237
multiply_cannon 2055 13.4 0.181 0.186 42.846 43.985
ot_scf_mini 99 9.5 0.002 0.002 43.811 43.897
multiply_cannon_loop 2055 14.4 1.522 1.556 41.874 43.086
velocity_verlet 10 3.0 0.001 0.002 42.279 42.280
ot_mini 99 10.5 0.001 0.001 25.766 25.847
multiply_cannon_multrec 49320 15.4 12.602 13.254 17.934 19.078
qs_ot_get_derivative 99 11.5 0.001 0.001 18.976 19.053
rebuild_ks_matrix 110 8.3 0.000 0.001 14.651 14.835
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.651 14.835
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.952 13.138
mp_waitall_1 220248 16.4 10.435 11.574 10.435 11.574
multiply_cannon_sync_h2d 49320 15.4 10.491 11.102 10.491 11.102
qs_ot_get_p 110 10.4 0.001 0.001 9.712 9.817
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.255 7.707
apply_single 110 13.6 0.000 0.001 7.254 7.707
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.063 7.609
multiply_cannon_metrocomm3 49320 15.4 0.080 0.083 6.137 7.516
init_scf_run 11 5.9 0.000 0.001 7.506 7.506
scf_env_initial_rho_setup 11 6.9 0.001 0.001 7.505 7.506
sum_up_and_integrate 110 10.3 0.036 0.043 7.112 7.127
integrate_v_rspace 110 11.3 0.003 0.004 7.075 7.099
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.662 6.797
calculate_rho_elec 110 8.6 0.021 0.026 6.662 6.796
qs_ot_p2m_diag 48 11.0 0.027 0.133 6.626 6.650
ot_diis_step 99 11.5 0.005 0.005 6.608 6.608
cp_dbcsr_syevd 48 12.0 0.003 0.008 5.842 5.856
dbcsr_mm_accdrv_process 87628 16.1 1.885 2.029 5.205 5.848
init_scf_loop 11 6.9 0.000 0.000 5.612 5.612
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.164 5.196
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.145 5.195
cp_fm_diag_elpa_base 48 14.0 5.146 5.177 5.157 5.189
calculate_dm_sparse 110 9.5 0.001 0.001 4.390 4.478
rs_pw_transfer 902 11.9 0.012 0.013 3.731 4.388
mp_sum_l 6594 12.7 3.440 4.159 3.440 4.159
wfi_extrapolate 11 7.9 0.001 0.001 4.050 4.050
density_rs2pw 110 9.6 0.005 0.005 3.420 4.016
make_m2s 4110 13.4 0.060 0.066 3.845 3.975
make_images 4110 14.4 0.177 0.190 3.749 3.882
multiply_cannon_metrocomm1 49320 15.4 0.062 0.065 2.436 3.834
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.553 3.557
grid_integrate_task_list 110 12.3 3.237 3.429 3.237 3.429
jit_kernel_multiply 13 15.9 3.047 3.367 3.047 3.367
calculate_first_density_matrix 1 7.0 0.000 0.000 3.351 3.357
prepare_preconditioner 11 7.9 0.000 0.000 3.298 3.311
make_preconditioner 11 8.9 0.000 0.000 3.298 3.311
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.217 3.269
pw_transfer 1331 11.6 0.056 0.067 3.121 3.201
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.069 3.117
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.032 3.114
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.021 3.065
fft_wrap_pw1pw2_140 451 13.1 0.458 0.503 2.583 2.683
potential_pw2rs 110 12.3 0.006 0.007 2.637 2.661
acc_transpose_blocks 49320 15.4 0.206 0.217 2.220 2.644
mp_alltoall_d11v 2046 13.8 2.065 2.583 2.065 2.583
mp_waitany 14300 13.8 1.810 2.507 1.810 2.507
grid_collocate_task_list 110 9.6 2.155 2.455 2.155 2.455
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 2.371 2.393
fft3d_ps 1111 14.6 0.782 0.865 2.218 2.271
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.949 1.968
mp_sum_d 3889 11.9 1.442 1.959 1.442 1.959
make_images_data 4110 15.4 0.043 0.046 1.775 1.920
hybrid_alltoall_any 4261 16.3 0.083 0.486 1.553 1.836
cp_fm_cholesky_invert 11 10.9 1.795 1.798 1.795 1.798
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=82.480000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=475.090909, yerr=2.502891
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 587.386880E+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 1426945.
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 0.038 0.129 69.584 69.585
qs_mol_dyn_low 1 2.0 0.003 0.005 69.114 69.125
qs_forces 11 3.9 0.004 0.010 68.779 68.780
qs_energies 11 4.9 0.003 0.013 65.416 65.420
scf_env_do_scf 11 5.9 0.001 0.001 56.143 56.147
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 48.755 48.756
dbcsr_multiply_generic 2055 12.4 0.145 0.167 37.473 37.630
velocity_verlet 10 3.0 0.002 0.003 36.080 36.087
qs_scf_new_mos 99 7.5 0.001 0.001 32.491 32.605
qs_scf_loop_do_ot 99 8.5 0.001 0.001 32.490 32.605
multiply_cannon 2055 13.4 0.221 0.243 30.983 32.173
ot_scf_mini 99 9.5 0.003 0.004 30.850 30.947
multiply_cannon_loop 2055 14.4 0.930 0.949 29.695 30.552
ot_mini 99 10.5 0.001 0.003 17.975 18.081
multiply_cannon_multrec 24660 15.4 7.628 9.705 13.860 15.915
rebuild_ks_matrix 110 8.3 0.000 0.000 13.765 13.874
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 13.765 13.874
qs_ot_get_derivative 99 11.5 0.001 0.002 12.195 12.295
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.100 12.195
mp_waitall_1 176588 16.5 7.492 10.219 7.492 10.219
multiply_cannon_sync_h2d 24660 15.4 7.009 8.229 7.009 8.229
multiply_cannon_metrocomm3 24660 15.4 0.068 0.072 5.126 7.548
init_scf_loop 11 6.9 0.001 0.003 7.354 7.355
apply_preconditioner_dbcsr 110 12.6 0.000 0.001 6.480 7.230
apply_single 110 13.6 0.000 0.001 6.479 7.230
init_scf_run 11 5.9 0.000 0.001 6.682 6.683
scf_env_initial_rho_setup 11 6.9 0.001 0.010 6.682 6.683
sum_up_and_integrate 110 10.3 0.053 0.059 6.580 6.593
integrate_v_rspace 110 11.3 0.003 0.003 6.527 6.543
qs_ot_get_p 110 10.4 0.001 0.001 6.256 6.399
dbcsr_mm_accdrv_process 52282 16.1 4.664 5.840 6.070 6.374
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.152 6.167
calculate_rho_elec 110 8.6 0.040 0.048 6.151 6.166
ot_diis_step 99 11.5 0.010 0.012 5.733 5.733
prepare_preconditioner 11 7.9 0.000 0.001 5.320 5.337
make_preconditioner 11 8.9 0.000 0.001 5.320 5.337
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.612 5.329
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.921 5.073
make_m2s 4110 13.4 0.057 0.060 4.086 4.598
make_images 4110 14.4 0.403 0.445 3.977 4.485
qs_ot_p2m_diag 48 11.0 0.029 0.044 4.428 4.447
cp_dbcsr_syevd 48 12.0 0.003 0.004 4.004 4.004
density_rs2pw 110 9.6 0.004 0.005 3.228 3.748
pw_transfer 1331 11.6 0.067 0.073 3.584 3.715
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 3.477 3.607
rs_pw_transfer 902 11.9 0.013 0.014 3.018 3.571
wfi_extrapolate 11 7.9 0.001 0.001 3.501 3.501
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.451 3.467
cp_fm_diag_elpa_base 48 14.0 3.403 3.420 3.448 3.463
grid_integrate_task_list 110 12.3 3.165 3.337 3.165 3.337
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.263 3.265
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.118 3.172
fft_wrap_pw1pw2_140 451 13.1 0.524 0.537 2.996 3.132
calculate_first_density_matrix 1 7.0 0.000 0.002 3.081 3.085
calculate_dm_sparse 110 9.5 0.001 0.001 3.002 3.034
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.935 2.964
make_images_data 4110 15.4 0.047 0.051 2.302 2.801
hybrid_alltoall_any 4261 16.3 0.103 0.447 2.020 2.775
fft3d_ps 1111 14.6 1.107 1.330 2.446 2.594
grid_collocate_task_list 110 9.6 2.101 2.532 2.101 2.532
cp_fm_cholesky_invert 11 10.9 2.483 2.489 2.483 2.489
mp_sum_l 6594 12.7 1.811 2.415 1.811 2.415
potential_pw2rs 110 12.3 0.008 0.009 2.360 2.369
qs_energies_init_hamiltonians 11 5.9 0.001 0.006 2.083 2.084
mp_alltoall_d11v 2046 13.8 1.687 2.037 1.687 2.037
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.926 1.945
mp_waitany 10164 13.8 1.348 1.919 1.348 1.919
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.822 1.833
rs_pw_transfer_RS2PW_140 121 11.5 0.206 0.217 1.099 1.653
mp_allgather_i34 2055 14.4 0.615 1.619 0.615 1.619
acc_transpose_blocks 24660 15.4 0.114 0.117 1.533 1.615
jit_kernel_multiply 8 16.2 1.054 1.611 1.054 1.611
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.593 1.605
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.002 1.388 1.492
multiply_cannon_metrocomm4 22605 15.4 0.078 0.082 0.792 1.446
make_basis_sm 11 9.8 0.000 0.000 1.428 1.431
parallel_gemm_fm 81 9.0 0.000 0.000 1.392 1.399
parallel_gemm_fm_cosma 81 10.0 1.391 1.398 1.391 1.398
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=69.585000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=555.727273, yerr=7.460087
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 660.152320E+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 0.018 0.028 59.847 59.848
qs_mol_dyn_low 1 2.0 0.003 0.003 59.583 59.592
qs_forces 11 3.9 0.003 0.003 59.510 59.511
qs_energies 11 4.9 0.001 0.001 56.266 56.269
scf_env_do_scf 11 5.9 0.000 0.001 48.193 48.193
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 39.800 39.801
velocity_verlet 10 3.0 0.001 0.002 32.261 32.263
dbcsr_multiply_generic 2055 12.4 0.106 0.111 28.621 28.855
qs_scf_new_mos 99 7.5 0.001 0.001 24.784 24.867
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.784 24.867
multiply_cannon 2055 13.4 0.213 0.224 22.596 23.754
ot_scf_mini 99 9.5 0.002 0.003 23.571 23.675
multiply_cannon_loop 2055 14.4 0.622 0.637 21.414 22.461
ot_mini 99 10.5 0.001 0.001 13.674 13.778
rebuild_ks_matrix 110 8.3 0.000 0.000 12.360 12.481
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 12.359 12.481
multiply_cannon_multrec 16440 15.4 3.980 5.169 9.817 11.080
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.855 10.965
mp_waitall_1 139946 16.5 7.061 10.462 7.061 10.462
qs_ot_get_derivative 99 11.5 0.001 0.001 9.296 9.397
init_scf_loop 11 6.9 0.000 0.000 8.360 8.361
multiply_cannon_metrocomm3 16440 15.4 0.042 0.045 4.503 7.740
prepare_preconditioner 11 7.9 0.000 0.000 6.593 6.615
make_preconditioner 11 8.9 0.000 0.000 6.593 6.615
sum_up_and_integrate 110 10.3 0.061 0.061 6.536 6.552
integrate_v_rspace 110 11.3 0.003 0.003 6.475 6.491
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.951 6.290
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.013 6.022
calculate_rho_elec 110 8.6 0.059 0.060 6.013 6.021
dbcsr_mm_accdrv_process 34862 16.1 4.748 5.507 5.689 5.981
init_scf_run 11 5.9 0.000 0.001 5.685 5.685
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.684 5.685
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.904 5.322
apply_single 110 13.6 0.000 0.000 4.903 5.322
qs_ot_get_p 110 10.4 0.001 0.001 5.132 5.254
make_m2s 4110 13.4 0.050 0.051 4.009 4.398
multiply_cannon_sync_h2d 16440 15.4 3.727 4.356 3.727 4.356
ot_diis_step 99 11.5 0.010 0.011 4.352 4.352
make_images 4110 14.4 0.393 0.512 3.893 4.279
density_rs2pw 110 9.6 0.004 0.005 3.063 4.260
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.481 4.112
rs_pw_transfer 902 11.9 0.011 0.012 2.621 3.783
pw_transfer 1331 11.6 0.066 0.072 3.619 3.630
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.586 3.589
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.512 3.524
grid_integrate_task_list 110 12.3 3.191 3.422 3.191 3.422
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.257 3.257
fft_wrap_pw1pw2_140 451 13.1 0.640 0.648 3.050 3.065
wfi_extrapolate 11 7.9 0.001 0.001 2.949 2.949
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.880 2.881
make_images_data 4110 15.4 0.043 0.047 2.315 2.841
hybrid_alltoall_any 4261 16.3 0.106 0.376 2.055 2.829
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.736 2.747
cp_fm_diag_elpa_base 48 14.0 2.669 2.697 2.735 2.745
calculate_dm_sparse 110 9.5 0.001 0.001 2.622 2.653
calculate_first_density_matrix 1 7.0 0.000 0.000 2.642 2.642
cp_fm_cholesky_invert 11 10.9 2.550 2.556 2.550 2.556
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.507 2.556
grid_collocate_task_list 110 9.6 2.128 2.526 2.128 2.526
mp_waitany 17072 13.8 1.178 2.397 1.178 2.397
fft3d_ps 1111 14.6 1.091 1.100 2.330 2.335
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.265 2.323
multiply_cannon_metrocomm4 14385 15.4 0.045 0.048 0.850 2.315
potential_pw2rs 110 12.3 0.011 0.012 2.247 2.251
mp_irecv_dv 48980 15.7 0.779 2.194 0.779 2.194
rs_pw_transfer_RS2PW_140 121 11.5 0.177 0.180 0.908 2.067
mp_alltoall_d11v 2046 13.8 1.742 2.049 1.742 2.049
mp_sum_l 6594 12.7 1.397 1.997 1.397 1.997
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.965 1.965
dbcsr_complete_redistribute 325 12.2 0.343 0.398 1.431 1.880
cp_fm_upper_to_full 70 14.2 1.371 1.742 1.371 1.742
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.698 1.711
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.589 1.601
acc_transpose_blocks 16440 15.4 0.075 0.078 1.346 1.584
cp_fm_cholesky_decompose 22 10.9 1.560 1.579 1.560 1.579
mp_allgather_i34 2055 14.4 0.501 1.526 0.501 1.526
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.369 1.507
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.980 1.413
jit_kernel_multiply 6 16.8 0.552 1.265 0.552 1.265
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.228 1.237
rs_gather_matrices 110 12.3 0.233 0.261 0.956 1.205
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=59.848000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=625.818182, yerr=8.526090
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 731.643904E+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 0.013 0.030 65.036 65.036
qs_mol_dyn_low 1 2.0 0.003 0.003 64.814 64.826
qs_forces 11 3.9 0.003 0.004 64.749 64.750
qs_energies 11 4.9 0.001 0.002 61.297 61.301
scf_env_do_scf 11 5.9 0.000 0.001 52.792 52.795
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 41.385 41.385
velocity_verlet 10 3.0 0.001 0.002 36.614 36.616
dbcsr_multiply_generic 2055 12.4 0.113 0.118 29.935 30.133
qs_scf_new_mos 99 7.5 0.001 0.001 26.329 26.441
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.329 26.441
ot_scf_mini 99 9.5 0.002 0.003 24.722 24.825
multiply_cannon 2055 13.4 0.244 0.260 23.068 24.452
multiply_cannon_loop 2055 14.4 0.897 0.912 21.682 22.268
ot_mini 99 10.5 0.001 0.001 14.251 14.359
multiply_cannon_multrec 24660 15.4 4.237 6.702 12.721 13.952
rebuild_ks_matrix 110 8.3 0.000 0.000 12.222 12.352
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.222 12.352
init_scf_loop 11 6.9 0.000 0.000 11.369 11.369
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.768 10.882
qs_ot_get_derivative 99 11.5 0.001 0.001 10.106 10.205
prepare_preconditioner 11 7.9 0.000 0.000 9.629 9.643
make_preconditioner 11 8.9 0.000 0.000 9.629 9.643
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.864 9.319
dbcsr_mm_accdrv_process 52304 16.0 7.098 8.347 8.337 9.268
sum_up_and_integrate 110 10.3 0.067 0.071 6.493 6.507
integrate_v_rspace 110 11.3 0.003 0.003 6.425 6.440
mp_waitall_1 121746 16.5 4.436 6.325 4.436 6.325
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.130 6.139
calculate_rho_elec 110 8.6 0.078 0.082 6.129 6.138
init_scf_run 11 5.9 0.000 0.002 5.722 5.722
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.721 5.722
make_m2s 4110 13.4 0.059 0.062 5.345 5.711
qs_ot_get_p 110 10.4 0.001 0.001 5.441 5.583
make_images 4110 14.4 0.580 0.704 5.205 5.568
cp_fm_upper_to_full 70 14.2 3.317 4.814 3.317 4.814
ot_diis_step 99 11.5 0.011 0.011 4.111 4.111
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.003 4.083
apply_single 110 13.6 0.000 0.000 4.002 4.082
pw_transfer 1331 11.6 0.066 0.075 3.711 3.755
dbcsr_complete_redistribute 325 12.2 0.423 0.460 2.610 3.752
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.730 3.745
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.646 3.691
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.604 3.653
density_rs2pw 110 9.6 0.004 0.004 2.997 3.523
multiply_cannon_metrocomm3 24660 15.4 0.036 0.037 1.620 3.434
grid_integrate_task_list 110 12.3 3.266 3.421 3.266 3.421
multiply_cannon_sync_h2d 24660 15.4 3.169 3.352 3.169 3.352
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.259 3.260
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.096 3.223
fft_wrap_pw1pw2_140 451 13.1 0.668 0.690 3.107 3.156
make_images_data 4110 15.4 0.046 0.050 2.751 3.096
hybrid_alltoall_any 4261 16.3 0.120 0.457 2.266 3.089
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.056 3.058
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.949 3.006
wfi_extrapolate 11 7.9 0.001 0.001 2.957 2.957
calculate_dm_sparse 110 9.5 0.001 0.001 2.891 2.917
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.757 2.870
rs_pw_transfer 902 11.9 0.010 0.011 2.328 2.857
mp_alltoall_i22 605 13.7 1.634 2.788 1.634 2.788
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.743 2.754
cp_fm_diag_elpa_base 48 14.0 2.597 2.655 2.740 2.752
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.710 2.751
calculate_first_density_matrix 1 7.0 0.000 0.000 2.665 2.668
cp_fm_cholesky_invert 11 10.9 2.554 2.562 2.554 2.562
grid_collocate_task_list 110 9.6 2.222 2.503 2.222 2.503
fft3d_ps 1111 14.6 1.085 1.139 2.365 2.395
jit_kernel_multiply 12 15.7 0.906 2.309 0.906 2.309
qs_energies_init_hamiltonians 11 5.9 0.017 0.024 2.268 2.272
potential_pw2rs 110 12.3 0.013 0.014 2.126 2.132
acc_transpose_blocks 24660 15.4 0.107 0.108 1.939 2.126
mp_alltoall_d11v 2046 13.8 1.717 2.064 1.717 2.064
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.616 1.718
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.678 1.714
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.695 1.708
mp_allgather_i34 2055 14.4 0.485 1.667 0.485 1.667
cp_fm_cholesky_decompose 22 10.9 1.594 1.647 1.594 1.647
multiply_cannon_metrocomm4 20550 15.4 0.060 0.063 0.861 1.544
mp_waitany 13376 13.8 1.050 1.538 1.050 1.538
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.519 1.534
mp_irecv_dv 62702 16.1 0.759 1.462 0.759 1.462
acc_transpose_blocks_kernels 24660 16.4 0.304 0.310 1.277 1.458
mp_sum_l 6594 12.7 0.833 1.409 0.833 1.409
rs_pw_transfer_RS2PW_140 121 11.5 0.169 0.176 0.816 1.355
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=65.036000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=690.363636, yerr=8.813963
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 835.297280E+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 0.013 0.030 55.644 55.645
qs_mol_dyn_low 1 2.0 0.003 0.003 55.411 55.421
qs_forces 11 3.9 0.003 0.003 55.309 55.310
qs_energies 11 4.9 0.001 0.001 51.596 51.602
scf_env_do_scf 11 5.9 0.000 0.001 42.752 42.753
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 35.038 35.039
velocity_verlet 10 3.0 0.023 0.046 31.005 31.008
dbcsr_multiply_generic 2055 12.4 0.103 0.108 23.137 23.243
qs_scf_new_mos 99 7.5 0.001 0.001 19.974 20.039
qs_scf_loop_do_ot 99 8.5 0.001 0.001 19.973 20.038
multiply_cannon 2055 13.4 0.248 0.265 17.792 19.093
ot_scf_mini 99 9.5 0.002 0.002 18.744 18.801
multiply_cannon_loop 2055 14.4 0.323 0.335 16.475 16.697
rebuild_ks_matrix 110 8.3 0.000 0.000 12.026 12.066
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 12.025 12.066
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.678 10.712
ot_mini 99 10.5 0.001 0.001 10.247 10.301
multiply_cannon_multrec 8220 15.4 3.183 4.528 8.127 9.297
mp_waitall_1 103326 16.6 5.829 7.715 5.829 7.715
init_scf_loop 11 6.9 0.000 0.000 7.668 7.670
sum_up_and_integrate 110 10.3 0.080 0.082 6.523 6.537
qs_ot_get_derivative 99 11.5 0.001 0.001 6.426 6.481
integrate_v_rspace 110 11.3 0.003 0.003 6.443 6.456
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.309 6.319
calculate_rho_elec 110 8.6 0.115 0.116 6.308 6.319
prepare_preconditioner 11 7.9 0.000 0.000 5.912 5.919
make_preconditioner 11 8.9 0.000 0.000 5.912 5.919
dbcsr_mm_accdrv_process 17442 15.9 3.181 4.254 4.815 5.744
init_scf_run 11 5.9 0.000 0.001 5.720 5.720
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.720 5.720
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.508 5.574
qs_ot_get_p 110 10.4 0.001 0.001 4.584 4.645
make_m2s 4110 13.4 0.039 0.040 4.136 4.395
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 3.010 4.393
make_images 4110 14.4 0.638 0.702 4.006 4.267
pw_transfer 1331 11.6 0.066 0.072 3.935 3.955
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.827 3.850
ot_diis_step 99 11.5 0.012 0.012 3.802 3.802
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.726 3.769
apply_single 110 13.6 0.000 0.000 3.726 3.769
grid_integrate_task_list 110 12.3 3.355 3.560 3.355 3.560
density_rs2pw 110 9.6 0.004 0.004 3.015 3.387
fft_wrap_pw1pw2_140 451 13.1 0.834 0.850 3.311 3.350
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.281 3.284
multiply_cannon_sync_h2d 8220 15.4 2.900 3.074 2.900 3.074
cp_dbcsr_syevd 48 12.0 0.003 0.003 2.980 2.981
calculate_dm_sparse 110 9.5 0.001 0.001 2.901 2.948
calculate_first_density_matrix 1 7.0 0.000 0.000 2.902 2.902
cp_fm_cholesky_invert 11 10.9 2.848 2.852 2.848 2.852
make_images_data 4110 15.4 0.038 0.044 2.335 2.755
wfi_extrapolate 11 7.9 0.001 0.001 2.715 2.715
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.673 2.673
hybrid_alltoall_any 4261 16.3 0.200 0.863 2.264 2.634
grid_collocate_task_list 110 9.6 2.315 2.571 2.315 2.571
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.557 2.558
rs_pw_transfer 902 11.9 0.010 0.011 2.150 2.543
jit_kernel_multiply 10 15.9 1.322 2.532 1.322 2.532
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.469 2.476
cp_fm_diag_elpa_base 48 14.0 2.412 2.434 2.467 2.474
fft3d_ps 1111 14.6 1.145 1.189 2.372 2.391
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.220 2.234
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.118 2.148
potential_pw2rs 110 12.3 0.015 0.016 2.050 2.053
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.788 2.007
mp_alltoall_d11v 2046 13.8 1.605 1.946 1.605 1.946
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.873 1.884
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.737 1.764
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.676 1.682
cp_fm_cholesky_decompose 22 10.9 1.640 1.656 1.640 1.656
qs_env_update_s_mstruct 11 6.9 0.001 0.005 1.494 1.615
mp_allgather_i34 2055 14.4 0.502 1.581 0.502 1.581
dbcsr_complete_redistribute 325 12.2 0.564 0.591 1.418 1.499
mp_waitany 9240 13.8 1.057 1.478 1.057 1.478
qs_create_task_list 11 7.9 0.001 0.001 1.210 1.316
generate_qs_task_list 11 8.9 0.373 0.440 1.210 1.316
multiply_cannon_metrocomm4 6165 15.4 0.018 0.019 0.480 1.299
rs_gather_matrices 110 12.3 0.328 0.376 0.950 1.263
mp_irecv_dv 24056 15.7 0.455 1.257 0.455 1.257
rs_pw_transfer_RS2PW_140 121 11.5 0.163 0.166 0.802 1.203
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.118 1.152
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=55.645000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=788.545455, yerr=10.316480
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 1.375048E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67098.
MP_Allreduce 9752 812.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.034 88.022 88.022
qs_mol_dyn_low 1 2.0 0.003 0.003 87.688 87.712
qs_forces 11 3.9 0.003 0.003 87.618 87.618
qs_energies 11 4.9 0.001 0.001 83.388 83.390
scf_env_do_scf 11 5.9 0.000 0.001 73.175 73.176
velocity_verlet 10 3.0 0.002 0.006 55.993 55.999
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 44.796 44.797
dbcsr_multiply_generic 2055 12.4 0.118 0.123 28.868 29.016
init_scf_loop 11 6.9 0.000 0.000 28.306 28.307
qs_scf_new_mos 99 7.5 0.001 0.001 26.472 26.556
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.472 26.556
prepare_preconditioner 11 7.9 0.000 0.000 26.280 26.295
make_preconditioner 11 8.9 0.000 0.000 26.280 26.295
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.425 25.735
ot_scf_mini 99 9.5 0.002 0.002 24.720 24.799
multiply_cannon 2055 13.4 0.349 0.381 21.767 22.486
multiply_cannon_loop 2055 14.4 0.342 0.346 19.958 20.351
cp_fm_upper_to_full 70 14.2 12.565 18.125 12.565 18.125
rebuild_ks_matrix 110 8.3 0.000 0.001 14.161 14.252
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 14.161 14.251
ot_mini 99 10.5 0.001 0.001 13.811 13.887
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.805 12.886
dbcsr_complete_redistribute 325 12.2 1.022 1.045 7.430 10.659
multiply_cannon_multrec 8220 15.4 4.376 4.568 9.667 9.889
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.405 9.625
qs_ot_get_derivative 99 11.5 0.001 0.001 9.122 9.201
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.840 9.037
mp_alltoall_i22 605 13.7 5.474 8.720 5.474 8.720
mp_waitall_1 84994 16.7 7.614 8.622 7.614 8.622
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.766 7.801
calculate_rho_elec 110 8.6 0.226 0.227 7.765 7.800
sum_up_and_integrate 110 10.3 0.150 0.151 7.376 7.390
integrate_v_rspace 110 11.3 0.004 0.004 7.226 7.240
init_scf_run 11 5.9 0.000 0.001 5.943 5.943
scf_env_initial_rho_setup 11 6.9 0.001 0.004 5.943 5.943
qs_ot_get_p 110 10.4 0.001 0.001 5.730 5.820
make_m2s 4110 13.4 0.043 0.043 5.255 5.782
make_images 4110 14.4 0.877 0.927 5.067 5.593
dbcsr_mm_accdrv_process 11614 15.7 3.311 3.782 5.150 5.516
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.771 5.445
apply_single 110 13.6 0.000 0.000 4.771 5.445
cp_fm_cholesky_invert 11 10.9 5.338 5.342 5.338 5.342
multiply_cannon_metrocomm3 8220 15.4 0.018 0.019 4.851 5.199
pw_transfer 1331 11.6 0.075 0.076 5.034 5.041
fft_wrap_pw1pw2 1111 12.6 0.009 0.010 4.917 4.923
ot_diis_step 99 11.5 0.015 0.016 4.668 4.669
fft_wrap_pw1pw2_140 451 13.1 1.335 1.344 4.279 4.283
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.138 4.145
multiply_cannon_sync_h2d 8220 15.4 3.946 3.950 3.946 3.950
density_rs2pw 110 9.6 0.004 0.004 3.753 3.813
grid_integrate_task_list 110 12.3 3.693 3.800 3.693 3.800
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.694 3.695
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.661 3.661
hybrid_alltoall_any 4261 16.3 0.257 0.551 2.860 3.576
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.014 3.501
make_images_data 4110 15.4 0.041 0.044 2.860 3.472
wfi_extrapolate 11 7.9 0.001 0.001 3.350 3.350
calculate_dm_sparse 110 9.5 0.001 0.001 3.117 3.136
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.085 3.085
cp_fm_diag_elpa_base 48 14.0 2.548 2.747 3.082 3.083
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.014 3.016
fft3d_ps 1111 14.6 1.298 1.304 2.847 2.856
grid_collocate_task_list 110 9.6 2.682 2.716 2.682 2.716
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.503 2.534
calculate_first_density_matrix 1 7.0 0.000 0.000 2.469 2.471
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.391 2.432
potential_pw2rs 110 12.3 0.021 0.022 2.419 2.426
rs_pw_transfer 902 11.9 0.011 0.011 2.303 2.349
qs_env_update_s_mstruct 11 6.9 0.001 0.001 2.165 2.219
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.115 2.189
mp_alltoall_d11v 2046 13.8 1.980 2.054 1.980 2.054
cp_fm_cholesky_decompose 22 10.9 2.000 2.013 2.000 2.013
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.956 1.965
jit_kernel_multiply 10 15.3 1.634 1.939 1.634 1.939
qs_create_task_list 11 7.9 0.001 0.001 1.881 1.935
generate_qs_task_list 11 8.9 0.729 0.782 1.881 1.934
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.774 1.801
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=88.022000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1227.272727, yerr=62.123178
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 629.407744E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175954870160
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 57905.
MP_Allreduce 11059 797.
MP_Sync 87
MP_Alltoall 2226 2204335.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.068 206.659 206.660
qs_mol_dyn_low 1 2.0 0.003 0.004 206.163 206.179
qs_forces 11 3.9 0.011 0.058 206.054 206.055
qs_energies 11 4.9 0.003 0.011 200.363 200.391
scf_env_do_scf 11 5.9 0.001 0.001 182.474 182.477
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 162.116 162.118
dbcsr_multiply_generic 2507 12.6 0.175 0.178 123.768 124.727
velocity_verlet 10 3.0 0.017 0.033 124.124 124.125
qs_scf_new_mos 117 7.6 0.001 0.001 121.919 122.169
qs_scf_loop_do_ot 117 8.6 0.001 0.001 121.919 122.168
ot_scf_mini 117 9.6 0.003 0.003 115.281 115.527
multiply_cannon 2507 13.6 0.239 0.249 101.374 103.238
multiply_cannon_loop 2507 14.6 2.046 2.113 99.214 100.990
ot_mini 117 10.6 0.001 0.001 64.507 64.722
multiply_cannon_multrec 60168 15.6 33.619 35.859 42.051 43.609
qs_ot_get_derivative 117 11.6 0.001 0.001 39.555 39.803
rebuild_ks_matrix 128 8.3 0.001 0.001 33.808 34.160
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 33.808 34.159
mp_waitall_1 267128 16.5 27.588 30.895 27.588 30.895
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.318 30.668
multiply_cannon_sync_h2d 60168 15.6 28.055 29.626 28.055 29.626
qs_ot_get_p 128 10.4 0.001 0.001 28.862 29.219
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.469 25.012
apply_single 128 13.6 0.001 0.001 24.469 25.012
ot_diis_step 117 11.6 0.007 0.008 24.717 24.718
qs_ot_p2m_diag 83 11.4 0.084 0.131 22.264 22.351
init_scf_loop 11 6.9 0.000 0.001 20.284 20.285
cp_dbcsr_syevd 83 12.4 0.005 0.006 19.815 19.821
multiply_cannon_metrocomm3 60168 15.6 0.115 0.120 15.511 18.277
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 17.766 17.947
cp_fm_diag_elpa 83 13.4 0.000 0.001 16.796 16.833
cp_fm_diag_elpa_base 83 14.4 16.724 16.769 16.791 16.827
prepare_preconditioner 11 7.9 0.000 0.000 15.687 15.731
make_preconditioner 11 8.9 0.000 0.000 15.687 15.731
make_full_inverse_cholesky 11 9.9 0.000 0.000 14.926 15.098
sum_up_and_integrate 128 10.3 0.089 0.108 14.294 14.309
integrate_v_rspace 128 11.3 0.004 0.005 14.204 14.224
make_m2s 5014 13.6 0.106 0.114 13.762 14.103
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.861 13.937
calculate_rho_elec 128 8.7 0.046 0.065 13.860 13.936
make_images 5014 14.6 0.403 0.423 13.582 13.934
init_scf_run 11 5.9 0.000 0.002 13.165 13.166
scf_env_initial_rho_setup 11 6.9 0.001 0.001 13.165 13.166
density_rs2pw 128 9.7 0.006 0.008 7.150 10.433
rs_pw_transfer 1046 11.9 0.017 0.019 5.933 9.157
wfi_extrapolate 11 7.9 0.001 0.001 9.155 9.155
cp_fm_cholesky_invert 11 10.9 8.915 8.923 8.915 8.923
calculate_dm_sparse 128 9.5 0.001 0.001 8.751 8.858
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.459 8.595
dbcsr_mm_accdrv_process 124484 16.2 2.984 3.184 7.989 8.519
mp_sum_l 7950 12.9 7.366 8.271 7.366 8.271
pw_transfer 1547 11.6 0.075 0.087 7.772 7.946
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.836 7.906
fft_wrap_pw1pw2 1291 12.7 0.011 0.013 7.568 7.747
make_images_data 5014 15.6 0.066 0.071 6.752 7.676
grid_integrate_task_list 128 12.3 7.080 7.570 7.080 7.570
multiply_cannon_metrocomm1 60168 15.6 0.090 0.094 5.563 7.189
hybrid_alltoall_any 5200 16.5 0.293 2.266 5.907 7.163
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 7.055 7.065
fft_wrap_pw1pw2_140 523 13.2 1.280 1.330 6.630 6.818
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 6.126 6.233
mp_waitany 16020 13.9 2.793 6.068 2.793 6.068
grid_collocate_task_list 128 9.7 4.685 5.928 4.685 5.928
fft3d_ps 1291 14.7 2.134 2.774 5.325 5.625
mp_alltoall_d11v 2415 14.1 4.167 5.465 4.167 5.465
rs_pw_transfer_RS2PW_140 139 11.5 0.283 0.312 2.221 5.440
potential_pw2rs 128 12.3 0.009 0.011 4.643 4.671
cp_fm_cholesky_decompose 22 10.9 4.618 4.634 4.618 4.634
mp_sum_d 4465 12.1 3.480 4.173 3.480 4.173
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=206.660000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=596.272727, yerr=5.816761
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/16/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 2.183246E+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 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 840.159232E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100943E+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 1463834845264
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58214.
MP_Allreduce 11083 960.
MP_Sync 87
MP_Alltoall 1969 6004217.
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 0.025 0.054 190.815 190.817
qs_mol_dyn_low 1 2.0 0.003 0.004 190.372 190.386
qs_forces 11 3.9 0.006 0.022 190.290 190.291
qs_energies 11 4.9 0.005 0.018 183.348 183.360
scf_env_do_scf 11 5.9 0.001 0.001 166.801 166.811
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 134.132 134.134
velocity_verlet 10 3.0 0.001 0.002 119.840 119.842
dbcsr_multiply_generic 2507 12.6 0.188 0.193 97.467 98.711
qs_scf_new_mos 117 7.6 0.001 0.001 94.030 94.557
qs_scf_loop_do_ot 117 8.6 0.001 0.001 94.030 94.556
ot_scf_mini 117 9.6 0.004 0.004 89.205 89.809
multiply_cannon 2507 13.6 0.477 0.530 77.364 81.496
multiply_cannon_loop 2507 14.6 1.249 1.284 74.131 76.794
ot_mini 117 10.6 0.001 0.001 49.907 50.423
mp_waitall_1 214728 16.6 24.734 38.673 24.734 38.673
multiply_cannon_multrec 30084 15.6 21.862 25.984 31.490 35.884
rebuild_ks_matrix 128 8.3 0.001 0.001 33.333 33.958
qs_ks_build_kohn_sham_matrix 128 9.3 0.019 0.028 33.332 33.957
init_scf_loop 11 6.9 0.000 0.000 32.582 32.583
qs_ks_update_qs_env 128 7.6 0.001 0.001 29.933 30.494
multiply_cannon_metrocomm3 30084 15.6 0.091 0.097 15.886 29.111
qs_ot_get_derivative 117 11.6 0.001 0.002 27.692 28.284
prepare_preconditioner 11 7.9 0.000 0.000 28.180 28.245
make_preconditioner 11 8.9 0.000 0.000 28.180 28.245
make_full_inverse_cholesky 11 9.9 0.000 0.000 26.863 27.401
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.209 23.330
apply_single 128 13.6 0.001 0.001 22.209 23.330
ot_diis_step 117 11.6 0.014 0.015 22.040 22.042
qs_ot_get_p 128 10.4 0.001 0.001 21.069 21.635
multiply_cannon_sync_h2d 30084 15.6 19.131 21.327 19.131 21.327
cp_fm_cholesky_invert 11 10.9 16.473 16.486 16.473 16.486
qs_ot_p2m_diag 83 11.4 0.188 0.216 16.309 16.344
make_m2s 5014 13.6 0.092 0.101 14.371 15.893
make_images 5014 14.6 1.171 1.356 14.159 15.679
cp_dbcsr_syevd 83 12.4 0.006 0.006 15.165 15.167
sum_up_and_integrate 128 10.3 0.116 0.134 14.734 14.765
integrate_v_rspace 128 11.3 0.004 0.005 14.617 14.653
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.163 14.200
calculate_rho_elec 128 8.7 0.089 0.106 14.162 14.199
cp_fm_diag_elpa 83 13.4 0.000 0.001 12.023 12.048
cp_fm_diag_elpa_base 83 14.4 11.766 11.838 12.016 12.044
init_scf_run 11 5.9 0.000 0.001 11.630 11.631
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.630 11.631
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 10.919 11.368
make_images_data 5014 15.6 0.065 0.075 8.783 10.661
multiply_cannon_metrocomm4 27577 15.6 0.104 0.120 3.803 10.609
mp_irecv_dv 69486 16.3 3.600 10.213 3.600 10.213
density_rs2pw 128 9.7 0.007 0.007 7.462 9.829
hybrid_alltoall_any 5200 16.5 0.342 1.510 7.565 9.726
dbcsr_mm_accdrv_process 62242 16.2 4.587 5.222 9.083 9.676
pw_transfer 1547 11.6 0.086 0.096 8.817 8.867
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 8.591 8.645
wfi_extrapolate 11 7.9 0.001 0.001 8.427 8.427
rs_pw_transfer 1046 11.9 0.015 0.017 5.760 8.097
grid_integrate_task_list 128 12.3 7.166 7.689 7.166 7.689
fft_wrap_pw1pw2_140 523 13.2 1.340 1.363 7.587 7.662
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.160 6.934
cp_fm_cholesky_decompose 22 10.9 6.837 6.924 6.837 6.924
calculate_dm_sparse 128 9.5 0.001 0.001 6.606 6.760
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.101 6.110
mp_sum_l 7950 12.9 3.954 6.010 3.954 6.010
grid_collocate_task_list 128 9.7 4.782 5.946 4.782 5.946
fft3d_ps 1291 14.7 2.807 2.993 5.892 5.935
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.366 5.433
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.234 5.387
mp_alltoall_d11v 2415 14.1 4.128 5.083 4.128 5.083
mp_allgather_i34 2507 14.6 1.792 5.024 1.792 5.024
potential_pw2rs 128 12.3 0.016 0.019 4.981 4.999
mp_waitany 11748 13.9 2.530 4.897 2.530 4.897
rs_pw_transfer_RS2PW_140 139 11.5 0.352 0.378 2.138 4.442
mp_sum_d 4470 12.1 2.715 4.135 2.715 4.135
dbcsr_complete_redistribute 395 12.7 0.782 0.868 3.122 3.978
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.789 3.849
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=190.817000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=797.636364, yerr=5.398194
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/17/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 2.928533E+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 3984192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1695.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 940.265472E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1042912
MPI messages size (bytes):
total size 2.716210E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.604448E+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 281856 36943429632
131072 < size <= 4194304 660064 996105256960
4194304 < size <= 16777216 65632 931531265168
16777216 < size 28672 751619276800
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 57952.
MP_Allreduce 11127 998.
MP_Sync 87
MP_Alltoall 1712 9388896.
MP_SendRecv 7936 75008.
MP_ISendRecv 7936 75008.
MP_Wait 21820
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.041 176.608 176.609
qs_mol_dyn_low 1 2.0 0.003 0.003 175.844 175.904
qs_forces 11 3.9 0.004 0.005 175.728 175.742
qs_energies 11 4.9 0.002 0.009 169.052 169.068
scf_env_do_scf 11 5.9 0.001 0.002 153.132 153.133
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 118.356 118.356
velocity_verlet 10 3.0 0.002 0.002 112.740 112.742
dbcsr_multiply_generic 2507 12.6 0.179 0.184 81.975 82.988
qs_scf_new_mos 117 7.6 0.001 0.001 80.047 80.431
qs_scf_loop_do_ot 117 8.6 0.001 0.001 80.047 80.431
ot_scf_mini 117 9.6 0.003 0.004 75.889 76.304
multiply_cannon 2507 13.6 0.503 0.523 62.086 66.727
multiply_cannon_loop 2507 14.6 0.863 0.891 59.040 61.441
ot_mini 117 10.6 0.001 0.001 42.674 43.102
init_scf_loop 11 6.9 0.002 0.015 34.678 34.679
mp_waitall_1 170520 16.6 25.239 34.657 25.239 34.657
rebuild_ks_matrix 128 8.3 0.001 0.001 30.836 31.266
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.022 30.835 31.265
prepare_preconditioner 11 7.9 0.000 0.001 30.624 30.670
make_preconditioner 11 8.9 0.000 0.001 30.624 30.670
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.199 29.659
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.728 28.128
multiply_cannon_multrec 20056 15.6 13.457 16.617 22.331 25.549
multiply_cannon_metrocomm3 20056 15.6 0.058 0.062 15.330 24.438
qs_ot_get_derivative 117 11.6 0.002 0.002 22.697 23.117
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 20.086 21.269
apply_single 128 13.6 0.001 0.001 20.085 21.269
ot_diis_step 117 11.6 0.018 0.019 19.870 19.870
qs_ot_get_p 128 10.4 0.001 0.001 18.573 19.069
multiply_cannon_sync_h2d 20056 15.6 14.149 16.251 14.149 16.251
make_m2s 5014 13.6 0.081 0.086 14.956 15.904
make_images 5014 14.6 1.181 1.272 14.722 15.671
sum_up_and_integrate 128 10.3 0.133 0.144 14.654 14.681
integrate_v_rspace 128 11.3 0.004 0.005 14.521 14.548
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.399 14.427
calculate_rho_elec 128 8.7 0.132 0.148 14.399 14.426
qs_ot_p2m_diag 83 11.4 0.275 0.342 14.395 14.404
cp_fm_cholesky_invert 11 10.9 14.217 14.226 14.217 14.226
cp_dbcsr_syevd 83 12.4 0.007 0.015 13.385 13.395
init_scf_run 11 5.9 0.000 0.001 10.852 10.852
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.852 10.852
make_images_data 5014 15.6 0.059 0.066 9.326 10.743
cp_fm_diag_elpa 83 13.4 0.000 0.001 10.154 10.171
cp_fm_diag_elpa_base 83 14.4 9.729 9.883 10.142 10.157
hybrid_alltoall_any 5200 16.5 0.435 1.991 8.041 9.824
density_rs2pw 128 9.7 0.007 0.007 7.209 9.425
multiply_cannon_metrocomm4 17549 15.6 0.062 0.074 3.477 9.295
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 8.836 9.153
mp_irecv_dv 50230 16.2 3.354 9.050 3.354 9.050
pw_transfer 1547 11.6 0.086 0.105 8.870 8.977
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 8.646 8.756
dbcsr_mm_accdrv_process 41502 16.2 5.096 5.488 8.329 8.470
fft_wrap_pw1pw2_140 523 13.2 1.417 1.442 7.673 7.802
grid_integrate_task_list 128 12.3 7.301 7.665 7.301 7.665
wfi_extrapolate 11 7.9 0.001 0.001 7.469 7.469
rs_pw_transfer 1046 11.9 0.014 0.015 5.190 7.405
cp_fm_cholesky_decompose 22 10.9 7.263 7.300 7.263 7.300
cp_fm_upper_to_full 105 14.8 5.679 7.264 5.679 7.264
dbcsr_complete_redistribute 395 12.7 1.175 1.208 4.661 6.436
calculate_dm_sparse 128 9.5 0.001 0.001 5.983 6.087
grid_collocate_task_list 128 9.7 4.970 5.905 4.970 5.905
fft3d_ps 1291 14.7 2.738 2.956 5.761 5.837
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.523 5.533
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.486 5.257
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.494 5.167
mp_alltoall_d11v 2415 14.1 4.385 4.908 4.385 4.908
mp_allgather_i34 2507 14.6 1.553 4.764 1.553 4.764
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.624 4.762
potential_pw2rs 128 12.3 0.021 0.024 4.712 4.729
mp_waitany 11748 13.9 2.391 4.656 2.391 4.656
mp_sum_l 7950 12.9 3.112 4.579 3.112 4.579
rs_pw_transfer_RS2PW_140 139 11.5 0.330 0.353 1.954 4.169
transfer_fm_to_dbcsr 11 9.9 0.019 0.029 2.404 4.132
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.028 4.072
mp_alltoall_i22 716 14.1 1.982 3.862 1.982 3.862
qs_energies_init_hamiltonians 11 5.9 0.001 0.005 3.779 3.789
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.613 3.657
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=176.609000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=892.454545, yerr=8.897023
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022121472 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444702699520 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796573E+12 0.0% 0.0% 100.0%
flops max/rank 4.320339E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499744 0.0% 0.0% 100.0%
number of processed stacks 5927808 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.2
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.152639E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1133160
MPI messages size (bytes):
total size 2.008142E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.772161E+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 315952 35695099904
131072 < size <= 4194304 709496 778939400192
4194304 < size <= 16777216 69840 660837692480
16777216 < size 30480 532676608000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4003 58189.
MP_Allreduce 11085 1083.
MP_Sync 86
MP_Alltoall 1700 12496376.
MP_SendRecv 5842 75008.
MP_ISendRecv 5842 75008.
MP_Wait 22272
MP_ISend 14840 244848.
MP_IRecv 14840 244848.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.021 0.042 188.006 188.007
qs_mol_dyn_low 1 2.0 0.003 0.003 187.486 187.545
qs_forces 11 3.9 0.004 0.004 187.383 187.387
qs_energies 11 4.9 0.008 0.055 180.154 180.160
scf_env_do_scf 11 5.9 0.001 0.003 162.665 162.673
velocity_verlet 10 3.0 0.002 0.002 123.146 123.149
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 116.511 116.513
dbcsr_multiply_generic 2485 12.5 0.184 0.191 79.736 80.442
qs_scf_new_mos 116 7.6 0.001 0.001 79.175 79.492
qs_scf_loop_do_ot 116 8.6 0.001 0.001 79.174 79.491
ot_scf_mini 116 9.6 0.004 0.008 74.717 74.995
multiply_cannon 2485 13.5 0.559 0.600 55.428 58.090
multiply_cannon_loop 2485 14.5 1.184 1.214 51.743 53.486
init_scf_loop 11 6.9 0.002 0.012 46.035 46.036
ot_mini 116 10.6 0.001 0.001 42.202 42.480
prepare_preconditioner 11 7.9 0.000 0.000 41.896 41.920
make_preconditioner 11 8.9 0.000 0.002 41.896 41.920
make_full_inverse_cholesky 11 9.9 0.012 0.023 35.439 40.464
multiply_cannon_multrec 29820 15.5 14.128 19.583 26.082 31.013
rebuild_ks_matrix 127 8.3 0.001 0.001 29.811 30.065
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.019 29.810 30.064
mp_waitall_1 146592 16.7 17.614 28.063 17.614 28.063
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.824 27.051
qs_ot_get_derivative 116 11.6 0.001 0.002 22.719 23.003
make_m2s 4970 13.5 0.096 0.101 20.049 21.145
make_images 4970 14.5 1.933 2.150 19.745 20.839
apply_preconditioner_dbcsr 127 12.6 0.000 0.001 18.886 19.399
apply_single 127 13.6 0.001 0.001 18.886 19.399
ot_diis_step 116 11.6 0.017 0.018 19.360 19.361
qs_ot_get_p 127 10.4 0.001 0.002 18.852 19.156
cp_fm_upper_to_full 104 14.8 11.457 16.866 11.457 16.866
cp_fm_cholesky_invert 11 10.9 15.885 15.894 15.885 15.894
multiply_cannon_metrocomm3 29820 15.5 0.047 0.050 6.628 15.601
qs_ot_p2m_diag 82 11.4 0.339 0.385 14.830 14.882
sum_up_and_integrate 127 10.3 0.139 0.149 14.655 14.681
qs_rho_update_rho_low 127 7.7 0.001 0.001 14.565 14.602
calculate_rho_elec 127 8.7 0.174 0.191 14.565 14.601
integrate_v_rspace 127 11.3 0.004 0.005 14.515 14.546
cp_dbcsr_syevd 82 12.4 0.005 0.006 13.571 13.573
dbcsr_complete_redistribute 393 12.7 1.511 1.629 9.011 12.746
make_images_data 4970 15.5 0.062 0.066 10.879 12.663
multiply_cannon_sync_h2d 29820 15.5 11.635 12.347 11.635 12.347
dbcsr_mm_accdrv_process 61748 16.2 7.452 8.303 11.529 12.146
hybrid_alltoall_any 5155 16.4 0.516 2.171 9.630 11.910
init_scf_run 11 5.9 0.000 0.002 11.543 11.545
scf_env_initial_rho_setup 11 6.9 0.001 0.002 11.543 11.544
copy_fm_to_dbcsr 208 11.6 0.002 0.002 7.658 11.396
cp_fm_diag_elpa 82 13.4 0.000 0.000 10.537 10.547
cp_fm_diag_elpa_base 82 14.4 9.581 9.884 10.530 10.541
transfer_fm_to_dbcsr 11 9.9 0.001 0.003 6.434 10.070
mp_alltoall_i22 712 14.1 5.636 9.430 5.636 9.430
qs_ot_get_derivative_diag 76 12.4 0.002 0.003 9.195 9.401
pw_transfer 1535 11.6 0.087 0.102 9.053 9.126
fft_wrap_pw1pw2 1281 12.7 0.010 0.012 8.827 8.908
density_rs2pw 127 9.7 0.006 0.007 6.923 8.416
fft_wrap_pw1pw2_140 519 13.2 1.552 1.581 7.860 7.941
grid_integrate_task_list 127 12.3 7.467 7.927 7.467 7.927
wfi_extrapolate 11 7.9 0.001 0.001 7.697 7.697
cp_fm_cholesky_decompose 22 10.9 7.367 7.462 7.367 7.462
multiply_cannon_metrocomm4 24850 15.5 0.079 0.090 2.693 7.244
mp_irecv_dv 75445 16.2 2.547 6.973 2.547 6.973
calculate_dm_sparse 127 9.5 0.001 0.001 6.142 6.245
rs_pw_transfer 1038 11.9 0.013 0.014 4.642 6.211
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.147 6.203
grid_collocate_task_list 127 9.7 5.121 5.995 5.121 5.995
fft3d_ps 1281 14.7 2.814 2.894 5.770 5.846
mp_alltoall_d11v 2401 14.1 4.899 5.686 4.899 5.686
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.077 5.161
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.781 4.861
potential_pw2rs 127 12.3 0.023 0.023 4.503 4.515
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.458 4.458
qs_ot_get_orbitals 116 10.6 0.001 0.001 4.147 4.198
mp_allgather_i34 2485 14.5 1.797 4.175 1.797 4.175
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=188.007000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1088.272727, yerr=18.404994
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420243808256 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 11528908111872 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 93.514772E+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 6755942624 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 1.485824E+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 721350290592
-------------------------------------------------------------------------------
- -
- 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 4020 57947.
MP_Allreduce 11127 1165.
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 0.030 0.042 174.361 174.361
qs_mol_dyn_low 1 2.0 0.003 0.003 173.917 173.938
qs_forces 11 3.9 0.004 0.004 173.810 173.816
qs_energies 11 4.9 0.004 0.005 166.141 166.151
scf_env_do_scf 11 5.9 0.001 0.001 148.363 148.364
velocity_verlet 10 3.0 0.002 0.003 114.093 114.097
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 112.755 112.756
dbcsr_multiply_generic 2507 12.6 0.179 0.192 74.727 75.249
qs_scf_new_mos 117 7.6 0.001 0.001 74.199 74.311
qs_scf_loop_do_ot 117 8.6 0.001 0.001 74.198 74.310
ot_scf_mini 117 9.6 0.003 0.004 69.745 69.854
multiply_cannon 2507 13.6 0.583 0.616 54.610 58.918
multiply_cannon_loop 2507 14.6 0.448 0.461 49.966 51.080
ot_mini 117 10.6 0.001 0.001 39.824 39.930
init_scf_loop 11 6.9 0.000 0.000 35.462 35.463
mp_waitall_1 125778 16.7 26.331 34.343 26.331 34.343
prepare_preconditioner 11 7.9 0.000 0.000 31.518 31.541
make_preconditioner 11 8.9 0.000 0.000 31.518 31.541
rebuild_ks_matrix 128 8.3 0.001 0.001 29.917 30.046
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.018 29.917 30.046
make_full_inverse_cholesky 11 9.9 0.018 0.028 29.416 29.673
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.102 27.221
multiply_cannon_multrec 10028 15.6 10.426 14.578 17.736 22.122
multiply_cannon_metrocomm3 10028 15.6 0.023 0.025 12.640 20.166
ot_diis_step 117 11.6 0.020 0.020 20.010 20.011
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.617 19.902
apply_single 128 13.6 0.001 0.001 19.617 19.902
qs_ot_get_derivative 117 11.6 0.002 0.002 19.745 19.851
make_m2s 5014 13.6 0.067 0.072 16.469 18.874
make_images 5014 14.6 2.319 2.747 16.161 18.564
cp_fm_cholesky_invert 11 10.9 17.961 17.967 17.961 17.967
qs_ot_get_p 128 10.4 0.001 0.001 16.905 17.046
qs_rho_update_rho_low 128 7.7 0.001 0.001 15.342 15.377
calculate_rho_elec 128 8.7 0.259 0.270 15.342 15.376
sum_up_and_integrate 128 10.3 0.181 0.191 14.937 14.987
integrate_v_rspace 128 11.3 0.004 0.004 14.756 14.812
qs_ot_p2m_diag 83 11.4 0.496 0.502 13.189 13.204
make_images_data 5014 15.6 0.051 0.060 10.027 12.545
multiply_cannon_sync_h2d 10028 15.6 11.507 12.395 11.507 12.395
hybrid_alltoall_any 5200 16.5 0.824 3.832 9.849 12.098
cp_dbcsr_syevd 83 12.4 0.005 0.006 12.066 12.069
init_scf_run 11 5.9 0.001 0.006 10.641 10.642
scf_env_initial_rho_setup 11 6.9 0.004 0.006 10.641 10.641
pw_transfer 1547 11.6 0.087 0.100 9.422 9.450
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 9.196 9.222
cp_fm_diag_elpa 83 13.4 0.000 0.000 9.071 9.082
cp_fm_diag_elpa_base 83 14.4 8.831 8.892 9.064 9.076
fft_wrap_pw1pw2_140 523 13.2 1.916 1.948 8.153 8.191
grid_integrate_task_list 128 12.3 7.761 8.164 7.761 8.164
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 7.922 7.997
cp_fm_cholesky_decompose 22 10.9 7.844 7.969 7.844 7.969
density_rs2pw 128 9.7 0.006 0.006 6.828 7.743
multiply_cannon_metrocomm1 10028 15.6 0.028 0.029 4.434 7.669
wfi_extrapolate 11 7.9 0.001 0.001 7.587 7.587
dbcsr_mm_accdrv_process 20762 16.1 2.679 3.456 6.943 7.579
mp_allgather_i34 2507 14.6 2.831 7.258 2.831 7.258
calculate_dm_sparse 128 9.5 0.001 0.001 6.103 6.214
grid_collocate_task_list 128 9.7 5.465 6.149 5.465 6.149
mp_alltoall_d11v 2415 14.1 5.019 5.869 5.019 5.869
fft3d_ps 1291 14.7 2.769 2.834 5.665 5.721
qs_energies_init_hamiltonians 11 5.9 0.257 0.268 5.604 5.606
dbcsr_complete_redistribute 395 12.7 2.134 2.165 5.147 5.566
multiply_cannon_metrocomm4 7521 15.6 0.024 0.027 1.880 5.379
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.251 5.304
mp_irecv_dv 28860 15.9 1.844 5.295 1.844 5.295
rs_pw_transfer 1046 11.9 0.013 0.013 4.069 4.990
potential_pw2rs 128 12.3 0.026 0.028 4.352 4.366
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.158 4.191
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.600 3.905
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.732 3.768
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.719 3.732
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.334 3.636
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.569 3.632
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.475 3.512
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=174.361000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1383.272727, yerr=45.172580
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430460020736 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613089636352 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228663E+12 0.0% 0.0% 100.0%
flops max/rank 11.785520E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806316384 0.0% 0.0% 100.0%
number of processed stacks 1979552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3438.3
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 3.147289E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101080
MPI messages size (bytes):
total size 1.144901E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.326685E+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 45568 35366371328
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726591885696
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4024 58825.
MP_Allreduce 11137 1501.
MP_Sync 87
MP_Alltoall 1724 36993750.
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 0.025 0.048 297.456 297.457
qs_mol_dyn_low 1 2.0 0.003 0.003 296.871 296.901
qs_forces 11 3.9 0.005 0.005 296.754 296.760
qs_energies 11 4.9 0.001 0.001 287.417 287.426
scf_env_do_scf 11 5.9 0.001 0.001 264.725 264.733
velocity_verlet 10 3.0 0.002 0.003 213.550 213.558
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 141.768 141.770
init_scf_loop 11 6.9 0.000 0.000 122.709 122.713
prepare_preconditioner 11 7.9 0.000 0.000 117.686 117.709
make_preconditioner 11 8.9 0.000 0.000 117.686 117.709
make_full_inverse_cholesky 11 9.9 0.036 0.039 94.107 114.795
qs_scf_new_mos 118 7.6 0.001 0.001 90.676 90.778
qs_scf_loop_do_ot 118 8.6 0.001 0.001 90.675 90.777
ot_scf_mini 118 9.6 0.004 0.004 85.820 85.846
dbcsr_multiply_generic 2527 12.6 0.212 0.223 84.501 85.087
cp_fm_upper_to_full 105 14.8 52.081 74.588 52.081 74.588
multiply_cannon 2527 13.6 0.713 0.771 59.247 60.035
multiply_cannon_loop 2527 14.6 0.479 0.484 55.333 56.747
ot_mini 118 10.6 0.001 0.001 45.213 45.237
dbcsr_complete_redistribute 395 12.7 3.990 4.048 29.161 41.801
rebuild_ks_matrix 129 8.3 0.001 0.001 38.503 38.551
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.017 38.503 38.551
copy_fm_to_dbcsr 209 11.7 0.001 0.002 25.833 38.477
transfer_fm_to_dbcsr 11 9.9 0.028 0.034 23.533 35.948
qs_ks_update_qs_env 129 7.6 0.001 0.001 35.509 35.551
mp_alltoall_i22 718 14.1 21.366 34.099 21.366 34.099
cp_fm_cholesky_invert 11 10.9 33.002 33.009 33.002 33.009
mp_waitall_1 104500 16.8 28.795 32.525 28.795 32.525
qs_ot_get_p 129 10.4 0.001 0.001 24.958 25.017
qs_ot_get_derivative 118 11.6 0.002 0.002 24.409 24.438
make_m2s 5054 13.6 0.076 0.078 20.651 21.785
make_images 5054 14.6 3.779 3.905 20.168 21.304
qs_ot_p2m_diag 83 11.4 0.879 0.884 20.927 20.957
ot_diis_step 118 11.6 0.022 0.022 20.772 20.772
qs_rho_update_rho_low 129 7.7 0.001 0.001 20.643 20.668
calculate_rho_elec 129 8.7 0.486 0.487 20.642 20.667
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 20.009 20.205
apply_single 129 13.6 0.001 0.001 20.008 20.204
multiply_cannon_metrocomm3 10108 15.6 0.024 0.026 18.708 19.973
sum_up_and_integrate 129 10.3 0.323 0.325 19.753 19.825
integrate_v_rspace 129 11.3 0.004 0.004 19.429 19.500
cp_dbcsr_syevd 83 12.4 0.006 0.006 19.197 19.201
multiply_cannon_multrec 10108 15.6 10.652 12.551 18.295 18.396
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.029 16.029
cp_fm_diag_elpa_base 83 14.4 11.688 13.289 16.025 16.025
multiply_cannon_sync_h2d 10108 15.6 15.765 15.777 15.765 15.777
hybrid_alltoall_any 5240 16.5 1.307 3.042 11.459 13.438
make_images_data 5054 15.6 0.059 0.063 11.345 13.253
init_scf_run 11 5.9 0.000 0.001 12.663 12.663
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.663 12.663
pw_transfer 1559 11.6 0.094 0.096 11.358 11.364
fft_wrap_pw1pw2 1301 12.7 0.012 0.012 11.120 11.128
fft_wrap_pw1pw2_140 527 13.2 3.094 3.119 9.903 9.912
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.348 9.375
mp_alltoall_d11v 2423 14.1 8.433 9.366 8.433 9.366
dbcsr_mm_accdrv_process 20926 16.1 3.908 5.784 7.401 9.275
wfi_extrapolate 11 7.9 0.001 0.001 9.140 9.140
grid_integrate_task_list 129 12.3 8.648 8.836 8.648 8.836
cp_fm_cholesky_decompose 22 10.9 8.743 8.779 8.743 8.779
density_rs2pw 129 9.7 0.005 0.006 8.002 8.176
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 7.977 7.978
calculate_dm_sparse 129 9.5 0.001 0.001 6.689 6.780
grid_collocate_task_list 129 9.7 6.484 6.555 6.484 6.555
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.298 6.354
rs_scatter_matrices 140 9.7 3.704 4.823 6.121 6.350
fft3d_ps 1301 14.7 2.862 2.871 6.164 6.188
copy_dbcsr_to_fm 186 11.8 0.004 0.004 5.991 6.041
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=297.457000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2862.636364, yerr=154.787404
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 1.259389E+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 19609392.
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.034 85.980 85.981
qs_energies 1 2.0 0.000 0.000 85.287 85.294
ls_scf 1 3.0 0.000 0.000 84.193 84.199
dbcsr_multiply_generic 111 6.7 0.015 0.016 72.964 73.111
multiply_cannon 111 7.7 0.017 0.020 56.134 57.228
multiply_cannon_loop 111 8.7 0.198 0.214 52.604 54.017
ls_scf_main 1 4.0 0.000 0.000 52.526 52.526
density_matrix_trs4 2 5.0 0.002 0.003 46.954 47.055
ls_scf_init_scf 1 4.0 0.000 0.000 28.582 28.584
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.446 27.492
mp_waitall_1 11031 10.9 22.504 26.079 22.504 26.079
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.198 25.216
multiply_cannon_multrec 2664 9.7 8.237 8.965 15.679 17.422
multiply_cannon_sync_h2d 2664 9.7 13.684 15.324 13.684 15.324
make_m2s 222 7.7 0.009 0.012 13.211 13.642
make_images 222 8.7 0.101 0.111 13.189 13.623
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.610 12.430
make_images_data 222 9.7 0.005 0.006 7.782 8.421
dbcsr_mm_accdrv_process 4760 10.4 0.501 0.610 7.061 8.051
hybrid_alltoall_any 227 10.6 0.218 1.869 6.716 7.806
multiply_cannon_metrocomm3 2664 9.7 0.010 0.011 5.448 7.293
dbcsr_mm_accdrv_process_sort 4760 11.4 6.358 7.273 6.358 7.273
calculate_norms 4752 9.8 5.537 6.273 5.537 6.273
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.025 5.204
mp_sum_l 887 5.1 3.083 4.830 3.083 4.830
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.282 3.630
make_images_sizes 222 9.7 0.000 0.000 0.742 3.407
mp_alltoall_i44 222 10.7 0.741 3.407 0.741 3.407
multiply_cannon_metrocomm4 2442 9.7 0.012 0.014 2.053 3.289
mp_irecv_dv 6231 10.9 2.036 3.277 2.036 3.277
arnoldi_extremal 4 6.8 0.000 0.000 3.243 3.270
arnoldi_normal_ev 4 7.8 0.001 0.003 3.242 3.270
build_subspace 16 8.4 0.009 0.012 3.145 3.147
ls_scf_post 1 4.0 0.000 0.000 3.084 3.092
ls_scf_store_result 1 5.0 0.000 0.000 2.857 2.885
dbcsr_special_finalize 555 9.7 0.005 0.006 2.328 2.774
dbcsr_merge_single_wm 555 10.7 0.486 0.620 2.320 2.766
make_images_pack 222 9.7 2.211 2.629 2.213 2.631
dbcsr_matrix_vector_mult 304 9.0 0.006 0.013 2.323 2.571
dbcsr_matrix_vector_mult_local 304 10.0 2.070 2.471 2.072 2.473
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.364 2.465
dbcsr_sort_data 658 11.4 2.092 2.464 2.092 2.464
buffer_matrices_ensure_size 222 8.7 1.834 2.211 1.834 2.211
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.785 1.787
rebuild_ks_matrix 3 7.3 0.000 0.000 1.776 1.777
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.776 1.777
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=85.981000, 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/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 2.082251E+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 0.034 0.061 91.891 91.891
qs_energies 1 2.0 0.000 0.000 91.326 91.330
ls_scf 1 3.0 0.000 0.000 89.800 89.803
dbcsr_multiply_generic 111 6.7 0.016 0.017 75.661 76.029
multiply_cannon 111 7.7 0.030 0.041 53.584 56.872
ls_scf_main 1 4.0 0.000 0.000 55.188 55.194
multiply_cannon_loop 111 8.7 0.117 0.123 50.278 53.518
density_matrix_trs4 2 5.0 0.002 0.003 49.328 49.506
ls_scf_init_scf 1 4.0 0.000 0.000 31.047 31.049
mp_waitall_1 9105 10.9 21.269 30.256 21.269 30.256
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.829 29.925
multiply_cannon_multrec 1332 9.7 13.283 17.367 22.641 27.881
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 27.393 27.408
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.859 20.888
make_m2s 222 7.7 0.008 0.009 15.374 16.017
make_images 222 8.7 1.588 1.985 15.343 15.986
dbcsr_mm_accdrv_process 4041 10.4 0.280 0.472 8.952 10.558
dbcsr_mm_accdrv_process_sort 4041 11.4 8.537 10.119 8.537 10.119
make_images_data 222 9.7 0.004 0.005 8.839 9.771
hybrid_alltoall_any 227 10.6 0.524 2.463 8.278 9.264
mp_sum_l 887 5.1 5.386 8.690 5.386 8.690
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.249 7.758
mp_irecv_dv 3311 11.0 3.229 7.681 3.229 7.681
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.181 6.975
calculate_norms 2376 9.8 6.000 6.807 6.000 6.807
multiply_cannon_sync_h2d 1332 9.7 4.734 5.768 4.734 5.768
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.115 5.353
arnoldi_extremal 4 6.8 0.000 0.000 4.655 4.684
arnoldi_normal_ev 4 7.8 0.001 0.005 4.655 4.684
build_subspace 16 8.4 0.014 0.021 4.385 4.388
ls_scf_post 1 4.0 0.000 0.000 3.564 3.569
ls_scf_store_result 1 5.0 0.000 0.000 3.258 3.387
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.138 3.383
dbcsr_matrix_vector_mult_local 304 10.0 2.736 3.218 2.738 3.219
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.691 2.812
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.211 2.721
mp_allgather_i34 111 8.7 1.001 2.380 1.001 2.380
make_images_pack 222 9.7 2.030 2.370 2.033 2.373
dbcsr_sort_data 436 11.2 1.840 2.061 1.840 2.061
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.888 1.890
rebuild_ks_matrix 3 7.3 0.000 0.000 1.875 1.877
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.875 1.877
dbcsr_data_new 4174 10.1 1.610 1.850 1.610 1.850
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=91.891000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1711.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 2.753651E+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 0.031 0.048 94.715 94.716
qs_energies 1 2.0 0.000 0.000 94.086 94.109
ls_scf 1 3.0 0.000 0.001 92.646 92.670
dbcsr_multiply_generic 111 6.7 0.016 0.017 77.226 77.516
ls_scf_main 1 4.0 0.000 0.006 57.699 57.703
multiply_cannon 111 7.7 0.040 0.107 52.862 57.276
multiply_cannon_loop 111 8.7 0.100 0.105 49.134 53.289
density_matrix_trs4 2 5.0 0.003 0.015 51.630 51.810
mp_waitall_1 7281 11.0 23.854 33.985 23.854 33.985
ls_scf_init_scf 1 4.0 0.000 0.002 31.437 31.439
ls_scf_init_matrix_S 1 5.0 0.000 0.001 30.169 30.249
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.169 27.181
multiply_cannon_multrec 888 9.7 12.744 15.432 21.399 24.696
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 10.826 23.087
make_m2s 222 7.7 0.007 0.008 17.604 18.923
make_images 222 8.7 1.981 2.297 17.566 18.885
make_images_data 222 9.7 0.004 0.005 10.054 11.467
hybrid_alltoall_any 227 10.6 0.623 2.879 9.723 11.186
dbcsr_mm_accdrv_process 3754 10.4 0.239 0.405 8.194 9.370
mp_sum_l 887 5.1 5.199 9.004 5.199 9.004
dbcsr_mm_accdrv_process_sort 3754 11.4 7.813 8.965 7.813 8.965
multiply_cannon_sync_h2d 888 9.7 5.958 7.427 5.958 7.427
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.068 7.116
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.419 6.924
mp_irecv_dv 2335 11.1 2.404 6.880 2.404 6.880
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.601 6.797
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.518 5.718
arnoldi_extremal 4 6.8 0.000 0.000 5.057 5.078
arnoldi_normal_ev 4 7.8 0.001 0.005 5.057 5.078
calculate_norms 1584 9.8 4.309 4.791 4.309 4.791
build_subspace 16 8.4 0.014 0.020 4.748 4.756
mp_allgather_i34 111 8.7 1.523 4.399 1.523 4.399
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.426 3.746
dbcsr_matrix_vector_mult_local 304 10.0 3.005 3.571 3.007 3.573
ls_scf_post 1 4.0 0.003 0.026 3.510 3.535
ls_scf_store_result 1 5.0 0.000 0.000 3.258 3.338
ls_scf_dm_to_ks 2 5.0 0.001 0.018 2.921 3.032
make_images_pack 222 9.7 2.088 2.544 2.091 2.547
dbcsr_sort_data 325 11.1 1.904 2.157 1.904 2.157
make_images_sizes 222 9.7 0.000 0.000 0.930 2.055
mp_alltoall_i44 222 10.7 0.930 2.055 0.930 2.055
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.963 1.965
rebuild_ks_matrix 3 7.3 0.000 0.000 1.946 1.947
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.946 1.947
dbcsr_data_release 9322 10.9 1.299 1.938 1.299 1.938
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=94.716000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2165.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 3.325751E+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 0.039 0.072 98.473 98.474
qs_energies 1 2.0 0.000 0.000 97.784 97.792
ls_scf 1 3.0 0.000 0.001 96.067 96.073
dbcsr_multiply_generic 111 6.7 0.017 0.018 79.108 79.322
ls_scf_main 1 4.0 0.000 0.001 59.317 59.318
multiply_cannon 111 7.7 0.058 0.131 52.225 56.580
density_matrix_trs4 2 5.0 0.003 0.007 53.127 53.236
multiply_cannon_loop 111 8.7 0.115 0.127 47.091 50.138
ls_scf_init_scf 1 4.0 0.000 0.002 33.506 33.509
ls_scf_init_matrix_S 1 5.0 0.000 0.001 31.832 31.888
matrix_sqrt_Newton_Schulz 2 6.5 0.003 0.018 29.342 29.357
mp_waitall_1 6369 11.0 23.011 29.107 23.011 29.107
multiply_cannon_multrec 1332 9.7 14.171 17.339 22.268 25.255
make_m2s 222 7.7 0.008 0.009 21.081 22.543
make_images 222 8.7 3.146 3.623 21.030 22.494
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 9.567 17.776
make_images_data 222 9.7 0.004 0.005 11.779 13.369
hybrid_alltoall_any 227 10.6 0.801 3.769 11.085 13.034
dbcsr_mm_accdrv_process 3641 10.4 0.207 0.423 7.685 9.198
dbcsr_mm_accdrv_process_sort 3641 11.4 7.262 8.788 7.262 8.788
mp_sum_l 887 5.1 4.218 7.827 4.218 7.827
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.322 6.185
multiply_cannon_sync_h2d 1332 9.7 5.525 6.125 5.525 6.125
multiply_cannon_metrocomm4 1110 9.7 0.004 0.007 2.072 6.010
mp_irecv_dv 3229 10.9 2.049 5.933 2.049 5.933
arnoldi_extremal 4 6.8 0.000 0.000 5.240 5.263
arnoldi_normal_ev 4 7.8 0.001 0.004 5.240 5.263
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.503 5.259
build_subspace 16 8.4 0.014 0.021 4.893 4.903
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.531 4.711
mp_allgather_i34 111 8.7 2.246 4.659 2.246 4.659
calculate_norms 2376 9.8 4.199 4.543 4.199 4.543
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.567 3.895
dbcsr_matrix_vector_mult_local 304 10.0 3.161 3.657 3.163 3.659
dbcsr_sort_data 658 11.4 3.061 3.473 3.061 3.473
ls_scf_post 1 4.0 0.000 0.001 3.243 3.250
dbcsr_special_finalize 555 9.7 0.006 0.007 2.799 3.227
dbcsr_merge_single_wm 555 10.7 0.537 0.660 2.790 3.218
ls_scf_dm_to_ks 2 5.0 0.000 0.001 3.080 3.150
ls_scf_store_result 1 5.0 0.000 0.000 2.972 3.046
dbcsr_data_release 10477 10.7 1.580 2.443 1.580 2.443
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.411 2.412
rebuild_ks_matrix 3 7.3 0.000 0.000 2.387 2.388
qs_ks_build_kohn_sham_matrix 3 8.3 0.002 0.010 2.387 2.388
dbcsr_finalize 304 7.8 0.050 0.062 1.826 2.011
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=98.474000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2727.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 4.694856E+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 0.056 99.707 99.708
qs_energies 1 2.0 0.000 0.000 98.952 98.956
ls_scf 1 3.0 0.000 0.000 96.955 96.960
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.100 78.349
ls_scf_main 1 4.0 0.000 0.000 62.256 62.257
multiply_cannon 111 7.7 0.095 0.167 55.591 60.747
density_matrix_trs4 2 5.0 0.002 0.003 55.023 55.116
multiply_cannon_loop 111 8.7 0.069 0.077 50.996 52.594
mp_waitall_1 5436 11.0 26.436 32.093 26.436 32.093
ls_scf_init_scf 1 4.0 0.000 0.000 30.964 30.966
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.693 29.737
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 27.390 27.402
multiply_cannon_multrec 444 9.7 14.055 16.579 21.237 23.248
make_m2s 222 7.7 0.006 0.006 17.656 20.217
make_images 222 8.7 3.734 4.447 17.592 20.156
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.068 16.291
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.303 15.201
make_images_data 222 9.7 0.003 0.004 9.871 12.333
hybrid_alltoall_any 227 10.6 0.791 3.775 9.680 12.124
dbcsr_mm_accdrv_process 3003 10.4 0.174 0.343 6.892 8.164
dbcsr_mm_accdrv_process_sort 3003 11.4 6.536 7.656 6.536 7.656
multiply_cannon_sync_h2d 444 9.7 6.533 7.602 6.533 7.602
mp_allgather_i34 111 8.7 2.817 7.007 2.817 7.007
arnoldi_extremal 4 6.8 0.000 0.000 5.775 5.782
arnoldi_normal_ev 4 7.8 0.002 0.005 5.775 5.782
build_subspace 16 8.4 0.015 0.020 5.380 5.388
mp_sum_l 887 5.1 2.886 5.116 2.886 5.116
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.697 4.850
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.150 4.333
dbcsr_matrix_vector_mult_local 304 10.0 3.651 4.123 3.653 4.125
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.663 3.910
mp_irecv_dv 1241 11.2 1.646 3.884 1.646 3.884
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.942 3.786
ls_scf_post 1 4.0 0.000 0.000 3.734 3.738
calculate_norms 792 9.8 3.558 3.733 3.558 3.733
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.601 3.702
make_images_sizes 222 9.7 0.000 0.000 1.119 3.511
mp_alltoall_i44 222 10.7 1.119 3.510 1.119 3.510
ls_scf_store_result 1 5.0 0.000 0.000 3.319 3.358
dbcsr_finalize 304 7.8 0.062 0.078 2.200 2.301
dbcsr_merge_all 275 8.9 0.473 0.527 2.052 2.138
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.075 2.076
rebuild_ks_matrix 3 7.3 0.000 0.000 2.042 2.044
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.042 2.044
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=99.708000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3602.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 8.772989E+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 0.144 0.239 114.462 114.462
qs_energies 1 2.0 0.000 0.000 112.320 112.333
ls_scf 1 3.0 0.000 0.000 109.395 109.408
dbcsr_multiply_generic 111 6.7 0.023 0.027 82.888 82.982
ls_scf_main 1 4.0 0.000 0.000 65.809 65.811
multiply_cannon 111 7.7 0.124 0.186 55.313 57.003
density_matrix_trs4 2 5.0 0.002 0.003 56.517 56.571
multiply_cannon_loop 111 8.7 0.068 0.070 51.929 52.952
ls_scf_init_scf 1 4.0 0.000 0.000 39.903 39.904
ls_scf_init_matrix_S 1 5.0 0.000 0.000 38.318 38.334
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 35.481 35.492
mp_waitall_1 4527 11.1 22.827 27.768 22.827 27.768
make_m2s 222 7.7 0.007 0.007 24.003 24.889
make_images 222 8.7 4.595 4.990 23.893 24.779
multiply_cannon_multrec 444 9.7 17.908 18.587 22.592 23.298
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 11.011 16.492
hybrid_alltoall_any 227 10.6 1.663 3.634 13.011 15.852
make_images_data 222 9.7 0.004 0.004 13.226 15.515
multiply_cannon_sync_h2d 444 9.7 8.840 8.882 8.840 8.882
arnoldi_extremal 4 6.8 0.000 0.000 7.307 7.319
arnoldi_normal_ev 4 7.8 0.003 0.009 7.307 7.319
build_subspace 16 8.4 0.026 0.036 6.759 6.772
compute_matrix_preconditioner 1 6.0 0.002 0.002 6.345 6.352
acc_transpose_blocks 444 9.7 0.002 0.002 5.035 5.732
acc_transpose_blocks_kernels 444 10.7 0.005 0.005 4.999 5.696
jit_kernel_transpose 1 13.0 4.994 5.691 4.994 5.691
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.394 5.531
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.249 5.342
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.024 5.254
dbcsr_matrix_vector_mult_local 304 10.0 4.935 5.227 4.938 5.230
dbcsr_mm_accdrv_process 1814 10.4 0.237 0.333 4.514 4.632
dbcsr_mm_accdrv_process_sort 1814 11.4 4.208 4.344 4.208 4.344
ls_scf_post 1 4.0 0.000 0.000 3.683 3.695
make_images_sizes 222 9.7 0.000 0.000 1.417 3.468
mp_alltoall_i44 222 10.7 1.417 3.467 1.417 3.467
ls_scf_store_result 1 5.0 0.000 0.000 3.418 3.429
calculate_norms 792 9.8 3.246 3.279 3.246 3.279
dbcsr_finalize 304 7.8 0.082 0.090 3.088 3.173
mp_allgather_i34 111 8.7 0.992 3.043 0.992 3.043
dbcsr_merge_all 275 8.9 0.891 0.918 2.873 2.952
dbcsr_complete_redistribute 5 7.6 1.496 1.554 2.827 2.943
qs_energies_init_hamiltonians 1 3.0 0.001 0.002 2.894 2.894
dbcsr_data_release 12724 10.6 2.324 2.877 2.324 2.877
matrix_ls_to_qs 2 6.0 0.000 0.000 2.454 2.576
dbcsr_sort_data 325 11.1 2.445 2.504 2.445 2.504
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.496 2.498
rebuild_ks_matrix 3 7.3 0.000 0.000 2.430 2.432
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.430 2.432
dbcsr_new_transposed 4 7.5 0.270 0.310 2.328 2.338
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=114.462000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6897.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e_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 585.805824E+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 357180720.
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 0.026 0.097 227.745 227.747
qs_mol_dyn_low 1 2.0 0.003 0.004 226.698 226.734
qs_forces 5 3.8 0.004 0.007 226.618 226.620
qs_energies 5 4.8 0.002 0.041 223.221 223.237
scf_env_do_scf 5 5.8 0.000 0.001 206.364 206.368
scf_env_do_scf_inner_loop 105 6.6 0.003 0.019 180.254 180.262
qs_scf_new_mos 105 7.6 0.000 0.001 139.946 140.104
qs_scf_loop_do_ot 105 8.6 0.001 0.001 139.945 140.104
dbcsr_multiply_generic 1445 12.2 0.126 0.134 133.399 133.837
ot_scf_mini 105 9.6 0.003 0.004 130.111 130.250
multiply_cannon 1445 13.2 0.274 0.285 115.251 117.516
multiply_cannon_loop 1445 14.2 2.347 2.471 113.488 114.640
velocity_verlet 4 3.0 0.002 0.015 104.704 104.706
ot_mini 105 10.6 0.001 0.001 59.187 59.309
multiply_cannon_multrec 69360 15.2 32.096 37.240 40.122 45.644
qs_ot_get_p 112 10.4 0.001 0.001 41.460 41.706
mp_waitall_1 488190 16.1 33.037 41.147 33.037 41.147
qs_ot_get_derivative 55 11.6 0.001 0.001 37.608 37.745
multiply_cannon_sync_h2d 69360 15.2 31.238 35.923 31.238 35.923
multiply_cannon_metrocomm3 69360 15.2 0.202 0.212 24.106 32.988
qs_ot_p2m_diag 40 11.0 0.022 0.133 30.878 30.955
rebuild_ks_matrix 110 8.4 0.000 0.000 29.084 29.243
qs_ks_build_kohn_sham_matrix 110 9.4 0.012 0.024 29.084 29.243
cp_dbcsr_syevd 40 12.0 0.002 0.004 28.024 28.026
qs_ks_update_qs_env 112 7.6 0.001 0.001 26.653 26.797
init_scf_loop 7 6.6 0.000 0.004 26.080 26.081
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 22.981 23.217
apply_single 62 13.6 0.000 0.000 22.981 23.217
cp_fm_syevd 40 13.0 0.000 0.000 22.885 23.028
prepare_preconditioner 7 7.6 0.000 0.000 21.173 21.200
make_preconditioner 7 8.6 0.000 0.002 21.173 21.200
ot_new_cg_direction 55 11.6 0.001 0.001 20.910 20.910
cp_fm_redistribute_end 40 14.0 9.120 18.197 9.125 18.199
cp_fm_syevd_base 40 14.0 9.066 18.148 9.066 18.148
qs_rho_update_rho_low 110 7.6 0.001 0.001 17.191 17.544
calculate_rho_elec 110 8.6 0.030 0.033 17.190 17.543
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.286 14.440
make_full_inverse_cholesky 7 9.6 0.000 0.000 14.222 14.286
init_scf_run 5 5.8 0.000 0.002 14.201 14.202
scf_env_initial_rho_setup 5 6.8 0.002 0.003 14.201 14.202
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 13.187 13.288
rs_pw_transfer 690 11.5 0.011 0.012 11.416 12.712
density_rs2pw 110 9.6 0.006 0.007 11.327 12.569
pw_transfer 1645 12.4 0.083 0.112 12.077 12.328
fft_wrap_pw1pw2 1425 13.5 0.013 0.017 11.935 12.191
mp_sum_l 4764 12.2 11.311 12.156 11.311 12.156
calculate_dm_sparse 110 9.5 0.000 0.001 11.561 11.750
qs_vxc_create 110 10.4 0.002 0.028 10.421 10.474
fft_wrap_pw1pw2_240 915 15.0 1.193 1.291 10.155 10.389
cp_fm_cholesky_invert 7 10.6 9.997 10.006 9.997 10.006
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 9.914 9.984
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.001 9.628 9.644
check_diag 80 13.5 8.588 8.860 9.353 9.497
calculate_first_density_matrix 1 7.0 0.000 0.002 9.373 9.385
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 9.183 9.228
acc_transpose_blocks 69360 15.2 0.336 0.365 8.432 9.220
dbcsr_mm_accdrv_process 154766 15.8 3.785 4.044 7.892 9.216
sum_up_and_integrate 60 10.3 0.028 0.031 8.679 8.693
integrate_v_rspace 60 11.3 0.001 0.002 8.651 8.666
fft3d_pb 915 16.0 2.381 2.685 8.218 8.484
multiply_cannon_metrocomm1 69360 15.2 0.092 0.099 4.385 8.396
xc_rho_set_and_dset_create 110 12.4 0.077 0.103 7.333 7.592
acc_transpose_blocks_kernels 69360 16.2 0.809 0.867 6.650 7.331
xc_vxc_pw_create 60 11.3 0.039 0.050 6.991 7.043
make_m2s 2890 13.2 0.082 0.095 6.383 6.959
make_images 2890 14.2 0.240 0.260 6.272 6.849
make_full_single_inverse 7 9.6 0.001 0.005 6.678 6.706
jit_kernel_transpose 5 15.0 5.841 6.477 5.841 6.477
xc_pw_derive 510 13.4 0.006 0.007 6.180 6.278
mp_waitany 7680 13.5 4.431 5.819 4.431 5.819
mp_alltoall_z22v 2340 17.7 5.295 5.574 5.295 5.574
potential_pw2rs 60 12.3 0.003 0.003 4.884 4.915
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=227.747000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=557.000000, yerr=1.788854
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: a7c7cb152c214f2aa9e9d66ac02b6f172b700f7e
Summary: empty
Status: OK