=== 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: 2ed94b8b96b2a4590f5340063d84c5d639a6fefe ################# ARCHITECTURE FILE ################## #!/bin/bash # # CP2K arch file for Cray-XC40 (Piz Daint, CSCS, multi-core partition) # # Tested with: GNU 11.2.0, Cray-MPICH 7.7.18, # Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10, # COSMA 2.6.6, ELPA 2023.05.001, HDF5 1.14.2, # LIBINT 2.6.0, LIBPEXSI 1.2.0, LIBXC 6.2.2, # LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.9.0, # SIRIUS 7.5.2, SPGLIB 2.3.1, LIBGRPP 20231215, # SPFFT 1.0.6, SPLA 1.5.5 # # 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: 28.03.2024 # # \ 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="11.2.0"; \ fi; \ module add daint-mc; \ module rm PrgEnv-cray; \ module add PrgEnv-gnu; \ module rm gcc; \ module add gcc/${gcc_version}; \ module add cray-fftw/3.3.8.10; \ echo "Expected setup:"; \ echo " cray-mpich/7.7.18"; \ echo " craype-broadwell"; \ echo " daint-mc/21.09"; \ echo " craype/2.7.10"; \ echo " cray-libsci/20.09.1"; \ echo " PrgEnv-gnu/6.0.10"; \ echo " gcc/${gcc_version}"; \ echo " cray-fftw/3.3.8.10"; \ module list; \ module -f save cp2k_mc_gnu_psmp; \ echo "To load the required modules in your batch job script, use:"; \ echo " module restore cp2k_mc_gnu_psmp"; \ cd tools/toolchain; \ rm -rf build; \ ./install_cp2k_toolchain.sh -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=${this_file##*.} libcp2k"; \ echo; \ return # Set options DO_CHECKS := no USE_COSMA := 2.6.6 USE_ELPA := 2023.05.001 USE_HDF5 := 1.14.2 USE_LIBGRPP := 20231215 USE_LIBINT := 2.6.0 USE_LIBPEXSI := 1.2.0 USE_LIBVORI := 220621 USE_LIBXC := 6.2.2 USE_LIBXSMM := 1.17 USE_PLUMED := 2.9.0 USE_SPFFT := 1.0.6 USE_SPLA := 1.5.5 #USE_QUIP := 0.9.10 USE_SIRIUS := 7.5.2 USE_SPGLIB := 2.3.1 # Only needed for SIRIUS LIBVDWXC_VER := 0.4.0 # Only needed for LIBPEXSI SCOTCH_VER := 6.0.0 SUPERLU_VER := 6.1.0 LMAX := 5 MAX_CONTR := 4 CC := cc FC := ftn LD := ftn AR := ar -r # cc, CC, and ftn include already the proper -march flag CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g DFLAGS := -D__parallel DFLAGS += -D__SCALAPACK DFLAGS += -D__FFTW3 DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR)) INSTALL_PATH := $(PWD)/tools/toolchain/install ifeq ($(DO_CHECKS), yes) DFLAGS += -D__CHECK_DIAG endif 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 := cpu 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 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_LIBGRPP),) USE_LIBGRPP := $(strip $(USE_LIBGRPP)) LIBGRPP_INC := $(INSTALL_PATH)/libgrpp-main-$(USE_LIBGRPP)/include LIBGRPP_LIB := $(INSTALL_PATH)/libgrpp-main-$(USE_LIBGRPP)/lib CFLAGS += -I$(LIBGRPP_INC) DFLAGS += -D__LIBGRPP LIBS += $(LIBGRPP_LIB)/liblibgrpp.a endif ifneq ($(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)) LIBVDWXC_VER := $(strip $(LIBVDWXC_VER)) LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib CFLAGS += -I$(LIBVDWXC_INC) CFLAGS += -I$(SIRIUS_INC) DFLAGS += -D__LIBVDWXC DFLAGS += -D__SIRIUS LIBS += $(SIRIUS_LIB)/libsirius.a LIBS += $(LIBVDWXC_LIB)/libvdwxc.a endif ifneq ($(USE_SPFFT),) USE_SPFFT := $(strip $(USE_SPFFT)) SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(USE_SPFFT)/include SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(USE_SPFFT)/lib CFLAGS += -I$(SPFFT_INC) DFLAGS += -D__SPFFT LIBS += $(SPFFT_LIB)/libspfft.a endif ifneq ($(USE_SPLA),) USE_SPLA := $(strip $(USE_SPLA)) SPLA_INC := $(INSTALL_PATH)/SpLA-$(USE_SPLA)/include/spla SPLA_LIB := $(INSTALL_PATH)/SpLA-$(USE_SPLA)/lib CFLAGS += -I$(SPLA_INC) DFLAGS += -D__SPLA LIBS += $(SPLA_LIB)/libspla.a endif ifneq ($(USE_HDF5),) USE_HDF5 := $(strip $(USE_HDF5)) HDF5_INC := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/include HDF5_LIB := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/lib CFLAGS += -I$(HDF5_INC) DFLAGS += -D__HDF5 LIBS += $(HDF5_LIB)/libhdf5_fortran.a LIBS += $(HDF5_LIB)/libhdf5_f90cstub.a LIBS += $(HDF5_LIB)/libhdf5.a endif ifneq ($(USE_COSMA),) USE_COSMA := $(strip $(USE_COSMA)) 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 += $(COSMA_LIB)/libcosta.a endif ifneq ($(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) 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 LDFLAGS := $(FCFLAGS) -static LIBS += -lz -ldl -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 = 4 # ranks/node = 4 # threads/rank = 9 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/01 job id: 52640058 --- Point --- name: 10 plot: h2o_32_ri_rpa_mp2 regex: Total RI-RPA Time= label: RI-RPA (4n/4r/9t) --- Point --- name: 11 plot: h2o_32_ri_rpa_mp2_mem regex: Estimated peak process memory label: RI-RPA (4n/4r/9t) ~~~~~~~ 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 = 4 # ranks/node = 12 # threads/rank = 3 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/02 job id: 52640059 --- Point --- name: 20 plot: h2o_32_ri_rpa_mp2 regex: Total MP2 Time= label: RI-MP2 (4n/12r/3t) --- Point --- name: 21 plot: h2o_32_ri_rpa_mp2_mem regex: Estimated peak process memory label: RI-MP2 (4n/12r/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 = 4 # ranks/node = 36 # threads/rank = 1 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/03 job id: 52640060 --- Point --- name: 100 plot: h2o_64_md regex: CP2K label: (4n/36r/1t) --- Point --- name: 101 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/36r/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 = 4 # ranks/node = 18 # threads/rank = 2 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/04 job id: 52640061 --- Point --- name: 102 plot: h2o_64_md regex: CP2K label: (4n/18r/2t) --- Point --- name: 103 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/18r/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 = 4 # ranks/node = 12 # threads/rank = 3 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/05 job id: 52640062 --- Point --- name: 104 plot: h2o_64_md regex: CP2K label: (4n/12r/3t) --- Point --- name: 105 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/12r/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 = 4 # ranks/node = 9 # threads/rank = 4 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/06 job id: 52640063 --- Point --- name: 106 plot: h2o_64_md regex: CP2K label: (4n/9r/4t) --- Point --- name: 107 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/9r/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 = 4 # ranks/node = 6 # threads/rank = 6 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/07 job id: 52640064 --- Point --- name: 108 plot: h2o_64_md regex: CP2K label: (4n/6r/6t) --- Point --- name: 109 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/6r/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 = 4 # ranks/node = 4 # threads/rank = 9 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/08 job id: 52640065 --- Point --- name: 110 plot: h2o_64_md regex: CP2K label: (4n/4r/9t) --- Point --- name: 111 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/4r/9t) ~~~~~~~ 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 = 4 # ranks/node = 3 # threads/rank = 12 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/09 job id: 52640066 --- Point --- name: 112 plot: h2o_64_md regex: CP2K label: (4n/3r/12t) --- Point --- name: 113 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/3r/12t) ~~~~~~~ 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 = 4 # ranks/node = 2 # threads/rank = 18 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/10 job id: 52640067 --- Point --- name: 114 plot: h2o_64_md regex: CP2K label: (4n/2r/18t) --- Point --- name: 115 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/2r/18t) ~~~~~~~ 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 = 4 # ranks/node = 1 # threads/rank = 36 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/11 job id: 52640068 --- Point --- name: 116 plot: h2o_64_md regex: CP2K label: (4n/1r/36t) --- Point --- name: 117 plot: h2o_64_md_mem regex: Estimated peak process memory label: (4n/1r/36t) ~~~~~~~ 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 = 4 # ranks/node = 36 # threads/rank = 1 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/12 job id: 52640069 --- Point --- name: 200 plot: h2o_128_md regex: CP2K label: (4n/36r/1t) --- Point --- name: 201 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/36r/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 = 4 # ranks/node = 18 # threads/rank = 2 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/13 job id: 52640070 --- Point --- name: 202 plot: h2o_128_md regex: CP2K label: (4n/18r/2t) --- Point --- name: 203 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/18r/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 = 4 # ranks/node = 12 # threads/rank = 3 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/14 job id: 52640071 --- Point --- name: 204 plot: h2o_128_md regex: CP2K label: (4n/12r/3t) --- Point --- name: 205 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/12r/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 = 4 # ranks/node = 9 # threads/rank = 4 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/15 job id: 52640072 --- Point --- name: 206 plot: h2o_128_md regex: CP2K label: (4n/9r/4t) --- Point --- name: 207 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/9r/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 = 4 # ranks/node = 6 # threads/rank = 6 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/16 job id: 52640073 --- Point --- name: 208 plot: h2o_128_md regex: CP2K label: (4n/6r/6t) --- Point --- name: 209 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/6r/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 = 4 # ranks/node = 4 # threads/rank = 9 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/17 job id: 52640074 --- Point --- name: 210 plot: h2o_128_md regex: CP2K label: (4n/4r/9t) --- Point --- name: 211 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/4r/9t) ~~~~~~~ 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 = 4 # ranks/node = 3 # threads/rank = 12 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/18 job id: 52640075 --- Point --- name: 212 plot: h2o_128_md regex: CP2K label: (4n/3r/12t) --- Point --- name: 213 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/3r/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 = 4 # ranks/node = 2 # threads/rank = 18 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/19 job id: 52640076 --- Point --- name: 214 plot: h2o_128_md regex: CP2K label: (4n/2r/18t) --- Point --- name: 215 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/2r/18t) ~~~~~~~ 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 = 4 # ranks/node = 1 # threads/rank = 36 nrepeat = 1 time[min] = 10 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/20 job id: 52640077 --- Point --- name: 216 plot: h2o_128_md regex: CP2K label: (4n/1r/36t) --- Point --- name: 217 plot: h2o_128_md_mem regex: Estimated peak process memory label: (4n/1r/36t) ~~~~~~~ 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 = 4 # ranks/node = 36 # threads/rank = 1 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/21 job id: 52640078 --- Point --- name: 400 plot: h2o_256_md regex: CP2K label: (4n/36r/1t) --- Point --- name: 401 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/36r/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 = 4 # ranks/node = 18 # threads/rank = 2 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/22 job id: 52640079 --- Point --- name: 402 plot: h2o_256_md regex: CP2K label: (4n/18r/2t) --- Point --- name: 403 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/18r/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 = 4 # ranks/node = 12 # threads/rank = 3 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/23 job id: 52640080 --- Point --- name: 404 plot: h2o_256_md regex: CP2K label: (4n/12r/3t) --- Point --- name: 405 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/12r/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 = 4 # ranks/node = 9 # threads/rank = 4 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/24 job id: 52640081 --- Point --- name: 406 plot: h2o_256_md regex: CP2K label: (4n/9r/4t) --- Point --- name: 407 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/9r/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 = 4 # ranks/node = 6 # threads/rank = 6 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/25 job id: 52640082 --- Point --- name: 408 plot: h2o_256_md regex: CP2K label: (4n/6r/6t) --- Point --- name: 409 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/6r/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 = 4 # ranks/node = 4 # threads/rank = 9 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/26 job id: 52640083 --- Point --- name: 410 plot: h2o_256_md regex: CP2K label: (4n/4r/9t) --- Point --- name: 411 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/4r/9t) ~~~~~~~ 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 = 4 # ranks/node = 3 # threads/rank = 12 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/27 job id: 52640084 --- Point --- name: 412 plot: h2o_256_md regex: CP2K label: (4n/3r/12t) --- Point --- name: 413 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/3r/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 = 4 # ranks/node = 2 # threads/rank = 18 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/28 job id: 52640085 --- Point --- name: 414 plot: h2o_256_md regex: CP2K label: (4n/2r/18t) --- Point --- name: 415 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/2r/18t) ~~~~~~~ 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 = 4 # ranks/node = 1 # threads/rank = 36 nrepeat = 1 time[min] = 30 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/29 job id: 52640086 --- Point --- name: 416 plot: h2o_256_md regex: CP2K label: (4n/1r/36t) --- Point --- name: 417 plot: h2o_256_md_mem regex: Estimated peak process memory label: (4n/1r/36t) ~~~~~~~ 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 = 4 # ranks/node = 36 # threads/rank = 1 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/30 job id: 52640087 --- Point --- name: 500 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/36r/1t) --- Point --- name: 501 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/36r/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 = 4 # ranks/node = 18 # threads/rank = 2 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/31 job id: 52640088 --- Point --- name: 502 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/18r/2t) --- Point --- name: 503 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/18r/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 = 4 # ranks/node = 9 # threads/rank = 4 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/32 job id: 52640089 --- Point --- name: 504 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/9r/4t) --- Point --- name: 505 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/9r/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 = 4 # ranks/node = 6 # threads/rank = 6 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/33 job id: 52640090 --- Point --- name: 506 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/6r/6t) --- Point --- name: 507 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/6r/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 = 4 # ranks/node = 4 # threads/rank = 9 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/34 job id: 52640091 --- Point --- name: 508 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/4r/9t) --- Point --- name: 509 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/4r/9t) ~~~~~~~ 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 = 4 # ranks/node = 3 # threads/rank = 12 nrepeat = 1 time[min] = 15 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/35 job id: 52640093 --- Point --- name: 510 plot: h2o_32_nrep3_ls regex: CP2K label: (4n/3r/12t) --- Point --- name: 511 plot: h2o_32_nrep3_ls_mem regex: Estimated peak process memory label: (4n/3r/12t) ~~~~~~~ END TEST ~~~~~~~ ~~~~~~~~~ TEST ~~~~~~~~~ description: 512 H2O (4 NVE MD steps on 16 nodes) input file: benchmarks/QS/00512_H2O/H2O-512_md.inp required files: [] output file: result.log # nodes = 16 # ranks/node = 36 # threads/rank = 1 nrepeat = 1 time[min] = 20 run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/36 job id: 52640094 --- Point --- name: 601 plot: h2o_512_md regex: CP2K label: (16n/36r/1t) --- Point --- name: 602 plot: h2o_512_md_mem regex: Estimated peak process memory label: (16n/36r/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 16 nodes)", xlabel="Revision", ylabel="Time [s]" ~~~~~~~~~ PLOT ~~~~~~~~~ Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 16 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]" === END PLOTS (description) === ============ RESULTS ============ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_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 172669. 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 0.024 0.079 258.732 258.744 farming_run 1 2.0 257.778 257.789 258.601 258.610 ------------------------------------------------------------------------------- @@@@@@@@@@ Run number: 2 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 32 x 32 x 32 4194304 0.0% 100.0% 0.0% flops 29 x 32 x 32 7602176 0.0% 100.0% 0.0% flops 14 x 32 x 32 14221312 0.0% 100.0% 0.0% flops 28 x 32 x 32 27525120 0.0% 100.0% 0.0% flops 43 x 32 x 32 28180480 0.0% 100.0% 0.0% flops 86 x 32 x 32 28180480 0.0% 100.0% 0.0% flops 14 x 32 x 456 78446592 0.0% 100.0% 0.0% flops 57 x 32 x 32 102727680 0.0% 100.0% 0.0% flops 14 x 14 x 32 208732160 0.0% 100.0% 0.0% flops 29 x 14 x 32 212860928 0.0% 100.0% 0.0% flops 14 x 29 x 32 212860928 0.0% 100.0% 0.0% flops 29 x 29 x 32 227352576 0.0% 100.0% 0.0% flops 32 x 32 x 456 298844160 0.0% 100.0% 0.0% flops 28 x 32 x 456 313786368 0.0% 100.0% 0.0% flops 43 x 32 x 456 321257472 0.0% 100.0% 0.0% flops 86 x 32 x 456 321257472 0.0% 100.0% 0.0% flops 57 x 32 x 456 1171095552 0.0% 100.0% 0.0% flops 14 x 32 x 14 895979560448 0.0% 100.0% 0.0% flops 29 x 32 x 14 928073646080 0.0% 100.0% 0.0% flops 14 x 32 x 29 928073646080 0.0% 100.0% 0.0% flops 29 x 32 x 29 961219133440 0.0% 100.0% 0.0% flops 32 x 32 x 14 1693022420992 0.0% 100.0% 0.0% flops 32 x 32 x 29 1753487507456 0.0% 100.0% 0.0% flops inhomo. stacks 1804075008 100.0% 0.0% 0.0% flops total 7.165239E+12 0.0% 100.0% 0.0% flops max/rank 447.990765E+09 0.0% 100.0% 0.0% matmuls inhomo. stacks 1440 100.0% 0.0% 0.0% matmuls total 249334846 0.0% 100.0% 0.0% number of processed stacks 368972 0.4% 99.6% 0.0% average stack size 1.0 678.4 0.0 marketing flops 7.165779E+12 ------------------------------------------------------------------------------- # multiplications 1160 max memory usage/rank 1.396593E+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 24 12. MP_Allreduce 2365 53. MP_Alltoall 4670 822089. MP_ISend 2604 90540. MP_IRecv 2604 90537. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 12 MP_Bcast 230 1131389. 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 0.027 234.029 234.030 qs_energies 1 2.0 0.002 0.004 233.792 233.832 mp2_main 1 3.0 0.089 0.231 231.320 231.358 mp2_gpw_main 1 4.0 0.245 0.752 230.124 230.195 mp2_ri_gpw_compute_in 1 5.0 0.441 0.759 151.252 153.363 mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.006 98.023 99.954 mp2_eri_3c_integrate_gpw 272 7.0 0.174 0.185 83.099 84.706 rpa_ri_compute_en 1 5.0 0.042 0.043 78.337 81.162 rpa_num_int 1 6.0 0.022 0.046 67.738 67.738 rpa_num_int_RPA_matrix_operati 8 7.0 0.004 0.007 67.565 67.700 calc_mat_Q 8 8.0 0.001 0.001 66.301 66.552 contract_S_to_Q 8 9.0 0.000 0.000 65.088 65.328 parallel_gemm_fm 14 9.1 0.000 0.000 64.666 64.845 parallel_gemm_fm_cosma 14 10.1 64.666 64.845 64.666 64.845 integrate_v_rspace 273 8.0 0.425 0.459 61.099 62.269 grid_integrate_task_list 273 9.0 55.868 57.109 55.868 57.109 get_2c_integrals 1 6.0 0.132 0.365 48.389 53.122 fft_wrap_pw1pw2 5465 10.4 0.081 0.087 40.279 44.279 fft_wrap_pw1pw2_100 2178 11.4 4.472 5.507 36.383 39.615 compute_2c_integrals 1 7.0 0.008 0.013 31.877 32.287 compute_2c_integrals_loop_lm 1 8.0 0.021 0.039 30.122 31.663 mp2_eri_2c_integrate_gpw 1 9.0 2.079 2.205 30.101 31.659 fft3d_s 5443 12.4 20.440 22.199 20.467 22.227 cp_fm_cholesky_decompose 12 8.2 16.608 20.986 16.608 20.986 cholesky_decomp 1 7.0 0.000 0.000 15.411 19.799 calculate_wavefunction 272 8.0 5.880 6.068 15.713 16.927 mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.003 0.014 14.685 15.690 calc_potential_gpw 544 9.5 0.005 0.006 14.538 15.661 collocate_single_gaussian 272 10.0 0.055 0.061 12.651 13.581 ao_to_mo_and_store_B_mult_1 272 7.0 11.967 13.358 11.967 13.358 potential_pw2rs 545 10.0 0.145 0.161 11.150 12.231 pw_scatter_s 2720 12.7 9.008 9.511 9.008 9.511 create_integ_mat 1 6.0 0.010 0.014 8.576 8.576 array2fm 1 7.0 0.000 0.000 7.485 7.863 mp_sync 25 8.8 4.529 6.177 4.529 6.177 pw_poisson_solve 545 10.5 0.010 0.013 4.903 6.109 pw_gather_s 2722 12.2 4.000 4.699 4.000 4.699 mp_min_d 1 6.0 1.873 4.697 1.873 4.697 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (4n/4r/9t)", y=230.211120, yerr=0.000000 PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (4n/4r/9t)", y=2748.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_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 200775. 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 0.022 0.039 533.358 533.450 farming_run 1 2.0 531.688 531.762 533.190 533.331 ------------------------------------------------------------------------------- @@@@@@@@@@ Run number: 2 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 32 x 32 x 32 16777216 0.0% 100.0% 0.0% flops 100 x 32 x 143 36608000 0.0% 100.0% 0.0% flops 14 x 32 x 32 36700160 0.0% 100.0% 0.0% flops 29 x 32 x 32 38010880 0.0% 100.0% 0.0% flops 128 x 32 x 143 46858240 0.0% 100.0% 0.0% flops 157 x 32 x 143 57474560 0.0% 100.0% 0.0% flops 100 x 32 x 32 58982400 0.0% 100.0% 0.0% flops 171 x 32 x 143 62599680 0.0% 100.0% 0.0% flops 186 x 32 x 143 68090880 0.0% 100.0% 0.0% flops 100 x 32 x 142 72704000 0.0% 100.0% 0.0% flops 200 x 32 x 143 73216000 0.0% 100.0% 0.0% flops 128 x 32 x 32 75497472 0.0% 100.0% 0.0% flops 100 x 32 x 157 80384000 0.0% 100.0% 0.0% flops 157 x 32 x 32 92602368 0.0% 100.0% 0.0% flops 128 x 32 x 142 93061120 0.0% 100.0% 0.0% flops 171 x 32 x 32 100859904 0.0% 100.0% 0.0% flops 128 x 32 x 157 102891520 0.0% 100.0% 0.0% flops 142 x 32 x 143 103966720 0.0% 100.0% 0.0% flops 143 x 32 x 143 104698880 0.0% 100.0% 0.0% flops 186 x 32 x 32 109707264 0.0% 100.0% 0.0% flops 157 x 32 x 142 114145280 0.0% 100.0% 0.0% flops 156 x 32 x 143 114216960 0.0% 100.0% 0.0% flops 200 x 32 x 32 117964800 0.0% 100.0% 0.0% flops 171 x 32 x 142 124323840 0.0% 100.0% 0.0% flops 157 x 32 x 157 126202880 0.0% 100.0% 0.0% flops 186 x 32 x 142 135229440 0.0% 100.0% 0.0% flops 171 x 32 x 157 137456640 0.0% 100.0% 0.0% flops 200 x 32 x 142 145408000 0.0% 100.0% 0.0% flops 32 x 32 x 143 146432000 0.0% 100.0% 0.0% flops 186 x 32 x 157 149514240 0.0% 100.0% 0.0% flops 200 x 32 x 157 160768000 0.0% 100.0% 0.0% flops 142 x 32 x 32 167510016 0.0% 100.0% 0.0% flops 143 x 32 x 32 168689664 0.0% 100.0% 0.0% flops 156 x 32 x 32 184025088 0.0% 100.0% 0.0% flops 142 x 32 x 142 206479360 0.0% 100.0% 0.0% flops 143 x 32 x 142 207933440 0.0% 100.0% 0.0% flops 156 x 32 x 142 226836480 0.0% 100.0% 0.0% flops 142 x 32 x 157 228290560 0.0% 100.0% 0.0% flops 143 x 32 x 157 229898240 0.0% 100.0% 0.0% flops 156 x 32 x 157 250798080 0.0% 100.0% 0.0% flops 32 x 32 x 142 290816000 0.0% 100.0% 0.0% flops 32 x 32 x 157 321536000 0.0% 100.0% 0.0% flops 14 x 14 x 32 626196480 0.0% 100.0% 0.0% flops 29 x 14 x 32 638582784 0.0% 100.0% 0.0% flops 14 x 29 x 32 638582784 0.0% 100.0% 0.0% flops 29 x 29 x 32 682057728 0.0% 100.0% 0.0% flops 14 x 32 x 14 896799536640 0.0% 100.0% 0.0% flops 29 x 32 x 14 928925089792 0.0% 100.0% 0.0% flops 14 x 32 x 29 928925089792 0.0% 100.0% 0.0% flops 29 x 32 x 29 962100985856 0.0% 100.0% 0.0% flops 32 x 32 x 14 1693022420992 0.0% 100.0% 0.0% flops 32 x 32 x 29 1753487507456 0.0% 100.0% 0.0% flops inhomo. stacks 1112785920 100.0% 0.0% 0.0% flops total 7.172345E+12 0.0% 100.0% 0.0% flops max/rank 150.710992E+09 0.1% 99.9% 0.0% matmuls inhomo. stacks 980 100.0% 0.0% 0.0% matmuls total 249562190 0.0% 100.0% 0.0% number of processed stacks 347432 0.3% 99.7% 0.0% average stack size 1.0 720.3 0.0 marketing flops 7.174951E+12 ------------------------------------------------------------------------------- # multiplications 1140 max memory usage/rank 1.208504E+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 53 12. MP_Allreduce 1182 39. MP_Alltoall 1797 713945. MP_ISend 3686 54897. MP_IRecv 3622 54246. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 12 MP_Bcast 757 480173. MP_Allreduce 2021 21391. MP_Sync 37 MP_Alltoall 77 MP_SendRecv 4192 1987179. MP_ISendRecv 1034 172713. 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 0.019 0.042 340.913 340.972 qs_energies 1 2.0 0.000 0.000 340.704 340.724 mp2_main 1 3.0 0.000 0.000 236.801 236.816 mp2_gpw_main 1 4.0 0.005 0.011 235.805 235.855 mp2_ri_gpw_compute_en 1 5.0 0.053 0.066 123.956 139.080 mp2_ri_gpw_compute_en_RI_loop 1 6.0 2.412 2.525 117.934 117.962 mp2_ri_gpw_compute_in 1 5.0 0.096 0.106 111.712 116.854 scf_env_do_scf 1 3.0 0.000 0.000 103.606 103.624 qs_ks_update_qs_env 5 5.0 0.000 0.000 102.671 102.689 rebuild_ks_matrix 4 6.0 0.000 0.000 102.669 102.687 qs_ks_build_kohn_sham_matrix 4 7.0 0.023 0.026 102.669 102.687 hfx_ks_matrix 4 8.0 0.001 0.001 102.224 102.243 integrate_four_center 4 9.0 0.279 0.654 102.223 102.242 mp2_ri_gpw_compute_en_expansio 172 7.0 0.712 0.806 88.771 93.071 local_gemm 172 8.0 88.058 92.278 88.058 92.278 mp2_ri_gpw_compute_in_loop 1 6.0 0.001 0.002 82.554 87.697 integrate_four_center_main 4 10.0 0.142 0.379 78.661 82.377 integrate_four_center_bin 214 11.0 78.519 82.171 78.519 82.171 init_scf_loop 1 4.0 0.000 0.000 80.685 80.699 mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.189 72.132 76.626 integrate_v_rspace 95 8.0 0.307 0.477 53.836 57.466 grid_integrate_task_list 95 9.0 48.840 50.451 48.840 50.451 fft_wrap_pw1pw2 1868 10.4 0.033 0.041 36.001 42.021 fft_wrap_pw1pw2_100 730 11.4 1.820 2.261 33.287 39.178 mp2_ri_gpw_compute_en_comm 36 7.0 1.139 1.282 22.453 33.997 mp_sendrecv_dm3 3384 8.0 19.482 31.099 19.482 31.099 get_2c_integrals 1 6.0 0.000 0.000 29.022 29.074 compute_2c_integrals 1 7.0 0.003 0.004 28.196 28.210 compute_2c_integrals_loop_lm 1 8.0 0.001 0.002 24.764 28.021 mp2_eri_2c_integrate_gpw 1 9.0 1.579 1.814 24.763 28.020 fft3d_s 1823 12.4 22.183 25.940 22.201 25.959 scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 22.919 22.923 mp_min_d 2 7.0 5.211 20.501 5.211 20.501 mp2_ri_get_integ_group_size 1 6.0 0.000 0.000 5.144 20.268 integrate_four_center_load 4 10.0 0.000 0.000 18.203 18.255 hfx_load_balance 1 11.0 0.001 0.001 18.203 18.255 calc_potential_gpw 182 9.5 0.002 0.003 14.309 17.561 mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 12.650 15.210 calculate_wavefunction 91 8.0 2.152 2.494 11.576 14.636 potential_pw2rs 186 10.0 0.042 0.048 9.929 13.462 collocate_single_gaussian 91 10.0 0.019 0.022 9.731 12.391 mp_comm_split_direct 6 7.2 3.276 12.159 3.276 12.159 mp_sum_l 425 2.2 5.612 10.971 5.612 10.971 hfx_load_balance_dist 1 12.0 0.000 0.000 5.396 10.711 ao_to_mo_and_store_B_mult_1 91 7.0 8.818 10.003 8.818 10.003 hfx_load_balance_count 1 12.0 6.304 9.015 6.304 9.015 hfx_load_balance_bin 1 12.0 6.286 8.971 6.286 8.971 mp_sync 37 10.5 3.926 8.644 3.926 8.644 pw_poisson_solve 186 10.4 0.006 0.009 5.792 7.912 pw_scatter_s 910 12.7 5.725 7.013 5.725 7.013 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (4n/12r/3t)", y=235.803833, yerr=0.000000 PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (4n/12r/3t)", y=1334.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/03/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 32 x 32 x 32 26877100032 0.0% 100.0% 0.0% flops 209 x 32 x 209 42582335488 0.0% 100.0% 0.0% flops 209 x 32 x 213 43397308416 0.0% 100.0% 0.0% flops 213 x 32 x 209 43397308416 0.0% 100.0% 0.0% flops 9 x 9 x 32 44168260608 0.0% 100.0% 0.0% flops 213 x 32 x 213 44227878912 0.0% 100.0% 0.0% flops 209 x 32 x 218 44416024576 0.0% 100.0% 0.0% flops 218 x 32 x 209 44416024576 0.0% 100.0% 0.0% flops 213 x 32 x 218 45266092032 0.0% 100.0% 0.0% flops 218 x 32 x 213 45266092032 0.0% 100.0% 0.0% flops 32 x 32 x 209 46131576832 0.0% 100.0% 0.0% flops 218 x 32 x 218 46328676352 0.0% 100.0% 0.0% flops 32 x 32 x 213 47014477824 0.0% 100.0% 0.0% flops 32 x 32 x 218 48118104064 0.0% 100.0% 0.0% flops 22 x 9 x 32 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 32 53885500416 0.0% 100.0% 0.0% flops 209 x 32 x 32 56760467456 0.0% 100.0% 0.0% flops 213 x 32 x 32 57846792192 0.0% 100.0% 0.0% flops 218 x 32 x 32 59204698112 0.0% 100.0% 0.0% flops 22 x 22 x 32 67007283200 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 1.880888E+12 0.0% 100.0% 0.0% flops max/rank 20.325101E+09 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 101210040 0.0% 100.0% 0.0% number of processed stacks 3134624 0.0% 100.0% 0.0% average stack size 0.0 32.3 0.0 marketing flops 2.107629E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 178.831360E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 7242048 MPI messages size (bytes): total size 355.819487E+09 min size 0.000000E+00 max size 380.192000E+03 average size 49.132441E+03 MPI breakdown and total messages size (bytes): size <= 128 2986104 0 128 < size <= 8192 1493448 12234326016 8192 < size <= 32768 0 0 32768 < size <= 131072 2138400 116785152000 131072 < size <= 4194304 624096 226802306368 4194304 < size <= 16777216 0 0 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 27 12. MP_Allreduce 12193 16. MP_Alltoall 8655 34121. MP_ISend 109684 25393. MP_IRecv 109684 24883. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3690 72644. MP_Allreduce 10330 309. MP_Sync 1482 MP_Alltoall 2094 25181027. MP_SendRecv 34034 3780. MP_ISendRecv 34034 3780. MP_Wait 45572 MP_comm_split 50 MP_ISend 23112 34348. MP_IRecv 23112 34348. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.036 0.054 27.502 27.504 qs_mol_dyn_low 1 2.0 0.004 0.004 26.645 26.654 qs_forces 11 3.9 0.002 0.003 26.002 26.029 qs_energies 11 4.9 0.008 0.030 24.645 24.676 scf_env_do_scf 11 5.9 0.000 0.001 20.720 20.721 scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 18.797 18.799 velocity_verlet 10 3.0 0.001 0.002 13.754 13.754 rebuild_ks_matrix 119 8.3 0.000 0.001 8.000 8.118 qs_ks_build_kohn_sham_matrix 119 9.3 0.011 0.015 8.000 8.118 qs_scf_new_mos 108 7.5 0.001 0.001 7.206 7.287 qs_scf_loop_do_ot 108 8.5 0.001 0.001 7.206 7.287 qs_ks_update_qs_env 119 7.6 0.001 0.001 7.091 7.198 dbcsr_multiply_generic 2286 12.5 0.092 0.144 6.802 7.058 ot_scf_mini 108 9.5 0.002 0.002 6.817 6.876 sum_up_and_integrate 119 10.3 0.001 0.002 6.327 6.343 integrate_v_rspace 119 11.3 0.002 0.004 6.321 6.338 qs_rho_update_rho_low 119 7.7 0.001 0.001 5.809 5.828 calculate_rho_elec 119 8.7 0.010 0.011 5.808 5.827 multiply_cannon 2286 13.5 0.168 0.201 3.523 4.159 ot_mini 108 10.5 0.001 0.001 3.812 3.882 multiply_cannon_loop 2286 14.5 0.151 0.254 2.780 3.467 mp_waitall_1 294200 16.4 2.367 3.365 2.367 3.365 qs_ot_get_derivative 108 11.5 0.001 0.001 2.989 3.050 grid_integrate_task_list 119 12.3 2.801 2.953 2.801 2.953 density_rs2pw 119 9.7 0.004 0.008 2.781 2.913 mp_waitany 7404 13.9 2.463 2.761 2.463 2.761 potential_pw2rs 119 12.3 0.004 0.007 2.710 2.730 grid_collocate_task_list 119 9.7 2.416 2.591 2.416 2.591 fft_wrap_pw1pw2 1201 11.6 0.008 0.012 2.420 2.483 transfer_rs2pw 487 10.6 0.005 0.007 2.161 2.371 init_scf_run 11 5.9 0.000 0.001 2.351 2.351 scf_env_initial_rho_setup 11 6.9 0.000 0.002 2.350 2.351 make_m2s 4572 13.5 0.065 0.090 2.011 2.246 transfer_pw2rs 487 13.2 0.005 0.006 2.155 2.170 multiply_cannon_metrocomm3 27432 15.5 0.075 0.164 0.777 2.078 fft_wrap_pw1pw2_140 487 12.2 0.037 0.044 1.853 2.011 qs_ot_get_p 119 10.4 0.001 0.001 1.859 1.945 init_scf_loop 11 6.9 0.000 0.000 1.902 1.905 make_images 4572 14.5 0.135 0.188 1.730 1.902 fft3d_pb 487 13.2 0.579 0.731 1.726 1.870 mp_alltoall_d11v 2130 13.8 1.611 1.735 1.611 1.735 multiply_cannon_multrec 27432 15.5 0.902 1.585 0.910 1.592 multiply_cannon_metrocomm1 27432 15.5 0.078 0.171 0.593 1.460 qs_energies_init_hamiltonians 11 5.9 0.000 0.000 1.329 1.356 mp_sum_l 11298 13.2 0.844 1.280 0.844 1.280 transfer_pw2rs_50 119 14.3 0.073 0.083 1.090 1.183 transfer_rs2pw_50 119 11.7 0.102 0.119 1.142 1.168 mp_alltoall_z22v 1688 15.5 1.018 1.168 1.018 1.168 calculate_first_density_matrix 1 7.0 0.000 0.000 1.162 1.163 qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.069 1.154 qs_ot_get_derivative_diag 49 12.0 0.001 0.002 1.078 1.110 wfi_extrapolate 11 7.9 0.001 0.001 1.110 1.110 qs_ot_get_derivative_taylor 59 13.0 0.001 0.002 1.004 1.035 calculate_rho_core 11 7.9 0.037 0.106 0.937 1.026 qs_ot_p2m_diag 50 11.0 0.003 0.006 1.004 1.023 transfer_rs2pw_140 130 11.5 0.156 0.189 0.773 0.991 prepare_preconditioner 11 7.9 0.000 0.000 0.968 0.976 make_preconditioner 11 8.9 0.000 0.000 0.968 0.976 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.952 0.963 make_images_data 4572 15.5 0.050 0.083 0.664 0.958 make_images_sizes 4572 15.5 0.005 0.013 0.725 0.947 mp_alltoall_i44 4572 16.5 0.720 0.942 0.720 0.942 make_full_inverse_cholesky 11 9.9 0.000 0.000 0.882 0.905 rs_gather_matrices 119 12.3 0.041 0.055 0.753 0.893 mp_allgather_i34 2286 14.5 0.434 0.864 0.434 0.864 hybrid_alltoall_any 4725 16.4 0.049 0.107 0.537 0.862 cp_dbcsr_syevd 50 12.0 0.005 0.006 0.832 0.840 ot_diis_step 108 11.5 0.006 0.008 0.768 0.769 apply_preconditioner_dbcsr 119 12.6 0.000 0.000 0.611 0.722 apply_single 119 13.6 0.000 0.000 0.611 0.721 calculate_atomic_block_dm 1 8.0 0.001 0.001 0.708 0.708 transfer_pw2rs_140 130 13.9 0.226 0.255 0.651 0.700 mp_sum_d 4139 12.0 0.511 0.698 0.511 0.698 rot_ana 1 3.0 0.001 0.001 0.579 0.659 diamat_all 1 4.0 0.578 0.658 0.578 0.658 rs_scatter_matrices 130 9.7 0.028 0.036 0.622 0.642 fft3d_ps 714 14.0 0.045 0.087 0.540 0.638 dbcsr_complete_redistribute 329 12.2 0.100 0.153 0.592 0.627 cube_transpose_4 249 14.8 0.026 0.032 0.472 0.585 cp_fm_diag_elpa 50 13.0 0.000 0.000 0.569 0.570 cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 0.552 0.554 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="100", plot="h2o_64_md", label="(4n/36r/1t)", y=27.504000, yerr=0.000000 PlotPoint: name="101", plot="h2o_64_md_mem", label="(4n/36r/1t)", y=170.454545, yerr=0.782030 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/04/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 142 x 32 x 213 3685656576 0.0% 100.0% 0.0% flops 142 x 32 x 218 3772174336 0.0% 100.0% 0.0% flops 182 x 32 x 213 4723869696 0.0% 100.0% 0.0% flops 182 x 32 x 218 4834758656 0.0% 100.0% 0.0% flops 187 x 32 x 213 4853646336 0.0% 100.0% 0.0% flops 191 x 32 x 213 4957467648 0.0% 100.0% 0.0% flops 187 x 32 x 218 4967581696 0.0% 100.0% 0.0% flops 191 x 32 x 218 5073840128 0.0% 100.0% 0.0% flops 196 x 32 x 213 5087244288 0.0% 100.0% 0.0% flops 196 x 32 x 218 5206663168 0.0% 100.0% 0.0% flops 209 x 32 x 213 5424663552 0.0% 100.0% 0.0% flops 209 x 32 x 218 5552003072 0.0% 100.0% 0.0% flops 218 x 32 x 213 5658261504 0.0% 100.0% 0.0% flops 218 x 32 x 218 5791084544 0.0% 100.0% 0.0% flops 240 x 32 x 213 6229278720 0.0% 100.0% 0.0% flops 240 x 32 x 218 6375505920 0.0% 100.0% 0.0% flops 249 x 32 x 213 6462876672 0.0% 100.0% 0.0% flops 249 x 32 x 218 6614587392 0.0% 100.0% 0.0% flops 284 x 32 x 213 7371313152 0.0% 100.0% 0.0% flops 284 x 32 x 218 7544348672 0.0% 100.0% 0.0% flops 142 x 32 x 32 9641132032 0.0% 100.0% 0.0% flops 142 x 32 x 209 10849327104 0.0% 100.0% 0.0% flops 231 x 32 x 213 11991361536 0.0% 100.0% 0.0% flops 231 x 32 x 218 12272848896 0.0% 100.0% 0.0% flops 182 x 32 x 32 12356943872 0.0% 100.0% 0.0% flops 187 x 32 x 32 12696420352 0.0% 100.0% 0.0% flops 191 x 32 x 32 12968001536 0.0% 100.0% 0.0% flops 196 x 32 x 32 13307478016 0.0% 100.0% 0.0% flops 182 x 32 x 209 13905475584 0.0% 100.0% 0.0% flops 209 x 32 x 32 14190116864 0.0% 100.0% 0.0% flops 187 x 32 x 209 14287494144 0.0% 100.0% 0.0% flops 191 x 32 x 209 14593108992 0.0% 100.0% 0.0% flops 218 x 32 x 32 14801174528 0.0% 100.0% 0.0% flops 196 x 32 x 209 14975127552 0.0% 100.0% 0.0% flops 209 x 32 x 209 15968375808 0.0% 100.0% 0.0% flops 240 x 32 x 32 16294871040 0.0% 100.0% 0.0% flops 218 x 32 x 209 16656009216 0.0% 100.0% 0.0% flops 249 x 32 x 32 16905928704 0.0% 100.0% 0.0% flops 240 x 32 x 209 18336890880 0.0% 100.0% 0.0% flops 249 x 32 x 209 19024524288 0.0% 100.0% 0.0% flops 284 x 32 x 32 19282264064 0.0% 100.0% 0.0% flops 284 x 32 x 209 21698654208 0.0% 100.0% 0.0% flops 32 x 32 x 213 23507238912 0.0% 100.0% 0.0% flops 32 x 32 x 218 24059052032 0.0% 100.0% 0.0% flops 32 x 32 x 32 26877100032 0.0% 100.0% 0.0% flops 231 x 32 x 32 31367626752 0.0% 100.0% 0.0% flops 231 x 32 x 209 35298514944 0.0% 100.0% 0.0% flops 9 x 9 x 32 44168260608 0.0% 100.0% 0.0% flops 22 x 9 x 32 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 32 53885500416 0.0% 100.0% 0.0% flops 22 x 22 x 32 67007283200 0.0% 100.0% 0.0% flops 32 x 32 x 209 69197365248 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 103113707520 100.0% 0.0% 0.0% flops total 1.890248E+12 5.5% 94.5% 0.0% flops max/rank 40.294274E+09 6.1% 93.9% 0.0% matmuls inhomo. stacks 76736 100.0% 0.0% 0.0% matmuls total 101210040 0.1% 99.9% 0.0% number of processed stacks 3136704 2.4% 97.6% 0.0% average stack size 1.0 33.1 0.0 marketing flops 2.107629E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 202.608640E+06 # max total images/rank 2 # max 3D layers 1 # MPI messages exchanged 3456432 MPI messages size (bytes): total size 321.940816E+09 min size 0.000000E+00 max size 765.456000E+03 average size 93.142531E+03 MPI breakdown and total messages size (bytes): size <= 128 1163952 0 128 < size <= 8192 704472 5771034624 8192 < size <= 32768 140976 2309750784 32768 < size <= 131072 1134984 87058022400 131072 < size <= 4194304 312048 226802306368 4194304 < size <= 16777216 0 0 16777216 < size 0 0 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3690 72736. MP_Allreduce 10328 308. MP_Sync 54 MP_Alltoall 2082 270407. MP_SendRecv 16898 6600. MP_ISendRecv 16898 6600. MP_Wait 35258 MP_comm_split 50 MP_ISend 15892 63460. MP_IRecv 15892 63460. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.024 0.054 41.793 41.795 qs_mol_dyn_low 1 2.0 0.027 0.095 40.625 40.839 qs_forces 11 3.9 0.009 0.030 39.978 40.034 qs_energies 11 4.9 0.008 0.031 37.766 37.829 scf_env_do_scf 11 5.9 0.001 0.004 33.925 33.926 scf_env_do_scf_inner_loop 108 6.5 0.003 0.035 30.382 30.384 velocity_verlet 10 3.0 0.001 0.002 23.037 23.060 qs_scf_new_mos 108 7.5 0.001 0.001 12.557 12.678 qs_scf_loop_do_ot 108 8.5 0.001 0.001 12.556 12.678 ot_scf_mini 108 9.5 0.003 0.003 11.952 12.044 rebuild_ks_matrix 119 8.3 0.001 0.001 11.823 11.952 qs_ks_build_kohn_sham_matrix 119 9.3 0.016 0.021 11.822 11.952 dbcsr_multiply_generic 2286 12.5 0.130 0.140 11.026 11.534 qs_ks_update_qs_env 119 7.6 0.001 0.001 10.476 10.596 qs_rho_update_rho_low 119 7.7 0.001 0.001 9.300 9.323 calculate_rho_elec 119 8.7 0.019 0.022 9.299 9.322 sum_up_and_integrate 119 10.3 0.004 0.011 9.151 9.171 integrate_v_rspace 119 11.3 0.003 0.004 9.135 9.158 multiply_cannon 2286 13.5 0.225 0.233 5.831 6.981 ot_mini 108 10.5 0.001 0.001 6.499 6.609 multiply_cannon_loop 2286 14.5 0.212 0.227 4.664 6.013 mp_waitall_1 220534 16.5 4.009 5.979 4.009 5.979 qs_ot_get_derivative 108 11.5 0.001 0.002 5.127 5.222 grid_collocate_task_list 119 9.7 4.711 4.955 4.711 4.955 grid_integrate_task_list 119 12.3 4.597 4.788 4.597 4.788 multiply_cannon_metrocomm3 27432 15.5 0.088 0.092 1.723 4.229 density_rs2pw 119 9.7 0.006 0.008 3.928 4.148 potential_pw2rs 119 12.3 0.008 0.008 3.684 3.755 qs_ot_get_p 119 10.4 0.003 0.009 3.632 3.750 init_scf_loop 11 6.9 0.000 0.000 3.504 3.505 fft_wrap_pw1pw2 1201 11.6 0.016 0.020 3.378 3.480 make_m2s 4572 13.5 0.087 0.090 3.208 3.463 fft3d_ps 1201 13.6 0.653 0.758 2.947 3.053 make_images 4572 14.5 0.222 0.274 2.744 3.018 multiply_cannon_multrec 27432 15.5 1.726 2.753 1.738 2.764 transfer_pw2rs 487 13.2 0.006 0.007 2.686 2.755 fft_wrap_pw1pw2_140 487 12.2 0.101 0.112 2.594 2.699 transfer_rs2pw 487 10.6 0.007 0.009 2.392 2.662 init_scf_run 11 5.9 0.000 0.008 2.589 2.589 scf_env_initial_rho_setup 11 6.9 0.001 0.005 2.588 2.589 mp_waitany 15892 13.8 2.052 2.578 2.052 2.578 mp_alltoall_d11v 2130 13.8 2.212 2.452 2.212 2.452 mp_alltoall_z22v 1201 15.6 1.880 2.071 1.880 2.071 prepare_preconditioner 11 7.9 0.000 0.000 2.036 2.051 make_preconditioner 11 8.9 0.000 0.001 2.036 2.051 mp_sum_l 11298 13.2 1.343 1.991 1.343 1.991 qs_ot_p2m_diag 50 11.0 0.006 0.014 1.961 1.983 wfi_extrapolate 11 7.9 0.001 0.003 1.973 1.973 make_full_inverse_cholesky 11 9.9 0.000 0.001 1.833 1.884 qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.810 1.855 qs_ot_get_derivative_taylor 59 13.0 0.002 0.002 1.758 1.807 cp_dbcsr_syevd 50 12.0 0.004 0.004 1.699 1.703 make_images_data 4572 15.5 0.062 0.071 1.060 1.504 make_images_sizes 4572 15.5 0.006 0.006 1.078 1.465 mp_alltoall_i44 4572 16.5 1.072 1.459 1.072 1.459 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.444 1.453 hybrid_alltoall_any 4725 16.4 0.063 0.126 0.952 1.425 ot_diis_step 108 11.5 0.013 0.015 1.310 1.311 dbcsr_complete_redistribute 329 12.2 0.070 0.094 1.194 1.296 multiply_cannon_metrocomm1 27432 15.5 0.047 0.052 0.487 1.288 yz_to_x 368 14.5 0.056 0.067 1.109 1.255 apply_preconditioner_dbcsr 119 12.6 0.000 0.001 1.098 1.248 apply_single 119 13.6 0.000 0.000 1.098 1.248 cp_fm_diag_elpa 50 13.0 0.000 0.000 1.198 1.199 mp_sum_d 4139 12.0 0.711 1.155 0.711 1.155 cp_fm_redistribute_end 50 14.0 0.588 1.140 0.599 1.146 transfer_rs2pw_140 130 11.5 0.220 0.256 0.876 1.126 cp_fm_diag_elpa_base 50 14.0 0.527 1.061 0.537 1.086 mp_allgather_i34 2286 14.5 0.685 1.082 0.685 1.082 transfer_pw2rs_50 119 14.3 0.101 0.118 0.923 1.073 transfer_pw2rs_140 130 13.9 0.320 0.358 0.970 1.060 rs_gather_matrices 119 12.3 0.066 0.075 0.792 1.021 qs_energies_init_hamiltonians 11 5.9 0.001 0.002 0.857 0.916 cp_fm_cholesky_invert 11 10.9 0.892 0.904 0.892 0.904 copy_dbcsr_to_fm 153 11.3 0.003 0.003 0.845 0.890 x_to_yz 357 14.4 0.051 0.062 0.869 0.882 arnoldi_extremal 119 11.4 0.002 0.003 0.802 0.869 arnoldi_normal_ev 119 12.4 0.003 0.003 0.799 0.867 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="102", plot="h2o_64_md", label="(4n/18r/2t)", y=41.795000, yerr=0.000000 PlotPoint: name="103", plot="h2o_64_md_mem", label="(4n/18r/2t)", y=193.363636, yerr=0.979121 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/05/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 142 x 32 x 200 3460710400 0.0% 100.0% 0.0% flops 164 x 32 x 200 3996876800 0.0% 100.0% 0.0% flops 209 x 32 x 200 5093580800 0.0% 100.0% 0.0% flops 213 x 32 x 200 5191065600 0.0% 100.0% 0.0% flops 64 x 32 x 200 5518131200 0.0% 100.0% 0.0% flops 231 x 32 x 200 5629747200 0.0% 100.0% 0.0% flops 262 x 32 x 200 6385254400 0.0% 100.0% 0.0% flops 64 x 32 x 32 6719275008 0.0% 100.0% 0.0% flops 293 x 32 x 200 7140761600 0.0% 100.0% 0.0% flops 142 x 32 x 209 7232884736 0.0% 100.0% 0.0% flops 142 x 32 x 222 7682777088 0.0% 100.0% 0.0% flops 164 x 32 x 209 8353472512 0.0% 100.0% 0.0% flops 164 x 32 x 222 8873066496 0.0% 100.0% 0.0% flops 196 x 32 x 200 9553510400 0.0% 100.0% 0.0% flops 142 x 32 x 32 9641132032 0.0% 100.0% 0.0% flops 209 x 32 x 209 10645583872 0.0% 100.0% 0.0% flops 213 x 32 x 209 10849327104 0.0% 100.0% 0.0% flops 164 x 32 x 32 11134828544 0.0% 100.0% 0.0% flops 209 x 32 x 222 11307749376 0.0% 100.0% 0.0% flops 213 x 32 x 222 11524165632 0.0% 100.0% 0.0% flops 64 x 32 x 209 11532894208 0.0% 100.0% 0.0% flops 231 x 32 x 209 11766171648 0.0% 100.0% 0.0% flops 64 x 32 x 222 12250251264 0.0% 100.0% 0.0% flops 231 x 32 x 222 12498038784 0.0% 100.0% 0.0% flops 262 x 32 x 209 13345181696 0.0% 100.0% 0.0% flops 262 x 32 x 222 14175264768 0.0% 100.0% 0.0% flops 209 x 32 x 32 14190116864 0.0% 100.0% 0.0% flops 213 x 32 x 32 14461698048 0.0% 100.0% 0.0% flops 293 x 32 x 209 14924191744 0.0% 100.0% 0.0% flops 231 x 32 x 32 15683813376 0.0% 100.0% 0.0% flops 293 x 32 x 222 15852490752 0.0% 100.0% 0.0% flops 218 x 32 x 200 15938764800 0.0% 100.0% 0.0% flops 32 x 32 x 200 16554393600 0.0% 100.0% 0.0% flops 262 x 32 x 32 17788567552 0.0% 100.0% 0.0% flops 293 x 32 x 32 19893321728 0.0% 100.0% 0.0% flops 196 x 32 x 209 19966836736 0.0% 100.0% 0.0% flops 32 x 32 x 32 20157825024 0.0% 100.0% 0.0% flops 196 x 32 x 222 21208793088 0.0% 100.0% 0.0% flops 196 x 32 x 32 26614956032 0.0% 100.0% 0.0% flops 218 x 32 x 209 33312018432 0.0% 100.0% 0.0% flops 32 x 32 x 209 34598682624 0.0% 100.0% 0.0% flops 218 x 32 x 222 35384057856 0.0% 100.0% 0.0% flops 32 x 32 x 222 36750753792 0.0% 100.0% 0.0% flops 9 x 9 x 32 44168260608 0.0% 100.0% 0.0% flops 218 x 32 x 32 44403523584 0.0% 100.0% 0.0% flops 22 x 9 x 32 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 32 53885500416 0.0% 100.0% 0.0% flops 22 x 22 x 32 67007283200 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 105981222912 100.0% 0.0% 0.0% flops total 1.894805E+12 5.6% 94.4% 0.0% flops max/rank 58.021006E+09 6.4% 93.6% 0.0% matmuls inhomo. stacks 70000 100.0% 0.0% 0.0% matmuls total 101118360 0.1% 99.9% 0.0% number of processed stacks 3045024 2.3% 97.7% 0.0% average stack size 1.0 34.0 0.0 marketing flops 2.107592E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 219.439104E+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 3679 72937. MP_Allreduce 10306 303. MP_Sync 54 MP_Alltoall 2060 1032890. 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 0.092 0.299 36.575 36.610 qs_mol_dyn_low 1 2.0 0.123 0.260 34.218 34.269 qs_forces 11 3.9 0.002 0.003 33.090 33.230 qs_energies 11 4.9 0.001 0.001 31.405 31.545 scf_env_do_scf 11 5.9 0.001 0.003 27.138 27.151 scf_env_do_scf_inner_loop 108 6.5 0.003 0.030 24.208 24.221 velocity_verlet 10 3.0 0.001 0.001 18.521 18.656 qs_scf_new_mos 108 7.5 0.001 0.001 11.009 11.093 qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.008 11.092 dbcsr_multiply_generic 2286 12.5 0.141 0.196 10.535 10.885 ot_scf_mini 108 9.5 0.003 0.004 10.439 10.533 rebuild_ks_matrix 119 8.3 0.001 0.001 8.735 8.922 qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.016 8.735 8.921 qs_ks_update_qs_env 119 7.6 0.001 0.002 7.765 7.934 qs_rho_update_rho_low 119 7.7 0.001 0.001 6.928 6.941 calculate_rho_elec 119 8.7 0.026 0.031 6.927 6.940 multiply_cannon 2286 13.5 0.228 0.268 5.600 6.541 sum_up_and_integrate 119 10.3 0.001 0.002 6.308 6.338 integrate_v_rspace 119 11.3 0.003 0.004 6.297 6.327 ot_mini 108 10.5 0.001 0.001 5.805 5.931 multiply_cannon_loop 2286 14.5 0.234 0.346 4.625 5.503 mp_waitall_1 200699 16.5 2.781 5.451 2.781 5.451 grid_collocate_task_list 119 9.7 4.245 4.703 4.245 4.703 qs_ot_get_derivative 108 11.5 0.001 0.001 4.346 4.446 multiply_cannon_metrocomm3 27432 15.5 0.102 0.203 1.760 4.392 grid_integrate_task_list 119 12.3 3.908 4.101 3.908 4.101 multiply_cannon_multrec 27432 15.5 1.828 3.333 1.841 3.346 make_m2s 4572 13.5 0.092 0.125 3.157 3.329 init_scf_run 11 5.9 0.000 0.005 3.158 3.160 scf_env_initial_rho_setup 11 6.9 0.000 0.004 3.158 3.160 density_rs2pw 119 9.7 0.005 0.008 2.374 3.135 qs_ot_get_p 119 10.4 0.001 0.001 2.840 2.951 init_scf_loop 11 6.9 0.000 0.000 2.908 2.909 make_images 4572 14.5 0.238 0.333 2.551 2.790 transfer_rs2pw 487 10.6 0.005 0.007 1.574 2.430 fft_wrap_pw1pw2 1201 11.6 0.012 0.015 2.189 2.247 fft3d_ps 1201 13.6 0.556 0.662 1.896 2.033 mp_waitany 5720 13.7 1.074 1.995 1.074 1.995 potential_pw2rs 119 12.3 0.007 0.011 1.919 1.939 mp_sum_l 11298 13.2 1.145 1.890 1.145 1.890 transfer_rs2pw_140 130 11.5 0.151 0.174 1.023 1.881 fft_wrap_pw1pw2_140 487 12.2 0.060 0.076 1.800 1.849 prepare_preconditioner 11 7.9 0.000 0.000 1.810 1.826 make_preconditioner 11 8.9 0.000 0.000 1.809 1.825 make_full_inverse_cholesky 11 9.9 0.000 0.000 1.650 1.691 qs_ot_get_derivative_diag 49 12.0 0.001 0.002 1.563 1.615 calculate_first_density_matrix 1 7.0 0.000 0.000 1.571 1.571 get_fft_scratch 1202 14.6 1.287 1.525 1.287 1.526 fft3d_s 1 2.0 0.000 0.000 1.279 1.515 qs_ot_get_derivative_taylor 59 13.0 0.002 0.003 1.465 1.514 make_images_sizes 4572 15.5 0.007 0.014 1.006 1.469 mp_alltoall_i44 4572 16.5 0.999 1.463 0.999 1.463 wfi_extrapolate 11 7.9 0.001 0.001 1.449 1.450 qs_ot_p2m_diag 50 11.0 0.008 0.014 1.401 1.419 ot_diis_step 108 11.5 0.014 0.017 1.414 1.415 mp_alltoall_z22v 1201 15.6 1.118 1.389 1.118 1.389 mp_alltoall_d11v 2130 13.8 1.036 1.326 1.036 1.326 apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.147 1.324 apply_single 119 13.6 0.000 0.001 1.147 1.324 transfer_pw2rs 487 13.2 0.004 0.005 1.314 1.320 cp_dbcsr_syevd 50 12.0 0.004 0.004 1.138 1.143 mp_sum_d 4139 12.0 0.654 1.136 0.654 1.136 make_basis_sm 11 9.8 0.000 0.000 1.112 1.116 make_images_data 4572 15.5 0.066 0.119 0.861 1.085 cp_fm_cholesky_decompose 22 10.9 1.046 1.066 1.046 1.066 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.047 1.066 qs_energies_init_hamiltonians 11 5.9 0.000 0.000 0.909 1.033 multiply_cannon_metrocomm4 25146 15.5 0.097 0.197 0.482 0.997 create_qs_kind_set 1 2.0 0.000 0.000 0.551 0.992 read_qs_kind 2 3.0 0.014 0.018 0.551 0.992 parser_read_line 2845 4.0 0.001 0.003 0.537 0.980 parser_read_line_low 6 5.0 0.006 0.072 0.536 0.980 broadcast_input_information 6 6.0 0.007 0.008 0.530 0.976 mp_bcast_i_src 24 7.0 0.259 0.963 0.259 0.963 hybrid_alltoall_any 4725 16.4 0.068 0.150 0.725 0.946 cp_fm_diag_elpa 50 13.0 0.000 0.000 0.863 0.865 mp_irecv_dv 59094 16.3 0.346 0.854 0.346 0.854 cp_fm_redistribute_end 50 14.0 0.424 0.835 0.428 0.839 cp_fm_diag_elpa_base 50 14.0 0.365 0.766 0.407 0.818 dbcsr_complete_redistribute 329 12.2 0.117 0.223 0.751 0.816 mp_allgather_i34 2286 14.5 0.454 0.807 0.454 0.807 yz_to_x 368 14.5 0.038 0.046 0.649 0.787 cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 0.773 0.779 cp_fm_cholesky_invert 11 10.9 0.762 0.772 0.762 0.772 qs_ot_get_orbitals 108 10.5 0.001 0.001 0.739 0.750 rs_gather_matrices 119 12.3 0.037 0.048 0.442 0.737 qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.575 0.733 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="104", plot="h2o_64_md", label="(4n/12r/3t)", y=36.610000, yerr=0.000000 PlotPoint: name="105", plot="h2o_64_md_mem", label="(4n/12r/3t)", y=210.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/06/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 80 x 32 x 32 1357905920 0.0% 100.0% 0.0% flops 80 x 32 x 64 1357905920 0.0% 100.0% 0.0% flops 80 x 64 x 32 1357905920 0.0% 100.0% 0.0% flops 80 x 64 x 64 1357905920 0.0% 100.0% 0.0% flops 89 x 32 x 32 1510670336 0.0% 100.0% 0.0% flops 89 x 32 x 64 1510670336 0.0% 100.0% 0.0% flops 89 x 64 x 32 1510670336 0.0% 100.0% 0.0% flops 89 x 64 x 64 1510670336 0.0% 100.0% 0.0% flops 64 x 64 x 64 1679818752 0.0% 100.0% 0.0% flops 64 x 64 x 32 1679818752 0.0% 100.0% 0.0% flops 64 x 32 x 32 1679818752 0.0% 100.0% 0.0% flops 64 x 32 x 64 1679818752 0.0% 100.0% 0.0% flops 80 x 32 x 422 2056929280 0.0% 100.0% 0.0% flops 80 x 64 x 422 2056929280 0.0% 100.0% 0.0% flops 80 x 32 x 427 2081300480 0.0% 100.0% 0.0% flops 80 x 64 x 427 2081300480 0.0% 100.0% 0.0% flops 80 x 32 x 431 2100797440 0.0% 100.0% 0.0% flops 80 x 64 x 431 2100797440 0.0% 100.0% 0.0% flops 89 x 32 x 422 2288333824 0.0% 100.0% 0.0% flops 89 x 64 x 422 2288333824 0.0% 100.0% 0.0% flops 89 x 32 x 427 2315446784 0.0% 100.0% 0.0% flops 89 x 64 x 427 2315446784 0.0% 100.0% 0.0% flops 89 x 32 x 431 2337137152 0.0% 100.0% 0.0% flops 89 x 64 x 431 2337137152 0.0% 100.0% 0.0% flops 71 x 64 x 64 3615424512 0.0% 100.0% 0.0% flops 71 x 64 x 32 3615424512 0.0% 100.0% 0.0% flops 71 x 32 x 32 3615424512 0.0% 100.0% 0.0% flops 71 x 32 x 64 3615424512 0.0% 100.0% 0.0% flops 32 x 32 x 32 5039456256 0.0% 100.0% 0.0% flops 32 x 32 x 64 5039456256 0.0% 100.0% 0.0% flops 32 x 64 x 64 5039456256 0.0% 100.0% 0.0% flops 32 x 64 x 32 5039456256 0.0% 100.0% 0.0% flops 71 x 64 x 422 5476574208 0.0% 100.0% 0.0% flops 71 x 32 x 422 5476574208 0.0% 100.0% 0.0% flops 71 x 64 x 427 5541462528 0.0% 100.0% 0.0% flops 71 x 32 x 427 5541462528 0.0% 100.0% 0.0% flops 71 x 64 x 431 5593373184 0.0% 100.0% 0.0% flops 71 x 32 x 431 5593373184 0.0% 100.0% 0.0% flops 64 x 64 x 422 5821628416 0.0% 100.0% 0.0% flops 64 x 32 x 422 5821628416 0.0% 100.0% 0.0% flops 64 x 64 x 427 5890605056 0.0% 100.0% 0.0% flops 64 x 32 x 427 5890605056 0.0% 100.0% 0.0% flops 64 x 64 x 431 5945786368 0.0% 100.0% 0.0% flops 64 x 32 x 431 5945786368 0.0% 100.0% 0.0% flops 111 x 64 x 64 9420472320 0.0% 100.0% 0.0% flops 111 x 64 x 32 9420472320 0.0% 100.0% 0.0% flops 111 x 32 x 32 9420472320 0.0% 100.0% 0.0% flops 111 x 32 x 64 9420472320 0.0% 100.0% 0.0% flops 98 x 64 x 64 9980608512 0.0% 100.0% 0.0% flops 98 x 64 x 32 9980608512 0.0% 100.0% 0.0% flops 98 x 32 x 32 9980608512 0.0% 100.0% 0.0% flops 98 x 32 x 64 9980608512 0.0% 100.0% 0.0% flops 120 x 32 x 32 10184294400 0.0% 100.0% 0.0% flops 120 x 32 x 64 10184294400 0.0% 100.0% 0.0% flops 120 x 64 x 64 10184294400 0.0% 100.0% 0.0% flops 120 x 64 x 32 10184294400 0.0% 100.0% 0.0% flops 111 x 64 x 422 14269946880 0.0% 100.0% 0.0% flops 111 x 32 x 422 14269946880 0.0% 100.0% 0.0% flops 111 x 64 x 427 14439022080 0.0% 100.0% 0.0% flops 111 x 32 x 427 14439022080 0.0% 100.0% 0.0% flops 111 x 64 x 431 14574282240 0.0% 100.0% 0.0% flops 111 x 32 x 431 14574282240 0.0% 100.0% 0.0% flops 98 x 64 x 422 15118430208 0.0% 100.0% 0.0% flops 98 x 32 x 422 15118430208 0.0% 100.0% 0.0% flops 98 x 64 x 427 15297558528 0.0% 100.0% 0.0% flops 98 x 32 x 427 15297558528 0.0% 100.0% 0.0% flops 120 x 32 x 422 15426969600 0.0% 100.0% 0.0% flops 120 x 64 x 422 15426969600 0.0% 100.0% 0.0% flops 98 x 64 x 431 15440861184 0.0% 100.0% 0.0% flops 98 x 32 x 431 15440861184 0.0% 100.0% 0.0% flops 120 x 32 x 427 15609753600 0.0% 100.0% 0.0% flops 120 x 64 x 427 15609753600 0.0% 100.0% 0.0% flops 120 x 32 x 431 15755980800 0.0% 100.0% 0.0% flops 120 x 64 x 431 15755980800 0.0% 100.0% 0.0% flops 32 x 32 x 422 17464885248 0.0% 100.0% 0.0% flops 32 x 64 x 422 17464885248 0.0% 100.0% 0.0% flops 32 x 32 x 427 17671815168 0.0% 100.0% 0.0% flops 32 x 64 x 427 17671815168 0.0% 100.0% 0.0% flops 32 x 32 x 431 17837359104 0.0% 100.0% 0.0% flops 32 x 64 x 431 17837359104 0.0% 100.0% 0.0% flops 9 x 9 x 64 22084130304 0.0% 100.0% 0.0% flops 9 x 9 x 32 22084130304 0.0% 100.0% 0.0% flops 22 x 9 x 64 26917862400 0.0% 100.0% 0.0% flops 22 x 9 x 32 26917862400 0.0% 100.0% 0.0% flops 9 x 22 x 64 26942750208 0.0% 100.0% 0.0% flops 9 x 22 x 32 26942750208 0.0% 100.0% 0.0% flops 22 x 22 x 64 33503641600 0.0% 100.0% 0.0% flops 22 x 22 x 32 33503641600 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 112840197120 100.0% 0.0% 0.0% flops total 1.896345E+12 6.0% 94.0% 0.0% flops max/rank 83.755100E+09 11.7% 88.3% 0.0% matmuls inhomo. stacks 68796 100.0% 0.0% 0.0% matmuls total 96003990 0.1% 99.9% 0.0% number of processed stacks 2257260 3.0% 97.0% 0.0% average stack size 1.0 43.8 0.0 marketing flops 2.107629E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 232.841216E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 822960 MPI messages size (bytes): total size 161.737343E+09 min size 0.000000E+00 max size 1.486088E+06 average size 196.531234E+03 MPI breakdown and total messages size (bytes): size <= 128 5610 0 128 < size <= 8192 169820 1391165440 8192 < size <= 32768 212110 4169891840 32768 < size <= 131072 243000 26542080000 131072 < size <= 4194304 192420 129634037440 4194304 < size <= 16777216 0 0 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 95 12. MP_Allreduce 12329 16. MP_Alltoall 8655 36603. MP_ISend 54820 93714. MP_IRecv 54820 91356. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3690 72720. MP_Allreduce 10327 343. MP_Sync 54 MP_Alltoall 1843 1552068. MP_SendRecv 8330 18700. MP_ISendRecv 8330 18700. MP_Wait 31172 MP_comm_split 50 MP_ISend 20872 59666. MP_IRecv 20872 59666. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.032 0.088 33.328 33.333 qs_mol_dyn_low 1 2.0 0.021 0.071 32.708 32.803 qs_forces 11 3.9 0.002 0.003 32.218 32.223 qs_energies 11 4.9 0.001 0.002 29.863 29.875 scf_env_do_scf 11 5.9 0.001 0.003 26.552 26.552 scf_env_do_scf_inner_loop 108 6.5 0.003 0.028 23.846 23.855 velocity_verlet 10 3.0 0.001 0.001 18.065 18.070 qs_scf_new_mos 108 7.5 0.001 0.001 9.492 9.624 qs_scf_loop_do_ot 108 8.5 0.001 0.001 9.491 9.624 rebuild_ks_matrix 119 8.3 0.001 0.001 9.416 9.571 qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.017 9.415 9.570 ot_scf_mini 108 9.5 0.003 0.003 9.011 9.111 dbcsr_multiply_generic 2286 12.5 0.128 0.178 8.546 8.868 qs_ks_update_qs_env 119 7.6 0.001 0.001 8.364 8.502 qs_rho_update_rho_low 119 7.7 0.001 0.001 7.600 7.607 calculate_rho_elec 119 8.7 0.035 0.040 7.599 7.607 sum_up_and_integrate 119 10.3 0.001 0.003 7.117 7.122 integrate_v_rspace 119 11.3 0.003 0.004 7.104 7.110 multiply_cannon 2286 13.5 0.211 0.244 4.249 5.582 grid_collocate_task_list 119 9.7 5.099 5.298 5.099 5.298 ot_mini 108 10.5 0.001 0.001 4.839 4.961 grid_integrate_task_list 119 12.3 4.763 4.921 4.763 4.921 multiply_cannon_loop 2286 14.5 0.129 0.187 3.392 4.388 qs_ot_get_derivative 108 11.5 0.001 0.001 3.584 3.685 multiply_cannon_multrec 13716 15.5 2.017 3.267 2.030 3.280 qs_ot_get_p 119 10.4 0.001 0.001 2.741 2.876 init_scf_loop 11 6.9 0.000 0.000 2.680 2.700 make_m2s 4572 13.5 0.076 0.100 2.579 2.661 mp_waitall_1 156604 16.6 1.667 2.391 1.667 2.391 fft_wrap_pw1pw2 1201 11.6 0.015 0.018 2.273 2.288 init_scf_run 11 5.9 0.000 0.006 2.261 2.262 scf_env_initial_rho_setup 11 6.9 0.000 0.004 2.261 2.262 density_rs2pw 119 9.7 0.006 0.009 2.077 2.252 make_images 4572 14.5 0.263 0.316 2.090 2.170 fft_wrap_pw1pw2_140 487 12.2 0.111 0.117 1.850 1.862 fft3d_ps 1201 13.6 0.675 0.721 1.822 1.841 potential_pw2rs 119 12.3 0.009 0.012 1.770 1.775 prepare_preconditioner 11 7.9 0.000 0.000 1.506 1.516 make_preconditioner 11 8.9 0.000 0.000 1.506 1.516 mp_sum_l 11298 13.2 1.019 1.495 1.019 1.495 multiply_cannon_metrocomm3 13716 15.5 0.046 0.092 0.535 1.453 make_full_inverse_cholesky 11 9.9 0.000 0.000 1.387 1.409 wfi_extrapolate 11 7.9 0.001 0.001 1.396 1.396 qs_ot_p2m_diag 50 11.0 0.010 0.023 1.374 1.388 mp_alltoall_d11v 2130 13.8 1.219 1.374 1.219 1.374 qs_ot_get_derivative_diag 49 12.0 0.001 0.002 1.254 1.304 qs_ot_get_derivative_taylor 59 13.0 0.002 0.003 1.232 1.288 ot_diis_step 108 11.5 0.014 0.017 1.220 1.221 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.158 1.175 cp_dbcsr_syevd 50 12.0 0.004 0.004 1.158 1.163 transfer_rs2pw 487 10.6 0.006 0.007 0.996 1.158 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.790 1.153 transfer_pw2rs 487 13.2 0.005 0.006 1.143 1.150 mp_waitany 20872 13.8 0.871 1.093 0.871 1.093 apply_preconditioner_dbcsr 119 12.6 0.000 0.000 0.955 1.084 apply_single 119 13.6 0.000 0.001 0.954 1.084 mp_sum_d 4139 12.0 0.793 1.078 0.793 1.078 mp_alltoall_z22v 1201 15.6 0.989 1.043 0.989 1.043 make_images_data 4572 15.5 0.061 0.102 0.883 1.023 hybrid_alltoall_any 4725 16.4 0.072 0.201 0.741 0.966 cp_fm_diag_elpa 50 13.0 0.000 0.000 0.879 0.880 qs_energies_init_hamiltonians 11 5.9 0.000 0.002 0.871 0.877 cp_fm_redistribute_end 50 14.0 0.432 0.853 0.435 0.855 cp_fm_diag_elpa_base 50 14.0 0.395 0.800 0.418 0.838 mp_allgather_i34 2286 14.5 0.378 0.790 0.378 0.790 transfer_rs2pw_140 130 11.5 0.149 0.168 0.601 0.778 calculate_first_density_matrix 1 7.0 0.000 0.002 0.753 0.754 transfer_pw2rs_140 130 13.9 0.204 0.224 0.676 0.728 make_images_sizes 4572 15.5 0.007 0.014 0.591 0.702 mp_alltoall_i44 4572 16.5 0.584 0.696 0.584 0.696 rs_gather_matrices 119 12.3 0.056 0.062 0.533 0.689 arnoldi_extremal 119 11.4 0.002 0.003 0.648 0.684 arnoldi_normal_ev 119 12.4 0.002 0.004 0.646 0.682 cp_fm_cholesky_invert 11 10.9 0.673 0.679 0.673 0.679 dbcsr_complete_redistribute 329 12.2 0.110 0.125 0.639 0.671 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="106", plot="h2o_64_md", label="(4n/9r/4t)", y=33.333000, yerr=0.000000 PlotPoint: name="107", plot="h2o_64_md_mem", label="(4n/9r/4t)", y=221.636364, yerr=0.881396 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/07/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 40 x 32 x 200 487424000 0.0% 100.0% 0.0% flops 40 x 64 x 200 487424000 0.0% 100.0% 0.0% flops 62 x 32 x 200 755507200 0.0% 100.0% 0.0% flops 62 x 64 x 200 755507200 0.0% 100.0% 0.0% flops 76 x 64 x 200 926105600 0.0% 100.0% 0.0% flops 76 x 32 x 200 926105600 0.0% 100.0% 0.0% flops 40 x 32 x 209 1018716160 0.0% 100.0% 0.0% flops 40 x 64 x 209 1018716160 0.0% 100.0% 0.0% flops 40 x 32 x 222 1082081280 0.0% 100.0% 0.0% flops 40 x 64 x 222 1082081280 0.0% 100.0% 0.0% flops 111 x 32 x 200 1352601600 0.0% 100.0% 0.0% flops 111 x 64 x 200 1352601600 0.0% 100.0% 0.0% flops 40 x 32 x 32 1357905920 0.0% 100.0% 0.0% flops 40 x 64 x 32 1357905920 0.0% 100.0% 0.0% flops 62 x 32 x 209 1579010048 0.0% 100.0% 0.0% flops 62 x 64 x 209 1579010048 0.0% 100.0% 0.0% flops 62 x 32 x 222 1677225984 0.0% 100.0% 0.0% flops 62 x 64 x 222 1677225984 0.0% 100.0% 0.0% flops 76 x 64 x 209 1935560704 0.0% 100.0% 0.0% flops 76 x 32 x 209 1935560704 0.0% 100.0% 0.0% flops 76 x 64 x 222 2055954432 0.0% 100.0% 0.0% flops 76 x 32 x 222 2055954432 0.0% 100.0% 0.0% flops 85 x 64 x 200 2071552000 0.0% 100.0% 0.0% flops 85 x 32 x 200 2071552000 0.0% 100.0% 0.0% flops 62 x 32 x 32 2104754176 0.0% 100.0% 0.0% flops 62 x 64 x 32 2104754176 0.0% 100.0% 0.0% flops 98 x 32 x 200 2388377600 0.0% 100.0% 0.0% flops 98 x 64 x 200 2388377600 0.0% 100.0% 0.0% flops 76 x 64 x 32 2580021248 0.0% 100.0% 0.0% flops 76 x 32 x 32 2580021248 0.0% 100.0% 0.0% flops 64 x 64 x 200 2759065600 0.0% 100.0% 0.0% flops 64 x 32 x 200 2759065600 0.0% 100.0% 0.0% flops 111 x 32 x 209 2826937344 0.0% 100.0% 0.0% flops 111 x 64 x 209 2826937344 0.0% 100.0% 0.0% flops 120 x 32 x 200 2924544000 0.0% 100.0% 0.0% flops 120 x 64 x 200 2924544000 0.0% 100.0% 0.0% flops 111 x 32 x 222 3002775552 0.0% 100.0% 0.0% flops 111 x 64 x 222 3002775552 0.0% 100.0% 0.0% flops 64 x 64 x 32 3359637504 0.0% 100.0% 0.0% flops 64 x 32 x 32 3359637504 0.0% 100.0% 0.0% flops 111 x 32 x 32 3768188928 0.0% 100.0% 0.0% flops 111 x 64 x 32 3768188928 0.0% 100.0% 0.0% flops 85 x 64 x 209 4329543680 0.0% 100.0% 0.0% flops 85 x 32 x 209 4329543680 0.0% 100.0% 0.0% flops 89 x 64 x 200 4338073600 0.0% 100.0% 0.0% flops 89 x 32 x 200 4338073600 0.0% 100.0% 0.0% flops 85 x 64 x 222 4598845440 0.0% 100.0% 0.0% flops 85 x 32 x 222 4598845440 0.0% 100.0% 0.0% flops 98 x 32 x 209 4991709184 0.0% 100.0% 0.0% flops 98 x 64 x 209 4991709184 0.0% 100.0% 0.0% flops 98 x 32 x 222 5302198272 0.0% 100.0% 0.0% flops 98 x 64 x 222 5302198272 0.0% 100.0% 0.0% flops 64 x 64 x 209 5766447104 0.0% 100.0% 0.0% flops 64 x 32 x 209 5766447104 0.0% 100.0% 0.0% flops 85 x 64 x 32 5771100160 0.0% 100.0% 0.0% flops 85 x 32 x 32 5771100160 0.0% 100.0% 0.0% flops 120 x 32 x 209 6112296960 0.0% 100.0% 0.0% flops 120 x 64 x 209 6112296960 0.0% 100.0% 0.0% flops 64 x 64 x 222 6125125632 0.0% 100.0% 0.0% flops 64 x 32 x 222 6125125632 0.0% 100.0% 0.0% flops 120 x 32 x 222 6492487680 0.0% 100.0% 0.0% flops 120 x 64 x 222 6492487680 0.0% 100.0% 0.0% flops 98 x 32 x 32 6653739008 0.0% 100.0% 0.0% flops 98 x 64 x 32 6653739008 0.0% 100.0% 0.0% flops 120 x 32 x 32 8147435520 0.0% 100.0% 0.0% flops 120 x 64 x 32 8147435520 0.0% 100.0% 0.0% flops 32 x 32 x 200 8277196800 0.0% 100.0% 0.0% flops 32 x 64 x 200 8277196800 0.0% 100.0% 0.0% flops 89 x 64 x 209 9066573824 0.0% 100.0% 0.0% flops 89 x 32 x 209 9066573824 0.0% 100.0% 0.0% flops 89 x 64 x 222 9630523392 0.0% 100.0% 0.0% flops 89 x 32 x 222 9630523392 0.0% 100.0% 0.0% flops 32 x 32 x 32 10078912512 0.0% 100.0% 0.0% flops 32 x 64 x 32 10078912512 0.0% 100.0% 0.0% flops 89 x 64 x 32 12085362688 0.0% 100.0% 0.0% flops 89 x 32 x 32 12085362688 0.0% 100.0% 0.0% flops 32 x 32 x 209 17299341312 0.0% 100.0% 0.0% flops 32 x 64 x 209 17299341312 0.0% 100.0% 0.0% flops 32 x 32 x 222 18375376896 0.0% 100.0% 0.0% flops 32 x 64 x 222 18375376896 0.0% 100.0% 0.0% flops 9 x 9 x 32 44168260608 0.0% 100.0% 0.0% flops 22 x 9 x 32 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 32 53885500416 0.0% 100.0% 0.0% flops 22 x 22 x 32 67007283200 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 409643995136 100.0% 0.0% 0.0% flops total 1.940194E+12 21.1% 78.9% 0.0% flops max/rank 126.888791E+09 24.8% 75.2% 0.0% matmuls inhomo. stacks 389676 100.0% 0.0% 0.0% matmuls total 101225376 0.4% 99.6% 0.0% number of processed stacks 3787752 10.3% 89.7% 0.0% average stack size 1.0 29.7 0.0 marketing flops 2.107592E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 273.178624E+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 3679 72932. MP_Allreduce 10304 342. MP_Sync 54 MP_Alltoall 1582 2412273. MP_SendRecv 8211 74133. MP_ISendRecv 8211 74133. MP_Wait 16271 MP_comm_split 50 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 0.010 0.057 43.397 43.432 qs_mol_dyn_low 1 2.0 0.016 0.030 43.052 43.102 qs_forces 11 3.9 0.002 0.003 42.345 42.399 qs_energies 11 4.9 0.001 0.001 40.216 40.268 scf_env_do_scf 11 5.9 0.001 0.003 36.233 36.262 scf_env_do_scf_inner_loop 108 6.5 0.003 0.027 31.622 31.649 velocity_verlet 10 3.0 0.001 0.001 24.119 24.150 qs_scf_new_mos 108 7.5 0.001 0.001 14.157 14.268 qs_scf_loop_do_ot 108 8.5 0.001 0.001 14.156 14.268 ot_scf_mini 108 9.5 0.003 0.004 13.470 13.558 dbcsr_multiply_generic 2286 12.5 0.148 0.195 11.870 12.320 rebuild_ks_matrix 119 8.3 0.001 0.001 10.428 10.623 qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.019 10.428 10.622 qs_rho_update_rho_low 119 7.7 0.001 0.001 10.084 10.098 calculate_rho_elec 119 8.7 0.051 0.059 10.083 10.098 qs_ks_update_qs_env 119 7.6 0.001 0.001 9.297 9.476 sum_up_and_integrate 119 10.3 0.001 0.002 7.782 7.790 integrate_v_rspace 119 11.3 0.003 0.003 7.769 7.778 ot_mini 108 10.5 0.001 0.001 7.051 7.156 grid_collocate_task_list 119 9.7 6.738 6.877 6.738 6.877 multiply_cannon 2286 13.5 0.231 0.269 5.384 6.413 qs_ot_get_derivative 108 11.5 0.001 0.001 5.566 5.657 multiply_cannon_loop 2286 14.5 0.249 0.351 4.243 5.392 grid_integrate_task_list 119 12.3 4.860 4.973 4.860 4.973 qs_ot_get_p 119 10.4 0.001 0.001 4.521 4.631 init_scf_loop 11 6.9 0.000 0.000 4.572 4.577 make_m2s 4572 13.5 0.098 0.130 4.269 4.400 mp_waitall_1 137007 16.6 2.897 3.703 2.897 3.703 make_images 4572 14.5 0.350 0.411 3.451 3.546 multiply_cannon_multrec 27432 15.5 2.367 3.457 2.382 3.471 prepare_preconditioner 11 7.9 0.000 0.000 3.315 3.326 make_preconditioner 11 8.9 0.000 0.000 3.315 3.326 make_full_inverse_cholesky 11 9.9 0.000 0.000 2.838 3.206 density_rs2pw 119 9.7 0.006 0.008 3.006 3.187 init_scf_run 11 5.9 0.000 0.006 2.748 2.750 scf_env_initial_rho_setup 11 6.9 0.000 0.004 2.747 2.750 fft_wrap_pw1pw2 1201 11.6 0.015 0.018 2.668 2.713 qs_ot_p2m_diag 50 11.0 0.014 0.023 2.672 2.691 potential_pw2rs 119 12.3 0.012 0.014 2.497 2.522 cp_dbcsr_syevd 50 12.0 0.003 0.004 2.342 2.345 fft3d_ps 1201 13.6 0.732 0.796 2.188 2.238 fft_wrap_pw1pw2_140 487 12.2 0.119 0.133 2.068 2.117 qs_ot_get_derivative_diag 49 12.0 0.001 0.002 2.002 2.049 qs_ot_get_derivative_taylor 59 13.0 0.002 0.003 1.895 1.940 cp_fm_diag_elpa 50 13.0 0.000 0.000 1.933 1.935 transfer_rs2pw 487 10.6 0.005 0.007 1.733 1.934 cp_fm_redistribute_end 50 14.0 0.949 1.881 0.975 1.887 wfi_extrapolate 11 7.9 0.001 0.001 1.882 1.883 cp_fm_diag_elpa_base 50 14.0 0.854 1.742 0.905 1.815 mp_sum_l 11298 13.2 1.340 1.787 1.340 1.787 transfer_pw2rs 487 13.2 0.004 0.005 1.714 1.736 dbcsr_complete_redistribute 329 12.2 0.151 0.171 1.282 1.600 make_images_data 4572 15.5 0.067 0.117 1.308 1.580 make_images_sizes 4572 15.5 0.007 0.014 1.139 1.523 mp_alltoall_i44 4572 16.5 1.132 1.517 1.132 1.517 ot_diis_step 108 11.5 0.016 0.018 1.434 1.435 mp_alltoall_z22v 1201 15.6 1.322 1.366 1.322 1.366 multiply_cannon_metrocomm3 27432 15.5 0.056 0.111 0.618 1.361 hybrid_alltoall_any 4725 16.4 0.082 0.163 1.063 1.345 mp_alltoall_d11v 2130 13.8 1.148 1.308 1.148 1.308 apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.129 1.296 apply_single 119 13.6 0.000 0.001 1.129 1.295 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.252 1.268 cp_fm_upper_to_full 72 13.8 0.923 1.256 0.923 1.256 copy_fm_to_dbcsr 176 11.2 0.001 0.002 0.817 1.128 cp_fm_cholesky_invert 11 10.9 1.028 1.039 1.028 1.039 qs_energies_init_hamiltonians 11 5.9 0.000 0.000 0.974 0.994 multiply_cannon_metrocomm1 27432 15.5 0.040 0.082 0.414 0.973 mp_sendrecv_dv 8211 12.7 0.936 0.965 0.936 0.965 mp_alltoall_i22 627 13.8 0.604 0.963 0.604 0.963 arnoldi_extremal 119 11.4 0.002 0.003 0.875 0.919 arnoldi_normal_ev 119 12.4 0.002 0.004 0.873 0.916 multiply_cannon_metrocomm4 22860 15.5 0.088 0.179 0.378 0.882 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="108", plot="h2o_64_md", label="(4n/6r/6t)", y=43.432000, yerr=0.000000 PlotPoint: name="109", plot="h2o_64_md_mem", label="(4n/6r/6t)", y=258.181818, yerr=3.638636 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/08/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 40 x 64 x 64 10863247360 0.0% 100.0% 0.0% flops 40 x 64 x 640 24956108800 0.0% 100.0% 0.0% flops 32 x 64 x 64 26877100032 0.0% 100.0% 0.0% flops 9 x 9 x 64 44168260608 0.0% 100.0% 0.0% flops 71 x 64 x 64 48205660160 0.0% 100.0% 0.0% flops 22 x 9 x 64 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 64 53885500416 0.0% 100.0% 0.0% flops 80 x 64 x 64 54316236800 0.0% 100.0% 0.0% flops 22 x 22 x 64 67007283200 0.0% 100.0% 0.0% flops 71 x 64 x 640 110742732800 0.0% 100.0% 0.0% flops 80 x 64 x 640 124780544000 0.0% 100.0% 0.0% flops 32 x 64 x 640 141264158720 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 261929041920 100.0% 0.0% 0.0% flops total 1.943572E+12 13.5% 86.5% 0.0% flops max/rank 122.902337E+09 13.9% 86.1% 0.0% matmuls inhomo. stacks 122304 100.0% 0.0% 0.0% matmuls total 90872996 0.1% 99.9% 0.0% number of processed stacks 1449216 8.4% 91.6% 0.0% average stack size 1.0 68.4 0.0 marketing flops 2.107587E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 329.531392E+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 19 12. MP_Allreduce 12177 16. MP_Alltoall 8655 62672. MP_ISend 36532 167957. MP_IRecv 36532 167930. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3629 73907. 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 0.034 0.092 37.601 37.603 qs_mol_dyn_low 1 2.0 0.056 0.110 36.654 37.053 qs_forces 11 3.9 0.008 0.026 35.916 35.956 qs_energies 11 4.9 0.001 0.001 33.857 33.896 scf_env_do_scf 11 5.9 0.001 0.002 29.425 29.426 scf_env_do_scf_inner_loop 108 6.5 0.004 0.025 25.375 25.418 velocity_verlet 10 3.0 0.002 0.003 19.113 19.123 qs_scf_new_mos 108 7.5 0.001 0.001 9.755 9.823 qs_scf_loop_do_ot 108 8.5 0.001 0.001 9.755 9.822 rebuild_ks_matrix 119 8.3 0.001 0.001 9.339 9.384 qs_ks_build_kohn_sham_matrix 119 9.3 0.015 0.017 9.339 9.383 ot_scf_mini 108 9.5 0.003 0.004 9.277 9.326 qs_rho_update_rho_low 119 7.7 0.001 0.001 9.216 9.218 calculate_rho_elec 119 8.7 0.075 0.078 9.215 9.218 qs_ks_update_qs_env 119 7.6 0.001 0.002 8.351 8.390 dbcsr_multiply_generic 2286 12.5 0.135 0.150 7.649 7.748 sum_up_and_integrate 119 10.3 0.001 0.002 6.798 6.814 integrate_v_rspace 119 11.3 0.003 0.003 6.787 6.802 grid_collocate_task_list 119 9.7 6.207 6.606 6.207 6.606 grid_integrate_task_list 119 12.3 4.551 4.716 4.551 4.716 qs_ot_get_p 119 10.4 0.001 0.001 4.292 4.392 ot_mini 108 10.5 0.001 0.002 4.192 4.247 multiply_cannon 2286 13.5 0.215 0.220 3.707 4.125 init_scf_loop 11 6.9 0.000 0.000 4.026 4.081 density_rs2pw 119 9.7 0.005 0.005 2.652 3.482 multiply_cannon_loop 2286 14.5 0.096 0.115 2.990 3.164 init_scf_run 11 5.9 0.000 0.005 3.099 3.099 scf_env_initial_rho_setup 11 6.9 0.001 0.003 3.099 3.099 qs_ot_get_derivative 108 11.5 0.001 0.003 2.964 3.014 fft_wrap_pw1pw2 1201 11.6 0.013 0.016 2.799 2.873 make_m2s 4572 13.5 0.070 0.079 2.576 2.719 multiply_cannon_multrec 9144 15.5 2.231 2.457 2.245 2.471 fft_wrap_pw1pw2_140 487 12.2 0.153 0.168 2.310 2.460 prepare_preconditioner 11 7.9 0.000 0.000 2.339 2.347 make_preconditioner 11 8.9 0.000 0.000 2.339 2.346 qs_ot_p2m_diag 50 11.0 0.021 0.024 2.266 2.289 fft3d_ps 1201 13.6 0.834 0.988 2.247 2.288 make_full_inverse_cholesky 11 9.9 0.000 0.000 2.176 2.223 cp_dbcsr_syevd 50 12.0 0.004 0.004 2.093 2.114 transfer_rs2pw 487 10.6 0.005 0.006 1.258 2.100 make_images 4572 14.5 0.333 0.372 1.963 2.086 cp_fm_diag_elpa 50 13.0 0.000 0.000 1.818 1.819 cp_fm_diag_elpa_base 50 14.0 1.774 1.791 1.816 1.816 mp_waitany 5200 13.7 0.915 1.727 0.915 1.727 potential_pw2rs 119 12.3 0.014 0.015 1.671 1.676 transfer_rs2pw_140 130 11.5 0.117 0.137 0.835 1.674 wfi_extrapolate 11 7.9 0.001 0.001 1.524 1.525 arnoldi_extremal 119 11.4 0.002 0.002 1.389 1.509 arnoldi_normal_ev 119 12.4 0.002 0.003 1.387 1.506 mp_waitall_1 115863 16.7 1.240 1.491 1.240 1.491 mp_alltoall_z22v 1201 15.6 1.240 1.354 1.240 1.354 mp_alltoall_d11v 2130 13.8 1.048 1.318 1.048 1.318 calculate_first_density_matrix 1 7.0 0.000 0.000 1.305 1.305 ot_diis_step 108 11.5 0.023 0.042 1.213 1.214 qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.110 1.150 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.140 1.146 cp_fm_cholesky_invert 11 10.9 1.111 1.120 1.111 1.120 qs_ot_get_derivative_taylor 59 13.0 0.002 0.002 1.005 1.051 make_images_data 4572 15.5 0.055 0.059 0.885 1.043 apply_preconditioner_dbcsr 119 12.6 0.000 0.001 0.981 1.018 apply_single 119 13.6 0.000 0.000 0.981 1.018 qs_ot_get_derivative_diag 49 12.0 0.001 0.001 0.981 0.998 hybrid_alltoall_any 4725 16.4 0.084 0.217 0.825 0.977 transfer_pw2rs 487 13.2 0.004 0.004 0.919 0.921 compute_evals 119 13.4 0.714 0.844 0.714 0.844 make_basis_sm 11 9.8 0.000 0.000 0.824 0.829 dbcsr_complete_redistribute 329 12.2 0.207 0.229 0.770 0.821 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.717 0.820 rs_gather_matrices 119 12.3 0.054 0.067 0.525 0.810 yz_to_x 606 14.1 0.079 0.102 0.764 0.808 cp_fm_cholesky_decompose 22 10.9 0.781 0.784 0.781 0.784 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="110", plot="h2o_64_md", label="(4n/4r/9t)", y=37.603000, yerr=0.000000 PlotPoint: name="111", plot="h2o_64_md_mem", label="(4n/4r/9t)", y=310.545455, yerr=5.314879 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/09/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 89 x 32 x 409 2217840128 0.0% 100.0% 0.0% flops 89 x 64 x 409 2217840128 0.0% 100.0% 0.0% flops 89 x 32 x 418 2266643456 0.0% 100.0% 0.0% flops 89 x 64 x 418 2266643456 0.0% 100.0% 0.0% flops 107 x 64 x 409 2666392064 0.0% 100.0% 0.0% flops 107 x 32 x 409 2666392064 0.0% 100.0% 0.0% flops 107 x 64 x 418 2725065728 0.0% 100.0% 0.0% flops 107 x 32 x 418 2725065728 0.0% 100.0% 0.0% flops 120 x 32 x 409 2990346240 0.0% 100.0% 0.0% flops 120 x 64 x 409 2990346240 0.0% 100.0% 0.0% flops 89 x 32 x 32 3021340672 0.0% 100.0% 0.0% flops 89 x 32 x 64 3021340672 0.0% 100.0% 0.0% flops 89 x 64 x 32 3021340672 0.0% 100.0% 0.0% flops 89 x 64 x 64 3021340672 0.0% 100.0% 0.0% flops 120 x 32 x 418 3056148480 0.0% 100.0% 0.0% flops 120 x 64 x 418 3056148480 0.0% 100.0% 0.0% flops 142 x 64 x 409 3538576384 0.0% 100.0% 0.0% flops 142 x 32 x 409 3538576384 0.0% 100.0% 0.0% flops 142 x 64 x 418 3616442368 0.0% 100.0% 0.0% flops 142 x 32 x 418 3616442368 0.0% 100.0% 0.0% flops 107 x 64 x 64 3632398336 0.0% 100.0% 0.0% flops 107 x 64 x 32 3632398336 0.0% 100.0% 0.0% flops 107 x 32 x 32 3632398336 0.0% 100.0% 0.0% flops 107 x 32 x 64 3632398336 0.0% 100.0% 0.0% flops 120 x 32 x 32 4073717760 0.0% 100.0% 0.0% flops 120 x 32 x 64 4073717760 0.0% 100.0% 0.0% flops 120 x 64 x 32 4073717760 0.0% 100.0% 0.0% flops 120 x 64 x 64 4073717760 0.0% 100.0% 0.0% flops 142 x 64 x 64 4820566016 0.0% 100.0% 0.0% flops 142 x 64 x 32 4820566016 0.0% 100.0% 0.0% flops 142 x 32 x 32 4820566016 0.0% 100.0% 0.0% flops 142 x 32 x 64 4820566016 0.0% 100.0% 0.0% flops 111 x 64 x 409 5532140544 0.0% 100.0% 0.0% flops 111 x 32 x 409 5532140544 0.0% 100.0% 0.0% flops 111 x 64 x 418 5653874688 0.0% 100.0% 0.0% flops 111 x 32 x 418 5653874688 0.0% 100.0% 0.0% flops 32 x 64 x 64 6719275008 0.0% 100.0% 0.0% flops 32 x 64 x 32 6719275008 0.0% 100.0% 0.0% flops 32 x 32 x 32 6719275008 0.0% 100.0% 0.0% flops 32 x 32 x 64 6719275008 0.0% 100.0% 0.0% flops 89 x 32 x 431 7011411456 0.0% 100.0% 0.0% flops 89 x 64 x 431 7011411456 0.0% 100.0% 0.0% flops 111 x 64 x 64 7536377856 0.0% 100.0% 0.0% flops 111 x 64 x 32 7536377856 0.0% 100.0% 0.0% flops 111 x 32 x 32 7536377856 0.0% 100.0% 0.0% flops 111 x 32 x 64 7536377856 0.0% 100.0% 0.0% flops 107 x 64 x 431 8429449728 0.0% 100.0% 0.0% flops 107 x 32 x 431 8429449728 0.0% 100.0% 0.0% flops 120 x 32 x 431 9453588480 0.0% 100.0% 0.0% flops 120 x 64 x 431 9453588480 0.0% 100.0% 0.0% flops 142 x 64 x 431 11186746368 0.0% 100.0% 0.0% flops 142 x 32 x 431 11186746368 0.0% 100.0% 0.0% flops 32 x 64 x 409 11284578304 0.0% 100.0% 0.0% flops 32 x 32 x 409 11284578304 0.0% 100.0% 0.0% flops 32 x 64 x 418 11532894208 0.0% 100.0% 0.0% flops 32 x 32 x 418 11532894208 0.0% 100.0% 0.0% flops 111 x 64 x 431 17489138688 0.0% 100.0% 0.0% flops 111 x 32 x 431 17489138688 0.0% 100.0% 0.0% flops 9 x 9 x 64 22084130304 0.0% 100.0% 0.0% flops 9 x 9 x 32 22084130304 0.0% 100.0% 0.0% flops 22 x 9 x 64 26917862400 0.0% 100.0% 0.0% flops 22 x 9 x 32 26917862400 0.0% 100.0% 0.0% flops 9 x 22 x 64 26942750208 0.0% 100.0% 0.0% flops 9 x 22 x 32 26942750208 0.0% 100.0% 0.0% flops 22 x 22 x 64 33503641600 0.0% 100.0% 0.0% flops 22 x 22 x 32 33503641600 0.0% 100.0% 0.0% flops 32 x 64 x 431 35674718208 0.0% 100.0% 0.0% flops 32 x 32 x 431 35674718208 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 555803951104 100.0% 0.0% 0.0% flops total 2.107308E+12 26.4% 73.6% 0.0% flops max/rank 262.290333E+09 28.3% 71.7% 0.0% matmuls inhomo. stacks 303960 100.0% 0.0% 0.0% matmuls total 96047982 0.3% 99.7% 0.0% number of processed stacks 2272806 13.4% 86.6% 0.0% average stack size 1.0 48.6 0.0 marketing flops 2.107592E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 399.904768E+06 # max total images/rank 3 # max 3D layers 1 # MPI messages exchanged 219456 MPI messages size (bytes): total size 139.149754E+09 min size 0.000000E+00 max size 4.537280E+06 average size 634.066750E+03 MPI breakdown and total messages size (bytes): size <= 128 1386 0 128 < size <= 8192 21148 173244416 8192 < size <= 32768 58442 1568899072 32768 < size <= 131072 38700 3527147520 131072 < size <= 4194304 78110 38969991440 4194304 < size <= 16777216 21670 94910778800 16777216 < size 0 0 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3679 73163. MP_Allreduce 10304 424. MP_Sync 54 MP_Alltoall 1582 4823651. MP_SendRecv 3927 131600. MP_ISendRecv 3927 131600. MP_Wait 8867 MP_comm_split 50 MP_ISend 4160 325000. MP_IRecv 4160 325000. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.382 0.576 45.406 45.410 qs_mol_dyn_low 1 2.0 0.052 0.077 44.101 44.633 qs_forces 11 3.9 0.008 0.012 42.883 42.968 qs_energies 11 4.9 0.009 0.025 40.647 40.728 scf_env_do_scf 11 5.9 0.003 0.008 36.648 36.650 scf_env_do_scf_inner_loop 108 6.5 0.006 0.027 30.492 30.493 velocity_verlet 10 3.0 0.002 0.003 25.370 25.393 qs_scf_new_mos 108 7.5 0.001 0.001 12.928 13.002 qs_scf_loop_do_ot 108 8.5 0.001 0.002 12.927 13.001 ot_scf_mini 108 9.5 0.004 0.005 12.290 12.361 dbcsr_multiply_generic 2286 12.5 0.149 0.197 12.052 12.184 rebuild_ks_matrix 119 8.3 0.001 0.001 10.340 10.447 qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.017 10.339 10.446 qs_rho_update_rho_low 119 7.7 0.001 0.001 10.291 10.315 calculate_rho_elec 119 8.7 0.100 0.105 10.290 10.314 qs_ks_update_qs_env 119 7.6 0.001 0.001 9.279 9.380 grid_collocate_task_list 119 9.7 7.563 7.728 7.563 7.728 sum_up_and_integrate 119 10.3 0.001 0.002 7.080 7.095 integrate_v_rspace 119 11.3 0.003 0.003 7.067 7.083 multiply_cannon 2286 13.5 0.233 0.268 6.066 6.860 ot_mini 108 10.5 0.002 0.002 6.641 6.734 init_scf_loop 11 6.9 0.002 0.003 6.118 6.120 multiply_cannon_loop 2286 14.5 0.146 0.196 4.917 5.341 grid_integrate_task_list 119 12.3 4.939 5.065 4.939 5.065 prepare_preconditioner 11 7.9 0.000 0.001 4.786 4.795 make_preconditioner 11 8.9 0.003 0.012 4.786 4.795 make_full_inverse_cholesky 11 9.9 0.000 0.000 4.086 4.663 qs_ot_get_derivative 108 11.5 0.001 0.002 4.458 4.534 mp_waitall_1 114435 16.7 3.056 4.497 3.056 4.497 make_m2s 4572 13.5 0.089 0.122 4.071 4.391 qs_ot_get_p 119 10.4 0.001 0.001 3.473 3.564 multiply_cannon_multrec 13716 15.5 2.291 3.444 2.310 3.462 make_images 4572 14.5 0.464 0.510 3.089 3.454 multiply_cannon_metrocomm3 13716 15.5 0.052 0.094 1.856 3.124 fft_wrap_pw1pw2 1201 11.6 0.014 0.017 2.694 2.740 density_rs2pw 119 9.7 0.005 0.008 2.289 2.492 cp_fm_upper_to_full 72 13.8 1.812 2.359 1.812 2.359 init_scf_run 11 5.9 0.001 0.005 2.319 2.320 scf_env_initial_rho_setup 11 6.9 0.003 0.007 2.318 2.319 fft_wrap_pw1pw2_140 487 12.2 0.176 0.181 2.267 2.313 ot_diis_step 108 11.5 0.019 0.021 2.159 2.160 fft3d_ps 1201 13.6 0.895 0.960 2.106 2.145 qs_ot_p2m_diag 50 11.0 0.026 0.042 1.905 1.928 dbcsr_complete_redistribute 329 12.2 0.255 0.284 1.402 1.890 wfi_extrapolate 11 7.9 0.001 0.001 1.863 1.864 apply_preconditioner_dbcsr 119 12.6 0.001 0.001 1.750 1.856 apply_single 119 13.6 0.000 0.001 1.749 1.855 cp_dbcsr_syevd 50 12.0 0.004 0.004 1.639 1.641 qs_ot_get_derivative_diag 49 12.0 0.001 0.002 1.598 1.639 potential_pw2rs 119 12.3 0.016 0.018 1.605 1.616 make_images_data 4572 15.5 0.069 0.115 1.294 1.584 hybrid_alltoall_any 4725 16.4 0.103 0.161 1.188 1.511 md_output 10 3.0 0.001 0.001 0.923 1.502 qs_ot_get_derivative_taylor 59 13.0 0.002 0.003 1.440 1.476 copy_fm_to_dbcsr 176 11.2 0.001 0.002 0.976 1.475 mp_sum_l 11298 13.2 0.932 1.463 0.932 1.463 mp_max_l 33 2.8 0.373 1.463 0.373 1.463 write_restart 10 4.0 0.028 0.321 0.913 1.435 cp_fm_cholesky_invert 11 10.9 1.414 1.431 1.414 1.431 qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.331 1.418 cp_fm_diag_elpa 50 13.0 0.000 0.000 1.346 1.347 cp_fm_redistribute_end 50 14.0 0.667 1.322 0.669 1.324 cp_fm_diag_elpa_base 50 14.0 0.476 1.012 0.650 1.303 update_input 1 5.0 0.169 0.227 0.884 1.291 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.256 1.266 mp_alltoall_i22 627 13.8 0.753 1.263 0.753 1.263 mp_alltoall_d11v 2130 13.8 1.119 1.216 1.119 1.216 multiply_cannon_metrocomm4 11430 15.5 0.050 0.094 0.498 1.204 transfer_rs2pw 487 10.6 0.006 0.006 0.978 1.186 transfer_fm_to_dbcsr 11 9.9 0.000 0.000 0.690 1.168 mp_alltoall_z22v 1201 15.6 1.067 1.137 1.067 1.137 mp_irecv_dv 29167 16.4 0.419 1.106 0.419 1.106 update_motion 1 6.0 0.343 0.457 0.714 1.063 mp_allgather_i34 2286 14.5 0.489 1.024 0.489 1.024 dbcsr_make_images_dense 3978 14.8 0.055 0.081 0.749 0.954 make_images_sizes 4572 15.5 0.007 0.014 0.782 0.940 mp_alltoall_i44 4572 16.5 0.775 0.933 0.775 0.933 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.875 0.930 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="112", plot="h2o_64_md", label="(4n/3r/12t)", y=45.410000, yerr=0.000000 PlotPoint: name="113", plot="h2o_64_md_mem", label="(4n/3r/12t)", y=376.727273, yerr=7.136219 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/10/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 58 x 64 x 64 15751708672 0.0% 100.0% 0.0% flops 32 x 64 x 64 26877100032 0.0% 100.0% 0.0% flops 58 x 64 x 640 36186357760 0.0% 100.0% 0.0% flops 71 x 64 x 64 38564528128 0.0% 100.0% 0.0% flops 80 x 64 x 64 43452989440 0.0% 100.0% 0.0% flops 9 x 9 x 64 44168260608 0.0% 100.0% 0.0% flops 22 x 9 x 64 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 64 53885500416 0.0% 100.0% 0.0% flops 22 x 22 x 64 67007283200 0.0% 100.0% 0.0% flops 71 x 64 x 640 88594186240 0.0% 100.0% 0.0% flops 80 x 64 x 640 99824435200 0.0% 100.0% 0.0% flops 32 x 64 x 640 141264158720 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 342522593280 100.0% 0.0% 0.0% flops total 1.972676E+12 17.4% 82.6% 0.0% flops max/rank 253.383579E+09 18.0% 82.0% 0.0% matmuls inhomo. stacks 152880 100.0% 0.0% 0.0% matmuls total 90862804 0.2% 99.8% 0.0% number of processed stacks 1405732 10.9% 89.1% 0.0% average stack size 1.0 72.4 0.0 marketing flops 2.107587E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 1.001234E+09 # 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 3629 74142. 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 0.074 0.099 48.265 48.266 qs_mol_dyn_low 1 2.0 0.046 0.052 47.654 47.659 qs_forces 11 3.9 0.007 0.008 46.961 46.966 qs_energies 11 4.9 0.005 0.006 44.245 44.251 scf_env_do_scf 11 5.9 0.002 0.004 39.718 39.718 scf_env_do_scf_inner_loop 108 6.5 0.006 0.025 30.101 30.115 velocity_verlet 10 3.0 0.281 0.376 29.209 29.222 qs_rho_update_rho_low 119 7.7 0.001 0.002 11.485 11.488 calculate_rho_elec 119 8.7 0.149 0.153 11.483 11.487 qs_scf_new_mos 108 7.5 0.001 0.001 11.319 11.383 qs_scf_loop_do_ot 108 8.5 0.001 0.002 11.318 11.382 rebuild_ks_matrix 119 8.3 0.000 0.000 10.965 11.021 qs_ks_build_kohn_sham_matrix 119 9.3 0.015 0.017 10.965 11.021 ot_scf_mini 108 9.5 0.010 0.027 10.655 10.709 dbcsr_multiply_generic 2286 12.5 0.141 0.148 10.552 10.654 init_scf_loop 11 6.9 0.270 0.361 9.584 9.617 qs_ks_update_qs_env 119 7.6 0.001 0.001 9.523 9.571 grid_collocate_task_list 119 9.7 8.225 8.430 8.225 8.430 prepare_preconditioner 11 7.9 0.000 0.000 7.754 7.776 make_preconditioner 11 8.9 0.003 0.010 7.754 7.776 make_full_inverse_cholesky 11 9.9 0.000 0.000 6.358 7.604 sum_up_and_integrate 119 10.3 0.001 0.001 7.343 7.549 integrate_v_rspace 119 11.3 0.003 0.003 7.331 7.536 multiply_cannon 2286 13.5 0.237 0.241 5.408 5.833 ot_mini 108 10.5 0.002 0.002 5.585 5.643 grid_integrate_task_list 119 12.3 5.049 5.135 5.049 5.135 multiply_cannon_loop 2286 14.5 0.098 0.102 4.517 4.704 cp_fm_upper_to_full 72 14.2 3.100 4.386 3.100 4.386 make_m2s 4572 13.5 0.073 0.075 3.583 3.873 mp_waitall_1 94719 16.7 2.851 3.839 2.851 3.839 qs_ot_get_derivative 108 11.5 0.001 0.002 3.608 3.665 fft_wrap_pw1pw2 1201 11.6 0.014 0.015 3.202 3.433 qs_ot_get_p 119 10.4 0.001 0.002 3.185 3.272 dbcsr_complete_redistribute 329 12.2 0.347 0.384 2.263 3.249 fft_wrap_pw1pw2_140 487 12.2 0.245 0.247 2.737 2.966 make_images 4572 14.5 0.546 0.571 2.606 2.910 multiply_cannon_multrec 9144 15.5 2.470 2.885 2.492 2.905 density_rs2pw 119 9.7 0.005 0.005 2.685 2.848 copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.788 2.772 fft3d_ps 1201 13.6 1.061 1.095 2.497 2.724 init_scf_run 11 5.9 0.001 0.005 2.683 2.683 scf_env_initial_rho_setup 11 6.9 0.006 0.007 2.682 2.683 mp_alltoall_i22 627 13.8 1.477 2.451 1.477 2.451 multiply_cannon_metrocomm3 9144 15.5 0.026 0.026 1.559 2.282 transfer_fm_to_dbcsr 11 9.9 0.000 0.001 1.385 2.126 wfi_extrapolate 11 7.9 0.001 0.001 2.121 2.121 cp_fm_cholesky_invert 11 10.9 1.977 1.984 1.977 1.984 ot_diis_step 108 11.5 0.022 0.025 1.966 1.966 potential_pw2rs 119 12.3 0.017 0.018 1.729 1.927 qs_ot_p2m_diag 50 11.0 0.039 0.043 1.804 1.808 apply_preconditioner_dbcsr 119 12.6 0.001 0.001 1.659 1.712 apply_single 119 13.6 0.000 0.000 1.658 1.711 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.693 1.704 hybrid_alltoall_any 4725 16.4 0.112 0.186 1.369 1.698 make_images_data 4572 15.5 0.058 0.062 1.350 1.669 cp_dbcsr_syevd 50 12.0 0.003 0.004 1.600 1.600 qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.581 1.586 mp_alltoall_z22v 1201 15.6 1.281 1.500 1.281 1.500 mp_alltoall_d11v 2130 13.8 1.379 1.427 1.379 1.427 cp_fm_diag_elpa 50 13.0 0.000 0.000 1.307 1.308 cp_fm_diag_elpa_base 50 14.0 1.183 1.200 1.305 1.306 qs_ot_get_derivative_taylor 59 13.0 0.002 0.002 1.215 1.259 qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.165 1.188 yz_to_x 606 14.1 0.078 0.085 0.936 1.147 transfer_rs2pw 487 10.6 0.006 0.007 1.001 1.143 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="114", plot="h2o_64_md", label="(4n/2r/18t)", y=48.266000, yerr=0.000000 PlotPoint: name="115", plot="h2o_64_md_mem", label="(4n/2r/18t)", y=819.454545, yerr=116.199174 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/11/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 207618048 0.0% 100.0% 0.0% flops 22 x 32 x 32 253755392 0.0% 100.0% 0.0% flops 18 x 128 x 128 8554807296 0.0% 100.0% 0.0% flops 18 x 128 x 1280 19652935680 0.0% 100.0% 0.0% flops 53 x 128 x 128 21590704128 0.0% 100.0% 0.0% flops 32 x 128 x 128 26877100032 0.0% 100.0% 0.0% flops 31 x 128 x 128 31571312640 0.0% 100.0% 0.0% flops 9 x 9 x 128 44168260608 0.0% 100.0% 0.0% flops 40 x 128 x 128 46168801280 0.0% 100.0% 0.0% flops 53 x 128 x 1280 49600266240 0.0% 100.0% 0.0% flops 22 x 9 x 128 53835724800 0.0% 100.0% 0.0% flops 9 x 22 x 128 53885500416 0.0% 100.0% 0.0% flops 22 x 22 x 128 67007283200 0.0% 100.0% 0.0% flops 31 x 128 x 1280 72528691200 0.0% 100.0% 0.0% flops 40 x 128 x 1280 106063462400 0.0% 100.0% 0.0% flops 32 x 128 x 1280 141264158720 0.0% 100.0% 0.0% flops 9 x 32 x 9 185405884416 0.0% 100.0% 0.0% flops 22 x 32 x 9 227871249408 0.0% 100.0% 0.0% flops 9 x 32 x 22 227871249408 0.0% 100.0% 0.0% flops 22 x 32 x 22 279130931200 0.0% 100.0% 0.0% flops inhomo. stacks 320807108608 100.0% 0.0% 0.0% flops total 1.984317E+12 16.2% 83.8% 0.0% flops max/rank 515.157433E+09 16.2% 83.8% 0.0% matmuls inhomo. stacks 63700 100.0% 0.0% 0.0% matmuls total 85771122 0.1% 99.9% 0.0% number of processed stacks 614366 10.4% 89.6% 0.0% average stack size 1.0 155.6 0.0 marketing flops 2.107587E+12 ------------------------------------------------------------------------------- # multiplications 2286 max memory usage/rank 4.504183E+09 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 18288 MPI messages size (bytes): total size 32.347431E+09 min size 0.000000E+00 max size 13.107200E+06 average size 1.768779E+06 MPI breakdown and total messages size (bytes): size <= 128 110 0 128 < size <= 8192 0 0 8192 < size <= 32768 22 720896 32768 < size <= 131072 8480 1111490560 131072 < size <= 4194304 8100 10616832000 4194304 < size <= 16777216 1576 20618391488 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 21 12. MP_Allreduce 12181 16. MP_Alltoall 8655 275585. MP_ISend 18244 454716. MP_IRecv 18244 454782. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3618 74366. MP_Allreduce 10132 514. MP_Sync 54 MP_Alltoall 1201 14305296. MP_SendRecv 1461 1332738. MP_ISendRecv 1461 1332738. MP_Wait 1461 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.070 0.091 61.499 61.501 qs_mol_dyn_low 1 2.0 0.086 0.090 60.035 60.041 qs_forces 11 3.9 0.064 0.065 59.052 59.052 qs_energies 11 4.9 0.084 0.085 55.166 55.170 scf_env_do_scf 11 5.9 0.109 0.116 47.578 47.578 scf_env_do_scf_inner_loop 108 6.5 0.032 0.047 39.369 39.370 velocity_verlet 10 3.0 0.002 0.003 34.027 34.040 qs_rho_update_rho_low 119 7.7 0.001 0.002 16.826 16.830 calculate_rho_elec 119 8.7 0.318 0.325 16.824 16.829 qs_scf_new_mos 108 7.5 0.027 0.028 14.681 14.701 qs_scf_loop_do_ot 108 8.5 0.016 0.031 14.654 14.673 ot_scf_mini 108 9.5 0.052 0.070 13.825 13.833 grid_collocate_task_list 119 9.7 12.419 12.522 12.419 12.522 dbcsr_multiply_generic 2286 12.5 0.160 0.165 11.816 12.213 rebuild_ks_matrix 119 8.3 0.000 0.001 11.916 11.953 qs_ks_build_kohn_sham_matrix 119 9.3 0.015 0.015 11.916 11.952 qs_ks_update_qs_env 119 7.6 0.001 0.001 10.782 10.813 init_scf_loop 11 6.9 0.003 0.005 8.045 8.047 sum_up_and_integrate 119 10.3 0.002 0.002 7.516 7.758 integrate_v_rspace 119 11.3 0.112 0.114 7.501 7.744 ot_mini 108 10.5 0.010 0.011 7.082 7.105 prepare_preconditioner 11 7.9 0.000 0.000 6.391 6.405 make_preconditioner 11 8.9 0.003 0.005 6.391 6.405 make_full_inverse_cholesky 11 9.9 0.000 0.000 5.955 6.160 multiply_cannon 2286 13.5 0.536 0.545 4.952 5.875 grid_integrate_task_list 119 12.3 5.120 5.363 5.120 5.363 qs_ot_get_p 119 10.4 0.001 0.001 4.946 4.969 qs_ot_get_derivative 108 11.5 0.002 0.002 4.887 4.912 make_m2s 4572 13.5 0.064 0.064 4.322 4.747 fft_wrap_pw1pw2 1201 11.6 0.014 0.015 4.417 4.440 init_scf_run 11 5.9 0.001 0.004 4.287 4.287 scf_env_initial_rho_setup 11 6.9 0.004 0.006 4.286 4.287 density_rs2pw 119 9.7 0.005 0.005 4.078 4.150 fft_wrap_pw1pw2_140 487 12.2 0.469 0.472 3.826 3.848 multiply_cannon_loop 2286 14.5 0.063 0.065 3.594 3.736 cp_fm_cholesky_invert 11 10.9 3.532 3.541 3.532 3.541 make_images 4572 14.5 0.746 0.766 3.037 3.454 fft3d_ps 1201 13.6 1.701 1.714 3.335 3.361 multiply_cannon_multrec 4572 15.5 2.933 3.099 2.963 3.128 qs_ot_p2m_diag 50 11.0 0.074 0.079 2.852 2.864 qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.793 2.796 wfi_extrapolate 11 7.9 0.001 0.001 2.725 2.726 cp_dbcsr_syevd 50 12.0 0.004 0.004 2.561 2.568 mp_waitall_1 74613 16.8 1.987 2.492 1.987 2.492 potential_pw2rs 119 12.3 0.027 0.028 2.269 2.271 hybrid_alltoall_any 4725 16.4 0.163 0.318 1.703 2.219 ot_diis_step 108 11.5 0.036 0.041 2.177 2.178 cp_fm_diag_elpa 50 13.0 0.000 0.000 2.159 2.161 cp_fm_diag_elpa_base 50 14.0 1.998 2.032 2.157 2.160 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.827 2.099 make_images_data 4572 15.5 0.054 0.058 1.411 1.930 transfer_rs2pw 487 10.6 0.008 0.008 1.814 1.892 qs_env_update_s_mstruct 11 6.9 0.001 0.002 1.656 1.708 qs_ot_get_derivative_taylor 59 13.0 0.002 0.002 1.675 1.692 apply_preconditioner_dbcsr 119 12.6 0.002 0.004 1.603 1.685 apply_single 119 13.6 0.000 0.000 1.601 1.683 dbcsr_complete_redistribute 329 12.2 0.601 0.606 1.524 1.624 qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.563 1.582 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.539 1.547 transfer_rs2pw_140 130 11.5 0.916 0.937 1.385 1.477 mp_alltoall_z22v 1201 15.6 1.452 1.469 1.452 1.469 cp_fm_cholesky_decompose 22 10.9 1.412 1.415 1.412 1.415 copy_dbcsr_to_fm 153 11.3 0.003 0.003 1.379 1.398 calculate_first_density_matrix 1 7.0 0.002 0.003 1.337 1.337 arnoldi_extremal 119 11.4 0.002 0.002 1.255 1.276 arnoldi_normal_ev 119 12.4 0.003 0.003 1.253 1.273 mp_allgather_i34 2286 14.5 0.475 1.271 0.475 1.271 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="116", plot="h2o_64_md", label="(4n/1r/36t)", y=61.501000, yerr=0.000000 PlotPoint: name="117", plot="h2o_64_md_mem", label="(4n/1r/36t)", y=3118.545455, yerr=916.983530 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/12/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 64 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 32 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 32 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 422 85181071360 0.0% 100.0% 0.0% flops 32 x 32 x 422 85181071360 0.0% 100.0% 0.0% flops 32 x 64 x 422 85181071360 0.0% 100.0% 0.0% flops 64 x 32 x 422 85181071360 0.0% 100.0% 0.0% flops 64 x 64 x 427 86190325760 0.0% 100.0% 0.0% flops 32 x 32 x 427 86190325760 0.0% 100.0% 0.0% flops 32 x 64 x 427 86190325760 0.0% 100.0% 0.0% flops 64 x 32 x 427 86190325760 0.0% 100.0% 0.0% flops 64 x 64 x 431 86997729280 0.0% 100.0% 0.0% flops 32 x 32 x 431 86997729280 0.0% 100.0% 0.0% flops 32 x 64 x 431 86997729280 0.0% 100.0% 0.0% flops 64 x 32 x 431 86997729280 0.0% 100.0% 0.0% flops 422 x 32 x 32 104651030528 0.0% 100.0% 0.0% flops 422 x 32 x 64 104651030528 0.0% 100.0% 0.0% flops 422 x 64 x 64 104651030528 0.0% 100.0% 0.0% flops 422 x 64 x 32 104651030528 0.0% 100.0% 0.0% flops 427 x 32 x 32 105890971648 0.0% 100.0% 0.0% flops 427 x 32 x 64 105890971648 0.0% 100.0% 0.0% flops 427 x 64 x 64 105890971648 0.0% 100.0% 0.0% flops 427 x 64 x 32 105890971648 0.0% 100.0% 0.0% flops 431 x 64 x 64 106882924544 0.0% 100.0% 0.0% flops 431 x 64 x 32 106882924544 0.0% 100.0% 0.0% flops 431 x 32 x 32 106882924544 0.0% 100.0% 0.0% flops 431 x 32 x 64 106882924544 0.0% 100.0% 0.0% flops 9 x 9 x 64 134590242816 0.0% 100.0% 0.0% flops 9 x 9 x 32 134590242816 0.0% 100.0% 0.0% flops 422 x 32 x 422 160475054080 0.0% 100.0% 0.0% flops 422 x 64 x 422 160475054080 0.0% 100.0% 0.0% flops 427 x 32 x 422 162376417280 0.0% 100.0% 0.0% flops 427 x 64 x 422 162376417280 0.0% 100.0% 0.0% flops 422 x 32 x 427 162376417280 0.0% 100.0% 0.0% flops 422 x 64 x 427 162376417280 0.0% 100.0% 0.0% flops 431 x 64 x 422 163897507840 0.0% 100.0% 0.0% flops 422 x 32 x 431 163897507840 0.0% 100.0% 0.0% flops 422 x 64 x 431 163897507840 0.0% 100.0% 0.0% flops 431 x 32 x 422 163897507840 0.0% 100.0% 0.0% flops 427 x 32 x 427 164300308480 0.0% 100.0% 0.0% flops 427 x 64 x 427 164300308480 0.0% 100.0% 0.0% flops 431 x 64 x 427 165839421440 0.0% 100.0% 0.0% flops 427 x 32 x 431 165839421440 0.0% 100.0% 0.0% flops 427 x 64 x 431 165839421440 0.0% 100.0% 0.0% flops 431 x 32 x 427 165839421440 0.0% 100.0% 0.0% flops 431 x 64 x 431 167392952320 0.0% 100.0% 0.0% flops 431 x 32 x 431 167392952320 0.0% 100.0% 0.0% flops 9 x 22 x 64 174697712640 0.0% 100.0% 0.0% flops 9 x 22 x 32 174697712640 0.0% 100.0% 0.0% flops 22 x 9 x 64 175021203456 0.0% 100.0% 0.0% flops 22 x 9 x 32 175021203456 0.0% 100.0% 0.0% flops 22 x 22 x 64 226790907904 0.0% 100.0% 0.0% flops 22 x 22 x 32 226790907904 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 12.884056E+12 0.0% 100.0% 0.0% flops max/rank 137.578432E+09 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 609915708 0.0% 100.0% 0.0% number of processed stacks 5472840 0.0% 100.0% 0.0% average stack size 0.0 111.4 0.0 marketing flops 15.646547E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 213.536768E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 6510240 MPI messages size (bytes): total size 1.243517E+12 min size 0.000000E+00 max size 1.486088E+06 average size 191.009359E+03 MPI breakdown and total messages size (bytes): size <= 128 50820 0 128 < size <= 8192 1301256 10659889152 8192 < size <= 32768 1625844 31963807744 32768 < size <= 131072 1967328 214884679680 131072 < size <= 4194304 1564992 985951934528 4194304 < size <= 16777216 0 0 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 65 12. MP_Allreduce 11033 25. MP_Alltoall 8043 61599. MP_ISend 98596 100033. MP_IRecv 98596 98808. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3528 76095. MP_Allreduce 9921 489. MP_Sync 492 MP_Alltoall 1939 944079. MP_SendRecv 31460 6552. MP_ISendRecv 31460 6552. MP_Wait 47872 MP_comm_split 48 MP_ISend 26114 37861. MP_IRecv 26114 37861. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.015 0.033 52.396 52.397 qs_mol_dyn_low 1 2.0 0.007 0.017 51.959 51.971 qs_forces 11 3.9 0.003 0.004 51.737 51.760 qs_energies 11 4.9 0.001 0.002 48.956 48.986 scf_env_do_scf 11 5.9 0.000 0.001 44.075 44.077 scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 38.303 38.304 velocity_verlet 10 3.0 0.001 0.002 29.978 29.985 dbcsr_multiply_generic 2055 12.4 0.105 0.148 18.340 19.040 qs_scf_new_mos 99 7.5 0.001 0.001 16.609 17.008 qs_scf_loop_do_ot 99 8.5 0.001 0.001 16.609 17.007 rebuild_ks_matrix 110 8.3 0.000 0.001 15.656 16.069 qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.015 15.655 16.069 ot_scf_mini 99 9.5 0.002 0.003 15.491 15.845 qs_ks_update_qs_env 110 7.6 0.001 0.001 13.810 14.185 multiply_cannon 2055 13.4 0.166 0.197 11.785 14.079 multiply_cannon_loop 2055 14.4 0.173 0.270 10.533 12.377 sum_up_and_integrate 110 10.3 0.001 0.002 10.622 10.649 integrate_v_rspace 110 11.3 0.003 0.004 10.603 10.628 qs_rho_update_rho_low 110 7.6 0.001 0.001 10.496 10.530 calculate_rho_elec 110 8.6 0.018 0.031 10.496 10.530 multiply_cannon_multrec 24660 15.4 5.815 8.844 5.825 8.854 ot_mini 99 10.5 0.001 0.001 8.426 8.811 mp_waitall_1 264348 16.4 6.499 8.782 6.499 8.782 init_scf_loop 11 6.9 0.000 0.000 5.733 5.734 qs_ot_get_derivative 99 11.5 0.001 0.001 5.359 5.715 grid_integrate_task_list 110 12.3 5.130 5.344 5.130 5.344 density_rs2pw 110 9.6 0.005 0.007 4.659 4.883 grid_collocate_task_list 110 9.6 4.621 4.839 4.621 4.839 multiply_cannon_metrocomm3 24660 15.4 0.074 0.154 1.774 4.822 make_m2s 4110 13.4 0.071 0.096 4.464 4.739 fft_wrap_pw1pw2 1111 11.6 0.012 0.014 4.401 4.599 make_images 4110 14.4 0.158 0.200 3.984 4.245 fft3d_ps 1111 13.6 0.687 0.972 3.970 4.130 potential_pw2rs 110 12.3 0.005 0.008 4.084 4.106 qs_ot_get_p 110 10.4 0.001 0.001 3.575 4.055 fft_wrap_pw1pw2_140 451 12.1 0.097 0.107 3.638 3.856 prepare_preconditioner 11 7.9 0.000 0.000 3.708 3.757 make_preconditioner 11 8.9 0.000 0.000 3.708 3.757 init_scf_run 11 5.9 0.000 0.001 3.575 3.575 scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.575 3.575 make_full_inverse_cholesky 11 9.9 0.000 0.000 3.336 3.419 mp_waitany 13684 13.7 3.049 3.395 3.049 3.395 mp_alltoall_d11v 2046 13.8 3.066 3.264 3.066 3.264 apply_preconditioner_dbcsr 110 12.6 0.000 0.000 2.784 3.155 apply_single 110 13.6 0.000 0.001 2.784 3.155 multiply_cannon_metrocomm1 24660 15.4 0.084 0.168 2.036 3.077 ot_diis_step 99 11.5 0.006 0.008 3.012 3.012 wfi_extrapolate 11 7.9 0.001 0.001 3.006 3.006 make_images_data 4110 15.4 0.051 0.081 2.353 2.894 transfer_pw2rs 451 13.1 0.005 0.006 2.872 2.891 transfer_rs2pw 451 10.6 0.005 0.007 2.555 2.728 hybrid_alltoall_any 4261 16.3 0.076 0.468 1.835 2.676 mp_sum_l 10179 13.1 1.514 2.642 1.514 2.642 mp_alltoall_z22v 1111 15.6 2.533 2.619 2.533 2.619 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 1.893 2.085 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.976 2.015 qs_ot_p2m_diag 48 11.0 0.008 0.016 1.932 1.945 mp_allgather_i34 2055 14.4 0.593 1.817 0.593 1.817 transfer_pw2rs_50 110 14.3 0.090 0.105 1.462 1.757 qs_ot_get_derivative_taylor 52 13.0 0.001 0.002 1.578 1.745 cp_dbcsr_syevd 48 12.0 0.003 0.004 1.732 1.733 cp_fm_cholesky_decompose 22 10.9 1.536 1.565 1.536 1.565 rs_gather_matrices 110 12.3 0.096 0.129 1.272 1.525 mp_sum_d 3893 11.9 1.054 1.515 1.054 1.515 yz_to_x 231 14.8 0.064 0.069 1.430 1.484 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 1.450 1.460 transfer_rs2pw_50 110 11.6 0.127 0.142 1.395 1.428 cp_fm_cholesky_invert 11 10.9 1.359 1.365 1.359 1.365 calculate_dm_sparse 110 9.5 0.001 0.001 1.277 1.363 dbcsr_complete_redistribute 325 12.2 0.138 0.166 1.209 1.332 make_images_sizes 4110 15.4 0.005 0.011 0.871 1.323 mp_alltoall_i44 4110 16.4 0.867 1.318 0.867 1.318 x_to_yz 220 13.4 0.086 0.168 1.213 1.266 rs_scatter_matrices 121 9.7 0.084 0.105 1.196 1.229 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.064 1.195 multiply_cannon_metrocomm4 22605 15.4 0.071 0.152 0.493 1.167 cp_fm_diag_elpa 48 13.0 0.000 0.000 1.151 1.152 transfer_pw2rs_140 121 13.9 0.369 0.482 1.000 1.147 cp_fm_redistribute_end 48 14.0 0.564 1.114 0.573 1.119 mp_irecv_dv 59576 16.0 0.474 1.107 0.474 1.107 cp_fm_diag_elpa_base 48 14.0 0.538 1.081 0.544 1.093 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="200", plot="h2o_128_md", label="(4n/36r/1t)", y=52.397000, yerr=0.000000 PlotPoint: name="201", plot="h2o_128_md_mem", label="(4n/36r/1t)", y=202.818182, yerr=2.166614 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/13/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 360 x 32 x 422 8556134400 0.0% 100.0% 0.0% flops 360 x 64 x 422 8556134400 0.0% 100.0% 0.0% flops 378 x 32 x 422 8983941120 0.0% 100.0% 0.0% flops 378 x 64 x 422 8983941120 0.0% 100.0% 0.0% flops 382 x 64 x 422 9079009280 0.0% 100.0% 0.0% flops 382 x 32 x 422 9079009280 0.0% 100.0% 0.0% flops 458 x 32 x 422 10885304320 0.0% 100.0% 0.0% flops 458 x 64 x 422 10885304320 0.0% 100.0% 0.0% flops 471 x 32 x 422 11194275840 0.0% 100.0% 0.0% flops 471 x 64 x 422 11194275840 0.0% 100.0% 0.0% flops 480 x 64 x 422 11408179200 0.0% 100.0% 0.0% flops 480 x 32 x 422 11408179200 0.0% 100.0% 0.0% flops 493 x 32 x 422 11717150720 0.0% 100.0% 0.0% flops 493 x 64 x 422 11717150720 0.0% 100.0% 0.0% flops 32 x 64 x 422 21295267840 0.0% 100.0% 0.0% flops 64 x 64 x 422 21295267840 0.0% 100.0% 0.0% flops 32 x 32 x 422 21295267840 0.0% 100.0% 0.0% flops 64 x 32 x 422 21295267840 0.0% 100.0% 0.0% flops 449 x 32 x 422 21342801920 0.0% 100.0% 0.0% flops 449 x 64 x 422 21342801920 0.0% 100.0% 0.0% flops 360 x 32 x 32 22318940160 0.0% 100.0% 0.0% flops 360 x 32 x 64 22318940160 0.0% 100.0% 0.0% flops 360 x 64 x 32 22318940160 0.0% 100.0% 0.0% flops 360 x 64 x 64 22318940160 0.0% 100.0% 0.0% flops 32 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 32 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 378 x 32 x 32 23434887168 0.0% 100.0% 0.0% flops 378 x 32 x 64 23434887168 0.0% 100.0% 0.0% flops 378 x 64 x 32 23434887168 0.0% 100.0% 0.0% flops 378 x 64 x 64 23434887168 0.0% 100.0% 0.0% flops 382 x 64 x 64 23682875392 0.0% 100.0% 0.0% flops 382 x 64 x 32 23682875392 0.0% 100.0% 0.0% flops 382 x 32 x 32 23682875392 0.0% 100.0% 0.0% flops 382 x 32 x 64 23682875392 0.0% 100.0% 0.0% flops 458 x 32 x 32 28394651648 0.0% 100.0% 0.0% flops 458 x 32 x 64 28394651648 0.0% 100.0% 0.0% flops 458 x 64 x 32 28394651648 0.0% 100.0% 0.0% flops 458 x 64 x 64 28394651648 0.0% 100.0% 0.0% flops 400 x 32 x 422 28520448000 0.0% 100.0% 0.0% flops 400 x 64 x 422 28520448000 0.0% 100.0% 0.0% flops 471 x 32 x 32 29200613376 0.0% 100.0% 0.0% flops 471 x 32 x 64 29200613376 0.0% 100.0% 0.0% flops 471 x 64 x 32 29200613376 0.0% 100.0% 0.0% flops 471 x 64 x 64 29200613376 0.0% 100.0% 0.0% flops 480 x 64 x 64 29758586880 0.0% 100.0% 0.0% flops 480 x 64 x 32 29758586880 0.0% 100.0% 0.0% flops 480 x 32 x 32 29758586880 0.0% 100.0% 0.0% flops 480 x 32 x 64 29758586880 0.0% 100.0% 0.0% flops 493 x 32 x 32 30564548608 0.0% 100.0% 0.0% flops 493 x 32 x 64 30564548608 0.0% 100.0% 0.0% flops 493 x 64 x 32 30564548608 0.0% 100.0% 0.0% flops 493 x 64 x 64 30564548608 0.0% 100.0% 0.0% flops 360 x 32 x 418 33900134400 0.0% 100.0% 0.0% flops 360 x 64 x 418 33900134400 0.0% 100.0% 0.0% flops 378 x 32 x 418 35595141120 0.0% 100.0% 0.0% flops 378 x 64 x 418 35595141120 0.0% 100.0% 0.0% flops 382 x 64 x 418 35971809280 0.0% 100.0% 0.0% flops 382 x 32 x 418 35971809280 0.0% 100.0% 0.0% flops 458 x 32 x 418 43128504320 0.0% 100.0% 0.0% flops 458 x 64 x 418 43128504320 0.0% 100.0% 0.0% flops 471 x 32 x 418 44352675840 0.0% 100.0% 0.0% flops 471 x 64 x 418 44352675840 0.0% 100.0% 0.0% flops 480 x 64 x 418 45200179200 0.0% 100.0% 0.0% flops 480 x 32 x 418 45200179200 0.0% 100.0% 0.0% flops 493 x 32 x 418 46424350720 0.0% 100.0% 0.0% flops 493 x 64 x 418 46424350720 0.0% 100.0% 0.0% flops 360 x 32 x 431 52431667200 0.0% 100.0% 0.0% flops 360 x 64 x 431 52431667200 0.0% 100.0% 0.0% flops 378 x 32 x 431 55053250560 0.0% 100.0% 0.0% flops 378 x 64 x 431 55053250560 0.0% 100.0% 0.0% flops 382 x 64 x 431 55635824640 0.0% 100.0% 0.0% flops 382 x 32 x 431 55635824640 0.0% 100.0% 0.0% flops 449 x 32 x 32 55673356288 0.0% 100.0% 0.0% flops 449 x 32 x 64 55673356288 0.0% 100.0% 0.0% flops 449 x 64 x 32 55673356288 0.0% 100.0% 0.0% flops 449 x 64 x 64 55673356288 0.0% 100.0% 0.0% flops 458 x 32 x 431 66704732160 0.0% 100.0% 0.0% flops 458 x 64 x 431 66704732160 0.0% 100.0% 0.0% flops 471 x 32 x 431 68598097920 0.0% 100.0% 0.0% flops 471 x 64 x 431 68598097920 0.0% 100.0% 0.0% flops 480 x 64 x 431 69908889600 0.0% 100.0% 0.0% flops 480 x 32 x 431 69908889600 0.0% 100.0% 0.0% flops 493 x 32 x 431 71802255360 0.0% 100.0% 0.0% flops 493 x 64 x 431 71802255360 0.0% 100.0% 0.0% flops 400 x 32 x 32 74396467200 0.0% 100.0% 0.0% flops 400 x 32 x 64 74396467200 0.0% 100.0% 0.0% flops 400 x 64 x 32 74396467200 0.0% 100.0% 0.0% flops 400 x 64 x 64 74396467200 0.0% 100.0% 0.0% flops 32 x 64 x 418 84373667840 0.0% 100.0% 0.0% flops 64 x 64 x 418 84373667840 0.0% 100.0% 0.0% flops 32 x 32 x 418 84373667840 0.0% 100.0% 0.0% flops 64 x 32 x 418 84373667840 0.0% 100.0% 0.0% flops 449 x 32 x 418 84562001920 0.0% 100.0% 0.0% flops 449 x 64 x 418 84562001920 0.0% 100.0% 0.0% flops 400 x 32 x 418 113000448000 0.0% 100.0% 0.0% flops 400 x 64 x 418 113000448000 0.0% 100.0% 0.0% flops 32 x 64 x 431 130496593920 0.0% 100.0% 0.0% flops 64 x 64 x 431 130496593920 0.0% 100.0% 0.0% flops 32 x 32 x 431 130496593920 0.0% 100.0% 0.0% flops 64 x 32 x 431 130496593920 0.0% 100.0% 0.0% flops 449 x 32 x 431 130787880960 0.0% 100.0% 0.0% flops 449 x 64 x 431 130787880960 0.0% 100.0% 0.0% flops 9 x 9 x 64 134590242816 0.0% 100.0% 0.0% flops 9 x 9 x 32 134590242816 0.0% 100.0% 0.0% flops 9 x 22 x 64 174697712640 0.0% 100.0% 0.0% flops 9 x 22 x 32 174697712640 0.0% 100.0% 0.0% flops 400 x 32 x 431 174772224000 0.0% 100.0% 0.0% flops 400 x 64 x 431 174772224000 0.0% 100.0% 0.0% flops 22 x 9 x 64 175021203456 0.0% 100.0% 0.0% flops 22 x 9 x 32 175021203456 0.0% 100.0% 0.0% flops 22 x 22 x 64 226790907904 0.0% 100.0% 0.0% flops 22 x 22 x 32 226790907904 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 388562944000 100.0% 0.0% 0.0% flops total 12.930049E+12 3.0% 97.0% 0.0% flops max/rank 265.896255E+09 3.3% 96.7% 0.0% matmuls inhomo. stacks 71280 100.0% 0.0% 0.0% matmuls total 609915708 0.0% 100.0% 0.0% number of processed stacks 5475744 1.3% 98.7% 0.0% average stack size 1.0 112.8 0.0 marketing flops 15.646547E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 258.506752E+06 # max total images/rank 2 # max 3D layers 1 # MPI messages exchanged 3107160 MPI messages size (bytes): total size 1.118966E+12 min size 0.000000E+00 max size 3.034240E+06 average size 360.125031E+03 MPI breakdown and total messages size (bytes): size <= 128 23892 0 128 < size <= 8192 371640 3044474880 8192 < size <= 32768 887412 19031326720 32768 < size <= 131072 539952 51713409024 131072 < size <= 4194304 1284264 1045217450048 4194304 < size <= 16777216 0 0 16777216 < size 0 0 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3528 76186. MP_Allreduce 9920 520. MP_Sync 52 MP_Alltoall 1938 1455359. MP_SendRecv 15620 11120. MP_ISendRecv 15620 11120. MP_Wait 31988 MP_comm_split 48 MP_ISend 14300 93624. MP_IRecv 14300 93624. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.045 0.089 95.975 95.981 qs_mol_dyn_low 1 2.0 0.494 0.978 95.216 95.247 qs_forces 11 3.9 0.008 0.026 92.588 92.601 qs_energies 11 4.9 0.009 0.031 88.085 88.105 scf_env_do_scf 11 5.9 0.001 0.004 78.775 78.776 scf_env_do_scf_inner_loop 99 6.5 0.003 0.033 67.688 67.692 velocity_verlet 10 3.0 0.275 0.637 55.243 55.262 dbcsr_multiply_generic 2055 12.4 0.161 0.209 36.414 38.033 qs_scf_new_mos 99 7.5 0.001 0.001 34.473 35.101 qs_scf_loop_do_ot 99 8.5 0.001 0.001 34.472 35.101 ot_scf_mini 99 9.5 0.003 0.005 32.483 33.021 multiply_cannon 2055 13.4 0.233 0.267 23.958 27.287 multiply_cannon_loop 2055 14.4 0.303 0.404 21.195 24.897 rebuild_ks_matrix 110 8.3 0.001 0.001 24.090 24.705 qs_ks_build_kohn_sham_matrix 110 9.3 0.018 0.031 24.089 24.704 qs_ks_update_qs_env 110 7.6 0.001 0.001 21.303 21.861 multiply_cannon_multrec 24660 15.4 14.024 19.167 14.043 19.184 mp_waitall_1 198528 16.4 9.855 18.703 9.855 18.703 ot_mini 99 10.5 0.001 0.002 17.955 18.539 qs_rho_update_rho_low 110 7.6 0.001 0.001 15.986 16.034 calculate_rho_elec 110 8.6 0.034 0.045 15.985 16.033 sum_up_and_integrate 110 10.3 0.005 0.015 15.329 15.351 integrate_v_rspace 110 11.3 0.004 0.007 15.279 15.307 multiply_cannon_metrocomm3 24660 15.4 0.101 0.193 3.919 12.928 qs_ot_get_derivative 99 11.5 0.001 0.002 10.870 11.415 init_scf_loop 11 6.9 0.000 0.001 10.994 10.998 grid_collocate_task_list 110 9.6 8.842 9.217 8.842 9.217 grid_integrate_task_list 110 12.3 8.502 9.011 8.502 9.011 make_m2s 4110 13.4 0.098 0.126 8.048 8.640 qs_ot_get_p 110 10.4 0.001 0.001 7.475 8.167 make_images 4110 14.4 0.469 0.545 7.103 7.719 prepare_preconditioner 11 7.9 0.000 0.000 7.499 7.569 make_preconditioner 11 8.9 0.000 0.000 7.499 7.568 make_full_inverse_cholesky 11 9.9 0.000 0.000 6.980 7.189 apply_preconditioner_dbcsr 110 12.6 0.000 0.001 6.504 7.128 apply_single 110 13.6 0.001 0.001 6.504 7.128 ot_diis_step 99 11.5 0.013 0.018 6.879 6.879 init_scf_run 11 5.9 0.000 0.008 6.866 6.867 scf_env_initial_rho_setup 11 6.9 0.001 0.008 6.866 6.867 density_rs2pw 110 9.6 0.007 0.011 6.040 6.369 fft_wrap_pw1pw2 1111 11.6 0.019 0.024 6.001 6.290 potential_pw2rs 110 12.3 0.010 0.013 5.381 5.482 wfi_extrapolate 11 7.9 0.001 0.003 5.466 5.466 fft3d_ps 1111 13.6 1.457 1.770 5.135 5.422 mp_sum_l 10179 13.1 3.290 5.420 3.290 5.420 fft_wrap_pw1pw2_140 451 12.1 0.190 0.202 4.546 4.857 make_images_data 4110 15.4 0.065 0.113 3.797 4.613 qs_ot_p2m_diag 48 11.0 0.018 0.034 4.261 4.295 mp_alltoall_d11v 2046 13.8 3.729 4.143 3.729 4.143 hybrid_alltoall_any 4261 16.3 0.140 0.664 3.210 4.027 cp_dbcsr_syevd 48 12.0 0.003 0.004 3.840 3.846 cp_fm_cholesky_invert 11 10.9 3.772 3.795 3.772 3.795 multiply_cannon_metrocomm4 22605 15.4 0.103 0.199 1.443 3.787 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 3.449 3.759 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 3.463 3.715 mp_irecv_dv 61724 16.0 1.355 3.707 1.355 3.707 transfer_pw2rs 451 13.1 0.007 0.009 3.603 3.704 transfer_rs2pw 451 10.6 0.007 0.010 3.154 3.551 multiply_cannon_metrocomm1 24660 15.4 0.059 0.113 1.198 3.236 mp_alltoall_z22v 1111 15.6 2.918 3.117 2.918 3.117 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.044 3.105 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 2.962 2.971 cp_fm_diag_elpa 48 13.0 0.000 0.000 2.814 2.815 cp_fm_redistribute_end 48 14.0 1.394 2.741 1.419 2.752 dbcsr_complete_redistribute 325 12.2 0.306 0.368 2.498 2.730 cp_fm_diag_elpa_base 48 14.0 1.290 2.593 1.312 2.655 mp_waitany 14300 13.8 2.147 2.643 2.147 2.643 mp_allgather_i34 2055 14.4 1.279 2.470 1.279 2.470 calculate_dm_sparse 110 9.5 0.001 0.001 2.306 2.464 mp_sum_d 3893 11.9 1.696 2.460 1.696 2.460 make_images_sizes 4110 15.4 0.006 0.013 1.742 2.400 mp_alltoall_i44 4110 16.4 1.736 2.394 1.736 2.394 cp_fm_cholesky_decompose 22 10.9 2.271 2.310 2.271 2.310 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.015 2.211 qs_ot_get_orbitals 99 10.5 0.001 0.001 1.905 2.009 dbcsr_dot_sd 1091 11.9 0.317 0.354 1.378 1.986 copy_dbcsr_to_fm 151 11.3 0.003 0.004 1.708 1.938 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="202", plot="h2o_128_md", label="(4n/18r/2t)", y=95.981000, yerr=0.000000 PlotPoint: name="203", plot="h2o_128_md_mem", label="(4n/18r/2t)", y=245.363636, yerr=2.346089 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/14/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 436 x 64 x 409 10043207680 0.0% 100.0% 0.0% flops 436 x 32 x 409 10043207680 0.0% 100.0% 0.0% flops 440 x 32 x 409 10135347200 0.0% 100.0% 0.0% flops 440 x 64 x 409 10135347200 0.0% 100.0% 0.0% flops 449 x 32 x 409 10342661120 0.0% 100.0% 0.0% flops 449 x 64 x 409 10342661120 0.0% 100.0% 0.0% flops 458 x 32 x 409 10549975040 0.0% 100.0% 0.0% flops 458 x 64 x 409 10549975040 0.0% 100.0% 0.0% flops 462 x 32 x 409 10642114560 0.0% 100.0% 0.0% flops 462 x 64 x 409 10642114560 0.0% 100.0% 0.0% flops 493 x 64 x 409 11356195840 0.0% 100.0% 0.0% flops 493 x 32 x 409 11356195840 0.0% 100.0% 0.0% flops 32 x 64 x 409 20639252480 0.0% 100.0% 0.0% flops 64 x 64 x 409 20639252480 0.0% 100.0% 0.0% flops 32 x 32 x 409 20639252480 0.0% 100.0% 0.0% flops 64 x 32 x 409 20639252480 0.0% 100.0% 0.0% flops 471 x 32 x 409 21698856960 0.0% 100.0% 0.0% flops 471 x 64 x 409 21698856960 0.0% 100.0% 0.0% flops 32 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 32 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 64 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 32 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 32 23051894784 0.0% 100.0% 0.0% flops 64 x 32 x 64 23051894784 0.0% 100.0% 0.0% flops 436 x 64 x 64 27030716416 0.0% 100.0% 0.0% flops 436 x 64 x 32 27030716416 0.0% 100.0% 0.0% flops 436 x 32 x 32 27030716416 0.0% 100.0% 0.0% flops 436 x 32 x 64 27030716416 0.0% 100.0% 0.0% flops 440 x 32 x 32 27278704640 0.0% 100.0% 0.0% flops 440 x 32 x 64 27278704640 0.0% 100.0% 0.0% flops 440 x 64 x 64 27278704640 0.0% 100.0% 0.0% flops 440 x 64 x 32 27278704640 0.0% 100.0% 0.0% flops 449 x 32 x 32 27836678144 0.0% 100.0% 0.0% flops 449 x 32 x 64 27836678144 0.0% 100.0% 0.0% flops 449 x 64 x 64 27836678144 0.0% 100.0% 0.0% flops 449 x 64 x 32 27836678144 0.0% 100.0% 0.0% flops 458 x 32 x 32 28394651648 0.0% 100.0% 0.0% flops 458 x 32 x 64 28394651648 0.0% 100.0% 0.0% flops 458 x 64 x 64 28394651648 0.0% 100.0% 0.0% flops 458 x 64 x 32 28394651648 0.0% 100.0% 0.0% flops 462 x 32 x 32 28642639872 0.0% 100.0% 0.0% flops 462 x 32 x 64 28642639872 0.0% 100.0% 0.0% flops 462 x 64 x 64 28642639872 0.0% 100.0% 0.0% flops 462 x 64 x 32 28642639872 0.0% 100.0% 0.0% flops 493 x 64 x 64 30564548608 0.0% 100.0% 0.0% flops 493 x 64 x 32 30564548608 0.0% 100.0% 0.0% flops 493 x 32 x 32 30564548608 0.0% 100.0% 0.0% flops 493 x 32 x 64 30564548608 0.0% 100.0% 0.0% flops 436 x 64 x 418 30792622080 0.0% 100.0% 0.0% flops 436 x 32 x 418 30792622080 0.0% 100.0% 0.0% flops 440 x 32 x 418 31075123200 0.0% 100.0% 0.0% flops 440 x 64 x 418 31075123200 0.0% 100.0% 0.0% flops 449 x 32 x 418 31710750720 0.0% 100.0% 0.0% flops 449 x 64 x 418 31710750720 0.0% 100.0% 0.0% flops 458 x 32 x 418 32346378240 0.0% 100.0% 0.0% flops 458 x 64 x 418 32346378240 0.0% 100.0% 0.0% flops 462 x 32 x 418 32628879360 0.0% 100.0% 0.0% flops 462 x 64 x 418 32628879360 0.0% 100.0% 0.0% flops 360 x 64 x 409 33170227200 0.0% 100.0% 0.0% flops 360 x 32 x 409 33170227200 0.0% 100.0% 0.0% flops 493 x 64 x 418 34818263040 0.0% 100.0% 0.0% flops 493 x 32 x 418 34818263040 0.0% 100.0% 0.0% flops 471 x 32 x 32 58401226752 0.0% 100.0% 0.0% flops 471 x 32 x 64 58401226752 0.0% 100.0% 0.0% flops 471 x 64 x 64 58401226752 0.0% 100.0% 0.0% flops 471 x 64 x 32 58401226752 0.0% 100.0% 0.0% flops 32 x 64 x 418 63280250880 0.0% 100.0% 0.0% flops 64 x 64 x 418 63280250880 0.0% 100.0% 0.0% flops 32 x 32 x 418 63280250880 0.0% 100.0% 0.0% flops 64 x 32 x 418 63280250880 0.0% 100.0% 0.0% flops 471 x 32 x 418 66529013760 0.0% 100.0% 0.0% flops 471 x 64 x 418 66529013760 0.0% 100.0% 0.0% flops 436 x 64 x 431 74084003840 0.0% 100.0% 0.0% flops 436 x 32 x 431 74084003840 0.0% 100.0% 0.0% flops 440 x 32 x 431 74763673600 0.0% 100.0% 0.0% flops 440 x 64 x 431 74763673600 0.0% 100.0% 0.0% flops 449 x 32 x 431 76292930560 0.0% 100.0% 0.0% flops 449 x 64 x 431 76292930560 0.0% 100.0% 0.0% flops 458 x 32 x 431 77822187520 0.0% 100.0% 0.0% flops 458 x 64 x 431 77822187520 0.0% 100.0% 0.0% flops 462 x 32 x 431 78501857280 0.0% 100.0% 0.0% flops 462 x 64 x 431 78501857280 0.0% 100.0% 0.0% flops 493 x 64 x 431 83769297920 0.0% 100.0% 0.0% flops 493 x 32 x 431 83769297920 0.0% 100.0% 0.0% flops 360 x 64 x 64 89275760640 0.0% 100.0% 0.0% flops 360 x 64 x 32 89275760640 0.0% 100.0% 0.0% flops 360 x 32 x 32 89275760640 0.0% 100.0% 0.0% flops 360 x 32 x 64 89275760640 0.0% 100.0% 0.0% flops 360 x 64 x 418 101700403200 0.0% 100.0% 0.0% flops 360 x 32 x 418 101700403200 0.0% 100.0% 0.0% flops 9 x 9 x 64 134590242816 0.0% 100.0% 0.0% flops 9 x 9 x 32 134590242816 0.0% 100.0% 0.0% flops 32 x 64 x 431 152246026240 0.0% 100.0% 0.0% flops 64 x 64 x 431 152246026240 0.0% 100.0% 0.0% flops 32 x 32 x 431 152246026240 0.0% 100.0% 0.0% flops 64 x 32 x 431 152246026240 0.0% 100.0% 0.0% flops 471 x 32 x 431 160062228480 0.0% 100.0% 0.0% flops 471 x 64 x 431 160062228480 0.0% 100.0% 0.0% flops 9 x 22 x 64 174697712640 0.0% 100.0% 0.0% flops 9 x 22 x 32 174697712640 0.0% 100.0% 0.0% flops 22 x 9 x 64 175021203456 0.0% 100.0% 0.0% flops 22 x 9 x 32 175021203456 0.0% 100.0% 0.0% flops 22 x 22 x 64 226790907904 0.0% 100.0% 0.0% flops 22 x 22 x 32 226790907904 0.0% 100.0% 0.0% flops 360 x 64 x 431 244681113600 0.0% 100.0% 0.0% flops 360 x 32 x 431 244681113600 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 415276646400 100.0% 0.0% 0.0% flops total 12.956763E+12 3.2% 96.8% 0.0% flops max/rank 393.280399E+09 3.4% 96.6% 0.0% matmuls inhomo. stacks 71280 100.0% 0.0% 0.0% matmuls total 609915708 0.0% 100.0% 0.0% number of processed stacks 5475744 1.3% 98.7% 0.0% average stack size 1.0 112.8 0.0 marketing flops 15.646302E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 303.067136E+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 3528 76294. MP_Allreduce 9919 558. MP_Sync 52 MP_Alltoall 1717 1655329. MP_SendRecv 10340 26400. MP_ISendRecv 10340 26400. MP_Wait 22352 MP_comm_split 48 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 0.021 0.039 75.741 75.750 qs_mol_dyn_low 1 2.0 0.019 0.066 75.245 75.257 qs_forces 11 3.9 0.003 0.003 74.863 74.884 qs_energies 11 4.9 0.001 0.001 71.233 71.267 scf_env_do_scf 11 5.9 0.000 0.001 63.988 63.993 scf_env_do_scf_inner_loop 99 6.5 0.003 0.017 54.681 54.728 velocity_verlet 10 3.0 0.001 0.001 42.232 42.237 dbcsr_multiply_generic 2055 12.4 0.151 0.208 31.243 32.349 qs_scf_new_mos 99 7.5 0.001 0.001 29.892 30.455 qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.891 30.454 ot_scf_mini 99 9.5 0.003 0.004 28.139 28.644 multiply_cannon 2055 13.4 0.229 0.268 21.006 24.582 multiply_cannon_loop 2055 14.4 0.273 0.386 19.168 22.125 multiply_cannon_multrec 24660 15.4 12.535 19.635 12.553 19.655 rebuild_ks_matrix 110 8.3 0.001 0.001 18.457 19.089 qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.016 18.456 19.089 qs_ks_update_qs_env 110 7.6 0.001 0.001 16.318 16.902 mp_waitall_1 176588 16.5 7.266 15.742 7.266 15.742 ot_mini 99 10.5 0.001 0.001 14.594 15.130 multiply_cannon_metrocomm3 24660 15.4 0.097 0.195 4.236 12.277 qs_rho_update_rho_low 110 7.6 0.001 0.001 11.461 11.479 calculate_rho_elec 110 8.6 0.049 0.059 11.460 11.478 sum_up_and_integrate 110 10.3 0.002 0.003 11.104 11.158 integrate_v_rspace 110 11.3 0.003 0.004 11.075 11.142 init_scf_loop 11 6.9 0.000 0.000 9.261 9.339 qs_ot_get_derivative 99 11.5 0.001 0.002 8.310 8.821 qs_ot_get_p 110 10.4 0.001 0.001 7.725 8.498 grid_integrate_task_list 110 12.3 7.083 7.448 7.083 7.448 grid_collocate_task_list 110 9.6 6.927 7.387 6.927 7.387 make_m2s 4110 13.4 0.098 0.135 6.465 6.800 apply_preconditioner_dbcsr 110 12.6 0.000 0.001 5.843 6.330 apply_single 110 13.6 0.000 0.001 5.842 6.330 ot_diis_step 99 11.5 0.013 0.015 6.167 6.168 prepare_preconditioner 11 7.9 0.000 0.000 5.975 6.048 make_preconditioner 11 8.9 0.000 0.000 5.974 6.048 make_full_inverse_cholesky 11 9.9 0.000 0.000 5.540 5.736 make_images 4110 14.4 0.468 0.566 5.468 5.721 init_scf_run 11 5.9 0.000 0.001 5.247 5.248 scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.247 5.248 density_rs2pw 110 9.6 0.005 0.008 3.609 4.424 fft_wrap_pw1pw2 1111 11.6 0.014 0.018 4.091 4.408 mp_sum_l 10179 13.1 2.742 4.367 2.742 4.367 qs_ot_p2m_diag 48 11.0 0.026 0.044 3.734 4.206 wfi_extrapolate 11 7.9 0.001 0.001 4.024 4.024 fft3d_ps 1111 13.6 1.100 1.423 3.421 3.982 multiply_cannon_metrocomm4 22605 15.4 0.098 0.196 1.712 3.895 cp_dbcsr_syevd 48 12.0 0.003 0.004 3.414 3.855 mp_irecv_dv 57340 16.2 1.582 3.754 1.582 3.754 fft_wrap_pw1pw2_140 451 12.1 0.139 0.168 3.499 3.696 make_images_data 4110 15.4 0.064 0.111 2.923 3.241 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 2.800 3.084 mp_alltoall_d11v 2046 13.8 2.489 3.031 2.489 3.031 potential_pw2rs 110 12.3 0.010 0.014 2.872 2.951 hybrid_alltoall_any 4261 16.3 0.116 0.434 2.518 2.925 mp_alltoall_z22v 1111 15.6 2.061 2.861 2.061 2.861 mp_waitany 10164 13.8 1.702 2.704 1.702 2.704 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 2.413 2.672 cp_fm_cholesky_invert 11 10.9 2.651 2.669 2.651 2.669 transfer_rs2pw 451 10.6 0.006 0.008 1.778 2.615 arnoldi_extremal 110 11.4 0.002 0.003 2.088 2.520 arnoldi_normal_ev 110 12.4 0.002 0.004 2.086 2.518 cp_fm_diag_elpa 48 13.0 0.000 0.000 2.434 2.435 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.355 2.427 mp_sum_d 3893 11.9 1.525 2.388 1.525 2.388 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 2.344 2.348 cp_fm_redistribute_end 48 14.0 1.148 2.286 1.155 2.289 cp_fm_diag_elpa_base 48 14.0 1.026 2.124 1.110 2.227 calculate_dm_sparse 110 9.5 0.001 0.001 1.977 2.125 transfer_rs2pw_140 121 11.5 0.225 0.250 1.146 2.000 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.797 1.982 transfer_pw2rs 451 13.1 0.005 0.007 1.735 1.753 dbcsr_dot_sd 1091 11.9 0.277 0.352 1.194 1.728 cp_fm_cholesky_decompose 22 10.9 1.650 1.715 1.650 1.715 qs_ot_get_orbitals 99 10.5 0.001 0.001 1.607 1.696 dbcsr_complete_redistribute 325 12.2 0.274 0.321 1.410 1.689 make_images_sizes 4110 15.4 0.006 0.013 1.168 1.637 mp_alltoall_i44 4110 16.4 1.162 1.632 1.162 1.632 yz_to_x 451 14.2 0.084 0.094 1.188 1.608 mp_allgather_i34 2055 14.4 0.725 1.529 0.725 1.529 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="204", plot="h2o_128_md", label="(4n/12r/3t)", y=75.750000, yerr=0.000000 PlotPoint: name="205", plot="h2o_128_md_mem", label="(4n/12r/3t)", y=285.545455, yerr=3.201239 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/15/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 64 x 64 x 64 1440743424 0.0% 100.0% 0.0% flops 151 x 64 x 64 2340388864 0.0% 100.0% 0.0% flops 173 x 64 x 64 2681372672 0.0% 100.0% 0.0% flops 178 x 64 x 64 2758868992 0.0% 100.0% 0.0% flops 271 x 64 x 64 4200300544 0.0% 100.0% 0.0% flops 64 x 96 x 64 4322230272 0.0% 100.0% 0.0% flops 64 x 64 x 96 4322230272 0.0% 100.0% 0.0% flops 151 x 64 x 96 7021166592 0.0% 100.0% 0.0% flops 151 x 96 x 64 7021166592 0.0% 100.0% 0.0% flops 151 x 64 x 849 7220167680 0.0% 100.0% 0.0% flops 151 x 64 x 853 7254184960 0.0% 100.0% 0.0% flops 151 x 64 x 858 7296706560 0.0% 100.0% 0.0% flops 169 x 64 x 64 7858126848 0.0% 100.0% 0.0% flops 173 x 64 x 96 8044118016 0.0% 100.0% 0.0% flops 173 x 96 x 64 8044118016 0.0% 100.0% 0.0% flops 173 x 64 x 849 8272112640 0.0% 100.0% 0.0% flops 178 x 64 x 96 8276606976 0.0% 100.0% 0.0% flops 178 x 96 x 64 8276606976 0.0% 100.0% 0.0% flops 173 x 64 x 853 8311086080 0.0% 100.0% 0.0% flops 173 x 64 x 858 8359802880 0.0% 100.0% 0.0% flops 178 x 64 x 849 8511191040 0.0% 100.0% 0.0% flops 178 x 64 x 853 8551290880 0.0% 100.0% 0.0% flops 178 x 64 x 858 8601415680 0.0% 100.0% 0.0% flops 32 x 64 x 64 10085203968 0.0% 100.0% 0.0% flops 218 x 64 x 64 10136518656 0.0% 100.0% 0.0% flops 64 x 64 x 849 10710712320 0.0% 100.0% 0.0% flops 64 x 64 x 853 10761175040 0.0% 100.0% 0.0% flops 64 x 64 x 858 10824253440 0.0% 100.0% 0.0% flops 271 x 96 x 64 12600901632 0.0% 100.0% 0.0% flops 271 x 64 x 96 12600901632 0.0% 100.0% 0.0% flops 209 x 64 x 64 12957384704 0.0% 100.0% 0.0% flops 271 x 64 x 849 12958049280 0.0% 100.0% 0.0% flops 64 x 96 x 96 12966690816 0.0% 100.0% 0.0% flops 271 x 64 x 853 13019100160 0.0% 100.0% 0.0% flops 271 x 64 x 858 13095413760 0.0% 100.0% 0.0% flops 151 x 96 x 96 21063499776 0.0% 100.0% 0.0% flops 151 x 96 x 849 21660503040 0.0% 100.0% 0.0% flops 151 x 96 x 853 21762554880 0.0% 100.0% 0.0% flops 151 x 96 x 858 21890119680 0.0% 100.0% 0.0% flops 169 x 96 x 64 23574380544 0.0% 100.0% 0.0% flops 169 x 64 x 96 23574380544 0.0% 100.0% 0.0% flops 173 x 96 x 96 24132354048 0.0% 100.0% 0.0% flops 169 x 64 x 849 24242549760 0.0% 100.0% 0.0% flops 169 x 64 x 853 24356766720 0.0% 100.0% 0.0% flops 169 x 64 x 858 24499537920 0.0% 100.0% 0.0% flops 173 x 96 x 849 24816337920 0.0% 100.0% 0.0% flops 178 x 96 x 96 24829820928 0.0% 100.0% 0.0% flops 173 x 96 x 853 24933258240 0.0% 100.0% 0.0% flops 231 x 64 x 64 25062309888 0.0% 100.0% 0.0% flops 173 x 96 x 858 25079408640 0.0% 100.0% 0.0% flops 178 x 96 x 849 25533573120 0.0% 100.0% 0.0% flops 178 x 96 x 853 25653872640 0.0% 100.0% 0.0% flops 178 x 96 x 858 25804247040 0.0% 100.0% 0.0% flops 32 x 96 x 64 30255611904 0.0% 100.0% 0.0% flops 32 x 64 x 96 30255611904 0.0% 100.0% 0.0% flops 218 x 64 x 96 30409555968 0.0% 100.0% 0.0% flops 218 x 96 x 64 30409555968 0.0% 100.0% 0.0% flops 218 x 64 x 849 31271454720 0.0% 100.0% 0.0% flops 218 x 64 x 853 31418787840 0.0% 100.0% 0.0% flops 218 x 64 x 858 31602954240 0.0% 100.0% 0.0% flops 64 x 96 x 849 32132136960 0.0% 100.0% 0.0% flops 64 x 96 x 853 32283525120 0.0% 100.0% 0.0% flops 64 x 96 x 858 32472760320 0.0% 100.0% 0.0% flops 271 x 96 x 96 37802704896 0.0% 100.0% 0.0% flops 209 x 96 x 64 38872154112 0.0% 100.0% 0.0% flops 209 x 64 x 96 38872154112 0.0% 100.0% 0.0% flops 271 x 96 x 849 38874147840 0.0% 100.0% 0.0% flops 271 x 96 x 853 39057300480 0.0% 100.0% 0.0% flops 271 x 96 x 858 39286241280 0.0% 100.0% 0.0% flops 209 x 64 x 849 39973908480 0.0% 100.0% 0.0% flops 209 x 64 x 853 40162242560 0.0% 100.0% 0.0% flops 209 x 64 x 858 40397660160 0.0% 100.0% 0.0% flops 9 x 9 x 64 67295121408 0.0% 100.0% 0.0% flops 169 x 96 x 96 70723141632 0.0% 100.0% 0.0% flops 169 x 96 x 849 72727649280 0.0% 100.0% 0.0% flops 169 x 96 x 853 73070300160 0.0% 100.0% 0.0% flops 169 x 96 x 858 73498613760 0.0% 100.0% 0.0% flops 32 x 64 x 849 74974986240 0.0% 100.0% 0.0% flops 231 x 64 x 96 75186929664 0.0% 100.0% 0.0% flops 231 x 96 x 64 75186929664 0.0% 100.0% 0.0% flops 32 x 64 x 853 75328225280 0.0% 100.0% 0.0% flops 32 x 64 x 858 75769774080 0.0% 100.0% 0.0% flops 231 x 64 x 849 77317954560 0.0% 100.0% 0.0% flops 231 x 64 x 853 77682232320 0.0% 100.0% 0.0% flops 231 x 64 x 858 78137579520 0.0% 100.0% 0.0% flops 9 x 22 x 64 87348856320 0.0% 100.0% 0.0% flops 22 x 9 x 64 87510601728 0.0% 100.0% 0.0% flops 32 x 96 x 96 90766835712 0.0% 100.0% 0.0% flops 218 x 96 x 96 91228667904 0.0% 100.0% 0.0% flops 218 x 96 x 849 93814364160 0.0% 100.0% 0.0% flops 218 x 96 x 853 94256363520 0.0% 100.0% 0.0% flops 218 x 96 x 858 94808862720 0.0% 100.0% 0.0% flops 22 x 22 x 64 113395453952 0.0% 100.0% 0.0% flops 209 x 96 x 96 116616462336 0.0% 100.0% 0.0% flops 209 x 96 x 849 119921725440 0.0% 100.0% 0.0% flops 209 x 96 x 853 120486727680 0.0% 100.0% 0.0% flops 209 x 96 x 858 121192980480 0.0% 100.0% 0.0% flops 9 x 9 x 96 201885364224 0.0% 100.0% 0.0% flops 32 x 96 x 849 224924958720 0.0% 100.0% 0.0% flops 231 x 96 x 96 225560788992 0.0% 100.0% 0.0% flops 32 x 96 x 853 225984675840 0.0% 100.0% 0.0% flops 32 x 96 x 858 227309322240 0.0% 100.0% 0.0% flops 231 x 96 x 849 231953863680 0.0% 100.0% 0.0% flops 231 x 96 x 853 233046696960 0.0% 100.0% 0.0% flops 231 x 96 x 858 234412738560 0.0% 100.0% 0.0% flops 9 x 22 x 96 262046568960 0.0% 100.0% 0.0% flops 22 x 9 x 96 262531805184 0.0% 100.0% 0.0% flops 22 x 22 x 96 340186361856 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 629542526976 100.0% 0.0% 0.0% flops total 12.909090E+12 4.9% 95.1% 0.0% flops max/rank 451.741913E+09 9.0% 91.0% 0.0% matmuls inhomo. stacks 62964 100.0% 0.0% 0.0% matmuls total 562477038 0.0% 100.0% 0.0% number of processed stacks 2497330 2.5% 97.5% 0.0% average stack size 1.0 231.0 0.0 marketing flops 15.646547E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 356.704256E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 739800 MPI messages size (bytes): total size 565.223162E+09 min size 0.000000E+00 max size 5.889312E+06 average size 764.021562E+03 MPI breakdown and total messages size (bytes): size <= 128 5610 0 128 < size <= 8192 0 0 8192 < size <= 32768 37270 1217658880 32768 < size <= 131072 295520 18156748800 131072 < size <= 4194304 335340 195349708800 4194304 < size <= 16777216 66060 350485115840 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 115 12. MP_Allreduce 11133 25. MP_Alltoall 8043 133840. MP_ISend 49276 334981. MP_IRecv 49276 330552. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3480 77168. MP_Allreduce 9775 564. MP_Sync 52 MP_Alltoall 1717 3826721. MP_SendRecv 7700 27936. MP_ISendRecv 7700 27936. MP_Wait 17864 MP_ISend 8316 219755. MP_IRecv 8316 219755. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.020 0.044 82.029 82.033 qs_mol_dyn_low 1 2.0 0.049 0.096 81.576 81.584 qs_forces 11 3.9 0.003 0.003 81.042 81.068 qs_energies 11 4.9 0.001 0.002 76.665 76.703 scf_env_do_scf 11 5.9 0.001 0.003 68.549 68.563 scf_env_do_scf_inner_loop 99 6.5 0.003 0.020 60.041 60.051 velocity_verlet 10 3.0 0.001 0.001 45.728 45.731 dbcsr_multiply_generic 2055 12.4 0.149 0.192 30.618 31.464 qs_scf_new_mos 99 7.5 0.001 0.001 30.669 31.131 qs_scf_loop_do_ot 99 8.5 0.001 0.001 30.668 31.130 ot_scf_mini 99 9.5 0.003 0.004 28.714 29.095 multiply_cannon 2055 13.4 0.220 0.248 19.808 21.927 rebuild_ks_matrix 110 8.3 0.001 0.001 20.704 21.306 qs_ks_build_kohn_sham_matrix 110 9.3 0.015 0.018 20.704 21.305 multiply_cannon_loop 2055 14.4 0.159 0.219 17.300 19.287 qs_ks_update_qs_env 110 7.6 0.001 0.001 18.324 18.866 ot_mini 99 10.5 0.001 0.001 14.411 14.822 qs_rho_update_rho_low 110 7.6 0.001 0.001 14.645 14.723 calculate_rho_elec 110 8.6 0.065 0.074 14.644 14.723 multiply_cannon_multrec 12330 15.4 12.269 14.410 12.288 14.429 sum_up_and_integrate 110 10.3 0.003 0.005 13.034 13.055 integrate_v_rspace 110 11.3 0.003 0.005 12.992 13.014 mp_waitall_1 141068 16.5 7.202 11.008 7.202 11.008 grid_collocate_task_list 110 9.6 9.472 9.756 9.472 9.756 grid_integrate_task_list 110 12.3 8.634 8.965 8.634 8.965 qs_ot_get_derivative 99 11.5 0.001 0.002 8.569 8.960 qs_ot_get_p 110 10.4 0.001 0.001 8.402 8.868 init_scf_loop 11 6.9 0.000 0.000 8.452 8.462 make_m2s 4110 13.4 0.084 0.108 7.240 7.688 make_images 4110 14.4 0.645 0.841 6.372 6.809 apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.138 6.578 apply_single 110 13.6 0.001 0.001 6.137 6.578 multiply_cannon_metrocomm3 12330 15.4 0.044 0.086 1.947 6.294 ot_diis_step 99 11.5 0.013 0.015 5.794 5.795 init_scf_run 11 5.9 0.000 0.007 5.708 5.708 scf_env_initial_rho_setup 11 6.9 0.000 0.005 5.707 5.708 qs_ot_p2m_diag 48 11.0 0.034 0.048 5.028 5.640 prepare_preconditioner 11 7.9 0.000 0.000 5.485 5.527 make_preconditioner 11 8.9 0.000 0.000 5.485 5.527 cp_dbcsr_syevd 48 12.0 0.003 0.004 4.748 5.361 make_full_inverse_cholesky 11 9.9 0.000 0.000 5.033 5.113 fft_wrap_pw1pw2 1111 11.6 0.016 0.023 4.625 4.685 wfi_extrapolate 11 7.9 0.001 0.001 4.454 4.454 density_rs2pw 110 9.6 0.006 0.009 4.064 4.379 make_images_data 4110 15.4 0.058 0.094 3.555 4.313 fft_wrap_pw1pw2_140 451 12.1 0.198 0.217 3.925 4.035 hybrid_alltoall_any 4261 16.3 0.178 0.964 3.379 4.004 fft3d_ps 1111 13.6 1.494 1.630 3.742 3.805 mp_alltoall_d11v 2046 13.8 3.384 3.651 3.384 3.651 cp_fm_diag_elpa 48 13.0 0.000 0.000 3.381 3.383 cp_fm_diag_elpa_base 48 14.0 3.336 3.341 3.374 3.374 mp_sum_l 10179 13.1 2.228 3.339 2.228 3.339 multiply_cannon_metrocomm1 12330 15.4 0.050 0.093 1.758 3.228 potential_pw2rs 110 12.3 0.014 0.018 3.133 3.162 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 2.661 2.926 mp_allgather_i34 2055 14.4 1.240 2.916 1.240 2.916 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.671 2.733 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 2.537 2.717 cp_fm_cholesky_invert 11 10.9 2.505 2.515 2.505 2.515 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 2.453 2.457 calculate_dm_sparse 110 9.5 0.001 0.001 2.214 2.388 transfer_rs2pw 451 10.6 0.006 0.007 1.899 2.316 mp_waitany 8316 13.8 1.698 2.190 1.698 2.190 mp_alltoall_z22v 1111 15.6 1.902 2.046 1.902 2.046 mp_irecv_dv 30428 16.1 0.930 2.016 0.930 2.016 arnoldi_extremal 110 11.4 0.002 0.002 1.387 2.007 arnoldi_normal_ev 110 12.4 0.002 0.004 1.385 2.004 make_images_sizes 4110 15.4 0.006 0.013 1.316 1.924 mp_alltoall_i44 4110 16.4 1.310 1.918 1.310 1.918 multiply_cannon_metrocomm4 10275 15.4 0.042 0.088 0.907 1.877 transfer_pw2rs 451 13.1 0.005 0.007 1.858 1.871 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.720 1.871 dbcsr_complete_redistribute 325 12.2 0.356 0.421 1.715 1.813 qs_energies_init_hamiltonians 11 5.9 0.007 0.028 1.784 1.809 copy_dbcsr_to_fm 151 11.3 0.003 0.004 1.644 1.786 cp_fm_cholesky_decompose 22 10.9 1.645 1.681 1.645 1.681 mp_sum_d 3893 11.9 1.214 1.641 1.214 1.641 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="206", plot="h2o_128_md", label="(4n/9r/4t)", y=82.033000, yerr=0.000000 PlotPoint: name="207", plot="h2o_128_md_mem", label="(4n/9r/4t)", y=330.363636, yerr=4.095956 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/16/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 129 x 64 x 409 1485749760 0.0% 100.0% 0.0% flops 151 x 64 x 409 1739133440 0.0% 100.0% 0.0% flops 174 x 64 x 409 2004034560 0.0% 100.0% 0.0% flops 200 x 64 x 409 2303488000 0.0% 100.0% 0.0% flops 209 x 64 x 409 2407144960 0.0% 100.0% 0.0% flops 218 x 64 x 409 2510801920 0.0% 100.0% 0.0% flops 160 x 64 x 409 3685580800 0.0% 100.0% 0.0% flops 129 x 64 x 32 3998810112 0.0% 100.0% 0.0% flops 129 x 64 x 64 3998810112 0.0% 100.0% 0.0% flops 187 x 64 x 409 4307522560 0.0% 100.0% 0.0% flops 129 x 96 x 409 4457249280 0.0% 100.0% 0.0% flops 129 x 64 x 418 4555330560 0.0% 100.0% 0.0% flops 151 x 64 x 32 4680777728 0.0% 100.0% 0.0% flops 151 x 64 x 64 4680777728 0.0% 100.0% 0.0% flops 64 x 64 x 409 5159813120 0.0% 100.0% 0.0% flops 151 x 96 x 409 5217400320 0.0% 100.0% 0.0% flops 231 x 64 x 409 5321057280 0.0% 100.0% 0.0% flops 151 x 64 x 418 5332208640 0.0% 100.0% 0.0% flops 174 x 64 x 32 5393743872 0.0% 100.0% 0.0% flops 174 x 64 x 64 5393743872 0.0% 100.0% 0.0% flops 64 x 64 x 32 5762973696 0.0% 100.0% 0.0% flops 64 x 64 x 64 5762973696 0.0% 100.0% 0.0% flops 174 x 96 x 409 6012103680 0.0% 100.0% 0.0% flops 262 x 64 x 409 6035138560 0.0% 100.0% 0.0% flops 174 x 64 x 418 6144399360 0.0% 100.0% 0.0% flops 200 x 64 x 32 6199705600 0.0% 100.0% 0.0% flops 200 x 64 x 64 6199705600 0.0% 100.0% 0.0% flops 209 x 64 x 32 6478692352 0.0% 100.0% 0.0% flops 209 x 64 x 64 6478692352 0.0% 100.0% 0.0% flops 218 x 64 x 32 6757679104 0.0% 100.0% 0.0% flops 218 x 64 x 64 6757679104 0.0% 100.0% 0.0% flops 200 x 96 x 409 6910464000 0.0% 100.0% 0.0% flops 200 x 64 x 418 7062528000 0.0% 100.0% 0.0% flops 209 x 96 x 409 7221434880 0.0% 100.0% 0.0% flops 209 x 64 x 418 7380341760 0.0% 100.0% 0.0% flops 218 x 96 x 409 7532405760 0.0% 100.0% 0.0% flops 218 x 64 x 418 7698155520 0.0% 100.0% 0.0% flops 160 x 64 x 32 9919528960 0.0% 100.0% 0.0% flops 160 x 64 x 64 9919528960 0.0% 100.0% 0.0% flops 129 x 64 x 431 10959674880 0.0% 100.0% 0.0% flops 160 x 96 x 409 11056742400 0.0% 100.0% 0.0% flops 160 x 64 x 418 11300044800 0.0% 100.0% 0.0% flops 187 x 64 x 32 11593449472 0.0% 100.0% 0.0% flops 187 x 64 x 64 11593449472 0.0% 100.0% 0.0% flops 129 x 96 x 64 11996430336 0.0% 100.0% 0.0% flops 129 x 96 x 32 11996430336 0.0% 100.0% 0.0% flops 151 x 64 x 431 12828766720 0.0% 100.0% 0.0% flops 187 x 96 x 409 12922567680 0.0% 100.0% 0.0% flops 187 x 64 x 418 13206927360 0.0% 100.0% 0.0% flops 129 x 96 x 418 13665991680 0.0% 100.0% 0.0% flops 151 x 96 x 64 14042333184 0.0% 100.0% 0.0% flops 151 x 96 x 32 14042333184 0.0% 100.0% 0.0% flops 231 x 64 x 32 14321319936 0.0% 100.0% 0.0% flops 231 x 64 x 64 14321319936 0.0% 100.0% 0.0% flops 174 x 64 x 431 14782817280 0.0% 100.0% 0.0% flops 32 x 64 x 409 15479439360 0.0% 100.0% 0.0% flops 64 x 96 x 409 15479439360 0.0% 100.0% 0.0% flops 64 x 64 x 418 15820062720 0.0% 100.0% 0.0% flops 231 x 96 x 409 15963171840 0.0% 100.0% 0.0% flops 151 x 96 x 418 15996625920 0.0% 100.0% 0.0% flops 174 x 96 x 64 16181231616 0.0% 100.0% 0.0% flops 174 x 96 x 32 16181231616 0.0% 100.0% 0.0% flops 262 x 64 x 32 16243228672 0.0% 100.0% 0.0% flops 262 x 64 x 64 16243228672 0.0% 100.0% 0.0% flops 231 x 64 x 418 16314439680 0.0% 100.0% 0.0% flops 200 x 64 x 431 16991744000 0.0% 100.0% 0.0% flops 32 x 64 x 32 17288921088 0.0% 100.0% 0.0% flops 32 x 64 x 64 17288921088 0.0% 100.0% 0.0% flops 64 x 96 x 64 17288921088 0.0% 100.0% 0.0% flops 64 x 96 x 32 17288921088 0.0% 100.0% 0.0% flops 209 x 64 x 431 17756372480 0.0% 100.0% 0.0% flops 262 x 96 x 409 18105415680 0.0% 100.0% 0.0% flops 174 x 96 x 418 18433198080 0.0% 100.0% 0.0% flops 262 x 64 x 418 18503823360 0.0% 100.0% 0.0% flops 218 x 64 x 431 18521000960 0.0% 100.0% 0.0% flops 200 x 96 x 64 18599116800 0.0% 100.0% 0.0% flops 200 x 96 x 32 18599116800 0.0% 100.0% 0.0% flops 209 x 96 x 64 19436077056 0.0% 100.0% 0.0% flops 209 x 96 x 32 19436077056 0.0% 100.0% 0.0% flops 218 x 96 x 64 20273037312 0.0% 100.0% 0.0% flops 218 x 96 x 32 20273037312 0.0% 100.0% 0.0% flops 200 x 96 x 418 21187584000 0.0% 100.0% 0.0% flops 209 x 96 x 418 22141025280 0.0% 100.0% 0.0% flops 218 x 96 x 418 23094466560 0.0% 100.0% 0.0% flops 160 x 64 x 431 27186790400 0.0% 100.0% 0.0% flops 160 x 96 x 64 29758586880 0.0% 100.0% 0.0% flops 160 x 96 x 32 29758586880 0.0% 100.0% 0.0% flops 187 x 64 x 431 31774561280 0.0% 100.0% 0.0% flops 129 x 96 x 431 32879024640 0.0% 100.0% 0.0% flops 160 x 96 x 418 33900134400 0.0% 100.0% 0.0% flops 187 x 96 x 64 34780348416 0.0% 100.0% 0.0% flops 187 x 96 x 32 34780348416 0.0% 100.0% 0.0% flops 64 x 64 x 431 38061506560 0.0% 100.0% 0.0% flops 151 x 96 x 431 38486300160 0.0% 100.0% 0.0% flops 231 x 64 x 431 39250928640 0.0% 100.0% 0.0% flops 187 x 96 x 418 39620782080 0.0% 100.0% 0.0% flops 231 x 96 x 64 42963959808 0.0% 100.0% 0.0% flops 231 x 96 x 32 42963959808 0.0% 100.0% 0.0% flops 174 x 96 x 431 44348451840 0.0% 100.0% 0.0% flops 262 x 64 x 431 44518369280 0.0% 100.0% 0.0% flops 32 x 96 x 409 46438318080 0.0% 100.0% 0.0% flops 32 x 64 x 418 47460188160 0.0% 100.0% 0.0% flops 64 x 96 x 418 47460188160 0.0% 100.0% 0.0% flops 262 x 96 x 64 48729686016 0.0% 100.0% 0.0% flops 262 x 96 x 32 48729686016 0.0% 100.0% 0.0% flops 231 x 96 x 418 48943319040 0.0% 100.0% 0.0% flops 200 x 96 x 431 50975232000 0.0% 100.0% 0.0% flops 32 x 96 x 64 51866763264 0.0% 100.0% 0.0% flops 32 x 96 x 32 51866763264 0.0% 100.0% 0.0% flops 209 x 96 x 431 53269117440 0.0% 100.0% 0.0% flops 262 x 96 x 418 55511470080 0.0% 100.0% 0.0% flops 218 x 96 x 431 55563002880 0.0% 100.0% 0.0% flops 160 x 96 x 431 81560371200 0.0% 100.0% 0.0% flops 187 x 96 x 431 95323683840 0.0% 100.0% 0.0% flops 32 x 64 x 431 114184519680 0.0% 100.0% 0.0% flops 64 x 96 x 431 114184519680 0.0% 100.0% 0.0% flops 231 x 96 x 431 117752785920 0.0% 100.0% 0.0% flops 262 x 96 x 431 133555107840 0.0% 100.0% 0.0% flops 9 x 9 x 64 134590242816 0.0% 100.0% 0.0% flops 9 x 9 x 32 134590242816 0.0% 100.0% 0.0% flops 32 x 96 x 418 142380564480 0.0% 100.0% 0.0% flops 9 x 22 x 64 174697712640 0.0% 100.0% 0.0% flops 9 x 22 x 32 174697712640 0.0% 100.0% 0.0% flops 22 x 9 x 64 175021203456 0.0% 100.0% 0.0% flops 22 x 9 x 32 175021203456 0.0% 100.0% 0.0% flops 22 x 22 x 64 226790907904 0.0% 100.0% 0.0% flops 22 x 22 x 32 226790907904 0.0% 100.0% 0.0% flops 32 x 96 x 431 342553559040 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 2770741067776 100.0% 0.0% 0.0% flops total 13.483664E+12 20.5% 79.5% 0.0% flops max/rank 678.224219E+09 23.3% 76.7% 0.0% matmuls inhomo. stacks 461340 100.0% 0.0% 0.0% matmuls total 609143868 0.1% 99.9% 0.0% number of processed stacks 4707072 9.8% 90.2% 0.0% average stack size 1.0 143.4 0.0 marketing flops 15.646302E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 452.444160E+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 3480 77279. 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 0.075 0.092 90.996 90.999 qs_mol_dyn_low 1 2.0 0.010 0.028 90.488 90.500 qs_forces 11 3.9 0.003 0.004 89.982 90.021 qs_energies 11 4.9 0.001 0.002 85.493 85.530 scf_env_do_scf 11 5.9 0.001 0.003 76.741 76.753 scf_env_do_scf_inner_loop 99 6.5 0.003 0.026 61.583 61.600 velocity_verlet 10 3.0 0.046 0.181 53.361 53.373 dbcsr_multiply_generic 2055 12.4 0.168 0.213 34.454 34.954 qs_scf_new_mos 99 7.5 0.001 0.001 30.795 31.249 qs_scf_loop_do_ot 99 8.5 0.001 0.001 30.794 31.248 ot_scf_mini 99 9.5 0.005 0.013 28.671 29.115 multiply_cannon 2055 13.4 0.244 0.277 20.629 22.643 rebuild_ks_matrix 110 8.3 0.001 0.001 21.942 22.516 qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.017 21.941 22.515 qs_ks_update_qs_env 110 7.6 0.001 0.001 19.536 20.045 multiply_cannon_loop 2055 14.4 0.287 0.389 18.054 19.530 multiply_cannon_multrec 24660 15.4 13.651 16.341 13.673 16.375 ot_mini 99 10.5 0.003 0.008 15.644 16.117 qs_rho_update_rho_low 110 7.6 0.001 0.001 15.794 15.823 calculate_rho_elec 110 8.6 0.097 0.102 15.794 15.822 init_scf_loop 11 6.9 0.000 0.000 15.104 15.127 sum_up_and_integrate 110 10.3 0.002 0.003 13.339 13.353 integrate_v_rspace 110 11.3 0.003 0.004 13.291 13.308 prepare_preconditioner 11 7.9 0.000 0.000 11.679 11.727 make_preconditioner 11 8.9 0.000 0.000 11.679 11.727 grid_collocate_task_list 110 9.6 10.742 11.264 10.742 11.264 make_full_inverse_cholesky 11 9.9 0.000 0.000 9.724 11.254 make_m2s 4110 13.4 0.105 0.136 9.699 10.106 grid_integrate_task_list 110 12.3 9.147 9.352 9.147 9.352 qs_ot_get_derivative 99 11.5 0.001 0.002 8.598 9.050 make_images 4110 14.4 0.898 1.184 8.312 8.664 mp_waitall_1 121746 16.5 6.306 8.510 6.306 8.510 apply_preconditioner_dbcsr 110 12.6 0.000 0.001 6.754 7.135 apply_single 110 13.6 0.001 0.001 6.754 7.135 ot_diis_step 99 11.5 0.015 0.018 6.971 6.973 qs_ot_get_p 110 10.4 0.001 0.001 6.068 6.589 init_scf_run 11 5.9 0.000 0.005 6.037 6.037 scf_env_initial_rho_setup 11 6.9 0.001 0.006 6.036 6.037 cp_fm_upper_to_full 70 14.2 3.624 5.177 3.624 5.177 make_images_data 4110 15.4 0.068 0.119 4.436 5.003 fft_wrap_pw1pw2 1111 11.6 0.017 0.022 4.684 4.729 density_rs2pw 110 9.6 0.006 0.009 3.969 4.691 hybrid_alltoall_any 4261 16.3 0.151 0.486 3.963 4.688 wfi_extrapolate 11 7.9 0.001 0.001 4.632 4.632 dbcsr_complete_redistribute 325 12.2 0.503 0.607 3.205 4.501 multiply_cannon_metrocomm3 24660 15.4 0.056 0.106 2.050 4.489 fft_wrap_pw1pw2_140 451 12.1 0.242 0.254 4.082 4.132 mp_sum_l 10179 13.1 2.494 3.925 2.494 3.925 fft3d_ps 1111 13.6 1.598 1.713 3.667 3.729 copy_fm_to_dbcsr 174 11.2 0.001 0.002 2.467 3.723 mp_alltoall_d11v 2046 13.8 3.015 3.299 3.015 3.299 qs_ot_p2m_diag 48 11.0 0.050 0.062 3.261 3.296 mp_alltoall_i22 605 13.7 1.979 3.272 1.979 3.272 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 2.922 3.201 transfer_fm_to_dbcsr 11 9.9 0.002 0.005 1.918 3.157 multiply_cannon_metrocomm4 20550 15.4 0.087 0.165 1.574 2.942 potential_pw2rs 110 12.3 0.017 0.021 2.926 2.940 cp_dbcsr_syevd 48 12.0 0.003 0.004 2.913 2.936 cp_fm_cholesky_invert 11 10.9 2.872 2.887 2.872 2.887 mp_irecv_dv 62702 16.1 1.474 2.859 1.474 2.859 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.730 2.815 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 2.592 2.802 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 2.720 2.727 mp_waitany 13376 13.8 1.794 2.603 1.794 2.603 transfer_rs2pw 451 10.6 0.006 0.007 1.822 2.574 calculate_dm_sparse 110 9.5 0.001 0.001 2.392 2.491 cp_fm_cholesky_decompose 22 10.9 2.321 2.344 2.321 2.344 cp_fm_diag_elpa 48 13.0 0.000 0.000 2.264 2.266 mp_allgather_i34 2055 14.4 1.025 2.259 1.025 2.259 cp_fm_diag_elpa_base 48 14.0 2.152 2.185 2.258 2.259 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.033 2.193 qs_energies_init_hamiltonians 11 5.9 0.001 0.002 2.116 2.155 transfer_rs2pw_140 121 11.5 0.205 0.239 1.293 2.064 make_images_sizes 4110 15.4 0.007 0.013 1.255 1.919 mp_alltoall_i44 4110 16.4 1.249 1.913 1.249 1.913 mp_alltoall_z22v 1111 15.6 1.762 1.870 1.762 1.870 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="208", plot="h2o_128_md", label="(4n/6r/6t)", y=90.999000, yerr=0.000000 PlotPoint: name="209", plot="h2o_128_md_mem", label="(4n/6r/6t)", y=426.636364, yerr=9.650932 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/17/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 89 x 128 x 128 22070951936 0.0% 100.0% 0.0% flops 89 x 128 x 1280 51327795200 0.0% 100.0% 0.0% flops 80 x 128 x 128 59517173760 0.0% 100.0% 0.0% flops 151 x 128 x 128 74892443648 0.0% 100.0% 0.0% flops 182 x 128 x 128 90267713536 0.0% 100.0% 0.0% flops 32 x 128 x 128 92207579136 0.0% 100.0% 0.0% flops 64 x 128 x 128 92207579136 0.0% 100.0% 0.0% flops 80 x 128 x 1280 138412032000 0.0% 100.0% 0.0% flops 151 x 128 x 1280 174168473600 0.0% 100.0% 0.0% flops 160 x 128 x 128 198390579200 0.0% 100.0% 0.0% flops 182 x 128 x 1280 209924915200 0.0% 100.0% 0.0% flops 129 x 128 x 128 255923847168 0.0% 100.0% 0.0% flops 9 x 9 x 128 269180485632 0.0% 100.0% 0.0% flops 9 x 22 x 128 349395425280 0.0% 100.0% 0.0% flops 22 x 9 x 128 350042406912 0.0% 100.0% 0.0% flops 22 x 22 x 128 453581815808 0.0% 100.0% 0.0% flops 160 x 128 x 1280 461373440000 0.0% 100.0% 0.0% flops 32 x 128 x 1280 516738252800 0.0% 100.0% 0.0% flops 64 x 128 x 1280 516738252800 0.0% 100.0% 0.0% flops 129 x 128 x 1280 595171737600 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 2199488299008 100.0% 0.0% 0.0% flops total 13.192496E+12 16.7% 83.3% 0.0% flops max/rank 860.633972E+09 17.4% 82.6% 0.0% matmuls inhomo. stacks 139920 100.0% 0.0% 0.0% matmuls total 546715604 0.0% 100.0% 0.0% number of processed stacks 1575496 8.9% 91.1% 0.0% average stack size 1.0 380.7 0.0 marketing flops 15.646297E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 603.979776E+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 27 12. MP_Allreduce 10957 25. MP_Alltoall 8043 258767. MP_ISend 32836 652428. MP_IRecv 32836 652812. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3480 77277. 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 0.064 0.102 83.296 83.302 qs_mol_dyn_low 1 2.0 0.035 0.093 82.488 82.496 qs_forces 11 3.9 0.005 0.008 81.829 81.882 qs_energies 11 4.9 0.020 0.041 77.434 77.491 scf_env_do_scf 11 5.9 0.014 0.036 66.344 66.355 scf_env_do_scf_inner_loop 99 6.5 0.023 0.045 55.019 55.029 velocity_verlet 10 3.0 0.173 0.427 47.145 47.161 dbcsr_multiply_generic 2055 12.4 0.159 0.173 28.615 28.780 qs_scf_new_mos 99 7.5 0.001 0.003 25.361 25.496 qs_scf_loop_do_ot 99 8.5 0.001 0.002 25.360 25.495 ot_scf_mini 99 9.5 0.004 0.005 23.490 23.552 rebuild_ks_matrix 110 8.3 0.001 0.001 20.413 20.590 qs_ks_build_kohn_sham_matrix 110 9.3 0.020 0.043 20.413 20.590 multiply_cannon 2055 13.4 0.220 0.231 18.824 20.299 qs_ks_update_qs_env 110 7.6 0.001 0.001 18.214 18.410 multiply_cannon_loop 2055 14.4 0.150 0.189 17.023 17.242 qs_rho_update_rho_low 110 7.6 0.005 0.009 15.363 15.393 calculate_rho_elec 110 8.6 0.149 0.153 15.358 15.392 multiply_cannon_multrec 8220 15.4 13.735 14.576 13.755 14.596 ot_mini 99 10.5 0.002 0.004 12.734 12.810 sum_up_and_integrate 110 10.3 0.002 0.002 12.516 12.550 integrate_v_rspace 110 11.3 0.003 0.003 12.472 12.506 init_scf_loop 11 6.9 0.019 0.040 11.254 11.276 grid_collocate_task_list 110 9.6 9.909 10.463 9.909 10.463 grid_integrate_task_list 110 12.3 8.377 8.577 8.377 8.577 prepare_preconditioner 11 7.9 0.001 0.002 8.130 8.161 make_preconditioner 11 8.9 0.008 0.031 8.129 8.161 init_scf_run 11 5.9 0.001 0.004 7.964 7.965 scf_env_initial_rho_setup 11 6.9 0.006 0.013 7.963 7.965 make_full_inverse_cholesky 11 9.9 0.000 0.000 7.321 7.449 make_m2s 4110 13.4 0.077 0.086 7.030 7.355 qs_ot_get_derivative 99 11.5 0.001 0.002 6.910 6.982 mp_waitall_1 103326 16.6 5.051 6.038 5.051 6.038 make_images 4110 14.4 0.954 1.030 5.682 6.033 ot_diis_step 99 11.5 0.022 0.027 5.785 5.786 apply_preconditioner_dbcsr 110 12.6 0.000 0.001 5.571 5.661 apply_single 110 13.6 0.000 0.001 5.571 5.661 fft_wrap_pw1pw2 1111 11.6 0.015 0.021 5.504 5.657 qs_ot_get_p 110 10.4 0.001 0.002 5.459 5.552 density_rs2pw 110 9.6 0.005 0.005 4.388 5.407 fft_wrap_pw1pw2_140 451 12.1 0.304 0.318 4.762 5.035 wfi_extrapolate 11 7.9 0.001 0.001 4.496 4.496 fft3d_ps 1111 13.6 1.832 2.048 4.343 4.453 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.635 3.838 hybrid_alltoall_any 4261 16.3 0.331 1.322 3.198 3.592 make_images_data 4110 15.4 0.053 0.060 3.141 3.565 mp_alltoall_d11v 2046 13.8 2.891 3.310 2.891 3.310 cp_fm_cholesky_invert 11 10.9 3.288 3.299 3.288 3.299 calculate_first_density_matrix 1 7.0 0.203 0.412 3.207 3.211 qs_ot_p2m_diag 48 11.0 0.072 0.081 3.177 3.184 dbcsr_complete_redistribute 325 12.2 0.899 1.014 2.754 3.142 cp_dbcsr_syevd 48 12.0 0.003 0.004 2.881 2.883 potential_pw2rs 110 12.3 0.021 0.022 2.857 2.866 transfer_rs2pw 451 10.6 0.006 0.007 1.741 2.725 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.649 2.722 cp_fm_cholesky_decompose 22 10.9 2.645 2.665 2.645 2.665 mp_waitany 9240 13.8 1.761 2.636 1.761 2.636 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.617 2.632 qs_energies_init_hamiltonians 11 5.9 0.001 0.003 2.451 2.504 make_basis_sm 11 9.8 0.000 0.000 2.449 2.452 mp_alltoall_z22v 1111 15.6 2.119 2.333 2.119 2.333 transfer_rs2pw_140 121 11.5 0.193 0.231 1.320 2.290 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 2.208 2.259 cp_fm_diag_elpa 48 13.0 0.000 0.000 2.251 2.251 cp_fm_diag_elpa_base 48 14.0 2.161 2.195 2.248 2.248 calculate_dm_sparse 110 9.5 0.001 0.001 2.150 2.240 copy_dbcsr_to_fm 151 11.3 0.003 0.003 2.041 2.143 mp_allgather_i34 2055 14.4 0.800 2.101 0.800 2.101 copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.684 2.020 multiply_cannon_metrocomm1 8220 15.4 0.028 0.030 1.461 2.014 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 1.916 1.961 mp_sum_l 10179 13.1 1.318 1.685 1.318 1.685 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="210", plot="h2o_128_md", label="(4n/4r/9t)", y=83.302000, yerr=0.000000 PlotPoint: name="211", plot="h2o_128_md_mem", label="(4n/4r/9t)", y=562.727273, yerr=20.675374 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/18/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 71 x 64 x 64 1100447744 0.0% 100.0% 0.0% flops 64 x 64 x 64 1440743424 0.0% 100.0% 0.0% flops 71 x 64 x 840 1679462400 0.0% 100.0% 0.0% flops 71 x 64 x 849 1697456640 0.0% 100.0% 0.0% flops 129 x 64 x 64 1999405056 0.0% 100.0% 0.0% flops 129 x 64 x 840 3051417600 0.0% 100.0% 0.0% flops 129 x 64 x 849 3084111360 0.0% 100.0% 0.0% flops 71 x 64 x 96 3301343232 0.0% 100.0% 0.0% flops 71 x 96 x 64 3301343232 0.0% 100.0% 0.0% flops 71 x 64 x 858 3430901760 0.0% 100.0% 0.0% flops 64 x 96 x 64 4322230272 0.0% 100.0% 0.0% flops 64 x 64 x 96 4322230272 0.0% 100.0% 0.0% flops 71 x 96 x 840 5038387200 0.0% 100.0% 0.0% flops 71 x 96 x 849 5092369920 0.0% 100.0% 0.0% flops 64 x 64 x 840 5298585600 0.0% 100.0% 0.0% flops 64 x 64 x 849 5355356160 0.0% 100.0% 0.0% flops 129 x 96 x 64 5998215168 0.0% 100.0% 0.0% flops 129 x 64 x 96 5998215168 0.0% 100.0% 0.0% flops 129 x 64 x 858 6233610240 0.0% 100.0% 0.0% flops 209 x 64 x 64 6478692352 0.0% 100.0% 0.0% flops 222 x 64 x 64 6881673216 0.0% 100.0% 0.0% flops 129 x 96 x 840 9154252800 0.0% 100.0% 0.0% flops 129 x 96 x 849 9252334080 0.0% 100.0% 0.0% flops 209 x 64 x 840 9887539200 0.0% 100.0% 0.0% flops 71 x 96 x 96 9904029696 0.0% 100.0% 0.0% flops 209 x 64 x 849 9993477120 0.0% 100.0% 0.0% flops 32 x 64 x 64 10085203968 0.0% 100.0% 0.0% flops 71 x 96 x 858 10292705280 0.0% 100.0% 0.0% flops 222 x 64 x 840 10502553600 0.0% 100.0% 0.0% flops 222 x 64 x 849 10615080960 0.0% 100.0% 0.0% flops 231 x 64 x 64 10740989952 0.0% 100.0% 0.0% flops 64 x 64 x 858 10824253440 0.0% 100.0% 0.0% flops 240 x 64 x 64 11159470080 0.0% 100.0% 0.0% flops 64 x 96 x 96 12966690816 0.0% 100.0% 0.0% flops 64 x 96 x 840 15895756800 0.0% 100.0% 0.0% flops 64 x 96 x 849 16066068480 0.0% 100.0% 0.0% flops 231 x 64 x 840 16392499200 0.0% 100.0% 0.0% flops 231 x 64 x 849 16568133120 0.0% 100.0% 0.0% flops 240 x 64 x 840 17031168000 0.0% 100.0% 0.0% flops 240 x 64 x 849 17213644800 0.0% 100.0% 0.0% flops 129 x 96 x 96 17994645504 0.0% 100.0% 0.0% flops 129 x 96 x 858 18700830720 0.0% 100.0% 0.0% flops 209 x 96 x 64 19436077056 0.0% 100.0% 0.0% flops 209 x 64 x 96 19436077056 0.0% 100.0% 0.0% flops 209 x 64 x 858 20198830080 0.0% 100.0% 0.0% flops 222 x 64 x 96 20645019648 0.0% 100.0% 0.0% flops 222 x 96 x 64 20645019648 0.0% 100.0% 0.0% flops 222 x 64 x 858 21455216640 0.0% 100.0% 0.0% flops 209 x 96 x 840 29662617600 0.0% 100.0% 0.0% flops 209 x 96 x 849 29980431360 0.0% 100.0% 0.0% flops 32 x 96 x 64 30255611904 0.0% 100.0% 0.0% flops 32 x 64 x 96 30255611904 0.0% 100.0% 0.0% flops 222 x 96 x 840 31507660800 0.0% 100.0% 0.0% flops 222 x 96 x 849 31845242880 0.0% 100.0% 0.0% flops 231 x 64 x 96 32222969856 0.0% 100.0% 0.0% flops 231 x 96 x 64 32222969856 0.0% 100.0% 0.0% flops 64 x 96 x 858 32472760320 0.0% 100.0% 0.0% flops 240 x 96 x 64 33478410240 0.0% 100.0% 0.0% flops 240 x 64 x 96 33478410240 0.0% 100.0% 0.0% flops 231 x 64 x 858 33487534080 0.0% 100.0% 0.0% flops 240 x 64 x 858 34792243200 0.0% 100.0% 0.0% flops 32 x 64 x 840 37090099200 0.0% 100.0% 0.0% flops 32 x 64 x 849 37487493120 0.0% 100.0% 0.0% flops 231 x 96 x 840 49177497600 0.0% 100.0% 0.0% flops 231 x 96 x 849 49704399360 0.0% 100.0% 0.0% flops 240 x 96 x 840 51093504000 0.0% 100.0% 0.0% flops 240 x 96 x 849 51640934400 0.0% 100.0% 0.0% flops 209 x 96 x 96 58308231168 0.0% 100.0% 0.0% flops 209 x 96 x 858 60596490240 0.0% 100.0% 0.0% flops 222 x 96 x 96 61935058944 0.0% 100.0% 0.0% flops 222 x 96 x 858 64365649920 0.0% 100.0% 0.0% flops 9 x 9 x 64 67295121408 0.0% 100.0% 0.0% flops 32 x 64 x 858 75769774080 0.0% 100.0% 0.0% flops 9 x 22 x 64 87348856320 0.0% 100.0% 0.0% flops 22 x 9 x 64 87510601728 0.0% 100.0% 0.0% flops 32 x 96 x 96 90766835712 0.0% 100.0% 0.0% flops 231 x 96 x 96 96668909568 0.0% 100.0% 0.0% flops 240 x 96 x 96 100435230720 0.0% 100.0% 0.0% flops 231 x 96 x 858 100462602240 0.0% 100.0% 0.0% flops 240 x 96 x 858 104376729600 0.0% 100.0% 0.0% flops 32 x 96 x 840 111270297600 0.0% 100.0% 0.0% flops 32 x 96 x 849 112462479360 0.0% 100.0% 0.0% flops 22 x 22 x 64 113395453952 0.0% 100.0% 0.0% flops 9 x 9 x 96 201885364224 0.0% 100.0% 0.0% flops 32 x 96 x 858 227309322240 0.0% 100.0% 0.0% flops 9 x 22 x 96 262046568960 0.0% 100.0% 0.0% flops 22 x 9 x 96 262531805184 0.0% 100.0% 0.0% flops 22 x 22 x 96 340186361856 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 3766099836928 100.0% 0.0% 0.0% flops total 13.644522E+12 27.6% 72.4% 0.0% flops max/rank 1.282868E+12 28.0% 72.0% 0.0% matmuls inhomo. stacks 336996 100.0% 0.0% 0.0% matmuls total 562477038 0.1% 99.9% 0.0% number of processed stacks 2521520 13.4% 86.6% 0.0% average stack size 1.0 257.3 0.0 marketing flops 15.646302E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 786.903040E+06 # max total images/rank 3 # max 3D layers 1 # MPI messages exchanged 197280 MPI messages size (bytes): total size 482.240758E+09 min size 0.000000E+00 max size 17.688240E+06 average size 2.444448E+06 MPI breakdown and total messages size (bytes): size <= 128 1386 0 128 < size <= 8192 0 0 8192 < size <= 32768 4706 153485312 32768 < size <= 131072 50860 4081582080 131072 < size <= 4194304 118308 127519948800 4194304 < size <= 16777216 15420 235141755840 16777216 < size 6600 115343360000 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3480 77746. MP_Allreduce 9774 769. MP_Sync 52 MP_Alltoall 1496 10935805. MP_SendRecv 2420 70608. MP_ISendRecv 2420 70608. MP_Wait 11198 MP_ISend 7392 401928. MP_IRecv 7392 401928. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.038 0.063 126.382 126.389 qs_mol_dyn_low 1 2.0 0.018 0.021 125.517 125.523 qs_forces 11 3.9 0.009 0.013 124.971 124.978 qs_energies 11 4.9 0.008 0.017 119.588 119.594 scf_env_do_scf 11 5.9 0.004 0.010 106.967 106.983 scf_env_do_scf_inner_loop 99 6.5 0.025 0.065 83.306 83.334 velocity_verlet 10 3.0 0.004 0.007 75.135 75.143 dbcsr_multiply_generic 2055 12.4 0.173 0.181 50.603 51.698 qs_scf_new_mos 99 7.5 0.009 0.017 46.959 47.734 qs_scf_loop_do_ot 99 8.5 0.002 0.003 46.950 47.717 ot_scf_mini 99 9.5 0.011 0.015 44.632 45.477 multiply_cannon 2055 13.4 0.245 0.259 32.806 34.625 multiply_cannon_loop 2055 14.4 0.205 0.221 30.300 32.372 rebuild_ks_matrix 110 8.3 0.001 0.001 26.016 26.927 qs_ks_build_kohn_sham_matrix 110 9.3 0.016 0.018 26.016 26.927 mp_waitall_1 102446 16.6 18.048 25.959 18.048 25.959 qs_ks_update_qs_env 110 7.6 0.001 0.001 23.306 24.128 ot_mini 99 10.5 0.020 0.026 22.955 23.764 init_scf_loop 11 6.9 0.001 0.003 23.568 23.600 multiply_cannon_multrec 12330 15.4 15.293 21.207 15.321 21.234 multiply_cannon_metrocomm3 12330 15.4 0.043 0.044 12.404 20.271 prepare_preconditioner 11 7.9 0.000 0.001 19.281 19.332 make_preconditioner 11 8.9 0.001 0.002 19.281 19.332 make_full_inverse_cholesky 11 9.9 0.000 0.000 15.973 18.543 qs_rho_update_rho_low 110 7.6 0.001 0.001 17.619 17.671 calculate_rho_elec 110 8.6 0.190 0.199 17.619 17.670 sum_up_and_integrate 110 10.3 0.002 0.005 13.630 13.665 integrate_v_rspace 110 11.3 0.003 0.004 13.581 13.615 qs_ot_get_p 110 10.4 0.001 0.001 11.123 12.347 grid_collocate_task_list 110 9.6 11.953 12.134 11.953 12.134 qs_ot_get_derivative 99 11.5 0.008 0.010 11.272 12.120 make_m2s 4110 13.4 0.089 0.092 11.546 12.037 apply_preconditioner_dbcsr 110 12.6 0.021 0.041 11.256 11.983 apply_single 110 13.6 0.000 0.001 11.235 11.983 ot_diis_step 99 11.5 0.068 0.086 11.580 11.599 make_images 4110 14.4 1.227 1.431 9.795 10.284 cp_fm_upper_to_full 70 14.2 6.944 9.527 6.944 9.527 grid_integrate_task_list 110 12.3 9.185 9.317 9.185 9.317 init_scf_run 11 5.9 0.001 0.005 8.492 8.493 scf_env_initial_rho_setup 11 6.9 0.002 0.003 8.491 8.493 multiply_cannon_metrocomm4 10275 15.4 0.044 0.048 2.231 7.795 mp_irecv_dv 29063 16.1 2.152 7.722 2.152 7.722 dbcsr_complete_redistribute 325 12.2 0.920 1.202 5.405 7.180 make_images_data 4110 15.4 0.057 0.059 6.139 7.131 hybrid_alltoall_any 4261 16.3 0.202 0.449 5.907 7.053 wfi_extrapolate 11 7.9 0.001 0.001 6.716 6.717 qs_ot_p2m_diag 48 11.0 0.096 0.115 6.658 6.703 cp_dbcsr_syevd 48 12.0 0.003 0.004 6.275 6.316 copy_fm_to_dbcsr 174 11.2 0.001 0.002 4.108 5.886 fft_wrap_pw1pw2 1111 11.6 0.016 0.017 5.619 5.713 mp_sum_l 10179 13.1 4.090 5.680 4.090 5.680 cp_fm_diag_elpa 48 13.0 0.000 0.000 5.422 5.424 cp_fm_diag_elpa_base 48 14.0 5.129 5.160 5.415 5.415 cp_fm_cholesky_invert 11 10.9 5.313 5.327 5.313 5.327 mp_alltoall_i22 605 13.7 3.231 5.105 3.231 5.105 transfer_fm_to_dbcsr 11 9.9 0.002 0.006 3.290 5.063 fft_wrap_pw1pw2_140 451 12.1 0.358 0.367 4.834 4.933 density_rs2pw 110 9.6 0.005 0.006 4.360 4.559 fft3d_ps 1111 13.6 1.906 2.041 4.413 4.489 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 4.457 4.474 qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.756 4.307 mp_alltoall_d11v 2046 13.8 3.769 3.894 3.769 3.894 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 3.261 3.784 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 3.300 3.617 qs_energies_init_hamiltonians 11 5.9 0.003 0.010 3.328 3.333 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.229 3.329 copy_dbcsr_to_fm 151 11.3 0.003 0.003 2.800 3.232 potential_pw2rs 110 12.3 0.021 0.022 3.149 3.170 dbcsr_dot_sd 1091 11.9 0.816 0.886 2.015 3.117 calculate_dm_sparse 110 9.5 0.001 0.001 2.634 2.825 mp_sum_d 3891 11.9 1.569 2.718 1.569 2.718 cp_fm_cholesky_decompose 22 10.9 2.540 2.585 2.540 2.585 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="212", plot="h2o_128_md", label="(4n/3r/12t)", y=126.389000, yerr=0.000000 PlotPoint: name="213", plot="h2o_128_md_mem", label="(4n/3r/12t)", y=723.000000, yerr=28.816031 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/19/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 142 x 128 x 128 70428655616 0.0% 100.0% 0.0% flops 129 x 128 x 128 127961923584 0.0% 100.0% 0.0% flops 142 x 128 x 1280 163787571200 0.0% 100.0% 0.0% flops 32 x 128 x 128 184415158272 0.0% 100.0% 0.0% flops 160 x 128 x 128 238068695040 0.0% 100.0% 0.0% flops 138 x 128 x 128 239556624384 0.0% 100.0% 0.0% flops 9 x 9 x 128 269180485632 0.0% 100.0% 0.0% flops 129 x 128 x 1280 297585868800 0.0% 100.0% 0.0% flops 9 x 22 x 128 349395425280 0.0% 100.0% 0.0% flops 22 x 9 x 128 350042406912 0.0% 100.0% 0.0% flops 22 x 22 x 128 453581815808 0.0% 100.0% 0.0% flops 160 x 128 x 1280 553648128000 0.0% 100.0% 0.0% flops 138 x 128 x 1280 557108428800 0.0% 100.0% 0.0% flops 32 x 128 x 1280 1033476505600 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 2588749070336 100.0% 0.0% 0.0% flops total 13.498461E+12 19.2% 80.8% 0.0% flops max/rank 1.745688E+12 20.7% 79.3% 0.0% matmuls inhomo. stacks 158576 100.0% 0.0% 0.0% matmuls total 546784212 0.0% 100.0% 0.0% number of processed stacks 1648032 9.6% 90.4% 0.0% average stack size 1.0 367.0 0.0 marketing flops 15.646297E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 1.964241E+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 3469 77990. 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 0.107 0.134 120.317 120.318 qs_mol_dyn_low 1 2.0 0.098 0.116 119.475 119.478 qs_forces 11 3.9 0.012 0.017 118.582 118.586 qs_energies 11 4.9 0.012 0.025 113.275 113.290 scf_env_do_scf 11 5.9 0.001 0.003 100.996 100.997 velocity_verlet 10 3.0 0.002 0.002 75.990 76.040 scf_env_do_scf_inner_loop 99 6.5 0.008 0.039 68.526 68.530 dbcsr_multiply_generic 2055 12.4 0.185 0.232 33.839 34.125 init_scf_loop 11 6.9 0.082 0.117 32.383 32.385 qs_scf_new_mos 99 7.5 0.001 0.003 30.781 31.182 qs_scf_loop_do_ot 99 8.5 0.001 0.002 30.780 31.181 ot_scf_mini 99 9.5 0.004 0.005 28.554 28.824 prepare_preconditioner 11 7.9 0.000 0.001 28.632 28.701 make_preconditioner 11 8.9 0.001 0.002 28.632 28.701 make_full_inverse_cholesky 11 9.9 0.000 0.000 22.849 27.912 rebuild_ks_matrix 110 8.3 0.001 0.001 23.208 23.448 qs_ks_build_kohn_sham_matrix 110 9.3 0.016 0.018 23.207 23.447 multiply_cannon 2055 13.4 0.255 0.287 21.285 23.066 qs_rho_update_rho_low 110 7.6 0.001 0.002 21.616 21.654 calculate_rho_elec 110 8.6 0.279 0.280 21.615 21.653 qs_ks_update_qs_env 110 7.6 0.001 0.001 20.881 21.072 multiply_cannon_loop 2055 14.4 0.138 0.174 19.436 19.920 cp_fm_upper_to_full 70 14.2 12.322 17.484 12.322 17.484 grid_collocate_task_list 110 9.6 15.850 16.062 15.850 16.062 ot_mini 99 10.5 0.002 0.002 15.235 15.503 multiply_cannon_multrec 8220 15.4 12.557 13.865 12.585 13.893 sum_up_and_integrate 110 10.3 0.002 0.002 13.408 13.417 integrate_v_rspace 110 11.3 0.003 0.003 13.357 13.367 mp_waitall_1 84994 16.7 9.437 12.628 9.437 12.628 dbcsr_complete_redistribute 325 12.2 1.248 1.333 7.933 10.998 make_m2s 4110 13.4 0.082 0.111 9.322 9.921 copy_fm_to_dbcsr 174 11.2 0.001 0.002 6.489 9.542 grid_integrate_task_list 110 12.3 9.401 9.459 9.401 9.459 transfer_fm_to_dbcsr 11 9.9 0.002 0.002 5.767 8.820 mp_alltoall_i22 605 13.7 5.458 8.588 5.458 8.588 multiply_cannon_metrocomm3 8220 15.4 0.029 0.052 5.402 8.256 qs_ot_get_derivative 99 11.5 0.002 0.002 7.814 8.089 init_scf_run 11 5.9 0.001 0.005 8.003 8.004 scf_env_initial_rho_setup 11 6.9 0.008 0.010 8.002 8.003 make_images 4110 14.4 1.482 1.563 7.325 7.966 ot_diis_step 99 11.5 0.032 0.040 7.383 7.383 apply_preconditioner_dbcsr 110 12.6 0.001 0.001 7.078 7.151 apply_single 110 13.6 0.000 0.001 7.077 7.150 qs_ot_get_p 110 10.4 0.001 0.001 6.643 7.063 cp_fm_cholesky_invert 11 10.9 6.387 6.395 6.387 6.395 fft_wrap_pw1pw2 1111 11.6 0.015 0.018 5.986 6.001 wfi_extrapolate 11 7.9 0.001 0.001 5.871 5.871 fft_wrap_pw1pw2_140 451 12.1 0.501 0.509 5.297 5.327 hybrid_alltoall_any 4261 16.3 0.418 0.966 4.327 5.307 make_images_data 4110 15.4 0.061 0.100 4.298 5.115 density_rs2pw 110 9.6 0.006 0.009 4.618 4.893 fft3d_ps 1111 13.6 2.430 2.479 4.560 4.582 qs_ot_p2m_diag 48 11.0 0.138 0.149 4.056 4.065 cp_dbcsr_syevd 48 12.0 0.004 0.004 3.657 3.660 cp_fm_cholesky_decompose 22 10.9 3.439 3.458 3.439 3.458 qs_energies_init_hamiltonians 11 5.9 0.002 0.002 3.452 3.454 multiply_cannon_metrocomm4 6165 15.4 0.028 0.052 1.196 3.453 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.388 3.393 mp_irecv_dv 17923 16.3 1.143 3.373 1.143 3.373 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.003 3.045 cp_fm_diag_elpa 48 13.0 0.000 0.000 3.026 3.027 cp_fm_diag_elpa_base 48 14.0 2.564 2.669 3.021 3.021 potential_pw2rs 110 12.3 0.028 0.031 2.972 2.976 mp_alltoall_d11v 2046 13.8 2.791 2.879 2.791 2.879 copy_dbcsr_to_fm 151 11.3 0.003 0.004 2.679 2.864 calculate_dm_sparse 110 9.5 0.001 0.001 2.557 2.724 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.531 2.623 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 2.420 2.547 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="214", plot="h2o_128_md", label="(4n/2r/18t)", y=120.318000, yerr=0.000000 PlotPoint: name="215", plot="h2o_128_md_mem", label="(4n/2r/18t)", y=1590.636364, yerr=217.413838 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/20/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 830472192 0.0% 100.0% 0.0% flops 22 x 32 x 32 1015021568 0.0% 100.0% 0.0% flops 32 x 256 x 256 184415158272 0.0% 100.0% 0.0% flops 49 x 256 x 256 194422767616 0.0% 100.0% 0.0% flops 71 x 256 x 256 246500294656 0.0% 100.0% 0.0% flops 9 x 9 x 256 269180485632 0.0% 100.0% 0.0% flops 9 x 22 x 256 349395425280 0.0% 100.0% 0.0% flops 22 x 9 x 256 350042406912 0.0% 100.0% 0.0% flops 80 x 256 x 256 396781158400 0.0% 100.0% 0.0% flops 49 x 256 x 2560 452145971200 0.0% 100.0% 0.0% flops 22 x 22 x 256 453581815808 0.0% 100.0% 0.0% flops 71 x 256 x 2560 573256499200 0.0% 100.0% 0.0% flops 80 x 256 x 2560 922746880000 0.0% 100.0% 0.0% flops 32 x 256 x 2560 1033476505600 0.0% 100.0% 0.0% flops 9 x 32 x 9 1138002296832 0.0% 100.0% 0.0% flops 9 x 32 x 22 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 9 1485592289280 0.0% 100.0% 0.0% flops 22 x 32 x 22 1910442074112 0.0% 100.0% 0.0% flops inhomo. stacks 1682398248960 100.0% 0.0% 0.0% flops total 13.129818E+12 12.8% 87.2% 0.0% flops max/rank 3.360079E+12 13.8% 86.2% 0.0% matmuls inhomo. stacks 46640 100.0% 0.0% 0.0% matmuls total 531185346 0.0% 100.0% 0.0% number of processed stacks 996500 4.7% 95.3% 0.0% average stack size 1.0 559.2 0.0 marketing flops 15.646297E+12 ------------------------------------------------------------------------------- # multiplications 2055 max memory usage/rank 8.328741E+09 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 16440 MPI messages size (bytes): total size 113.041801E+09 min size 0.000000E+00 max size 52.428800E+06 average size 6.876022E+06 MPI breakdown and total messages size (bytes): size <= 128 110 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 22 1441792 131072 < size <= 4194304 7388 3873439744 4194304 < size <= 16777216 7452 39069941760 16777216 < size 1468 70097023168 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 19 12. MP_Allreduce 10941 25. MP_Alltoall 8043 1122409. MP_ISend 16396 1767280. MP_IRecv 16396 1767297. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3469 77990. MP_Allreduce 9752 980. MP_Sync 52 MP_Alltoall 1474 32339426. MP_SendRecv 990 720533. MP_ISendRecv 990 720533. MP_Wait 2926 MP_ISend 1452 2662400. MP_IRecv 1452 2662400. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.129 0.144 143.783 143.784 qs_mol_dyn_low 1 2.0 0.097 0.100 142.616 142.621 qs_forces 11 3.9 0.089 0.102 141.313 141.315 qs_energies 11 4.9 0.046 0.049 132.972 132.978 scf_env_do_scf 11 5.9 0.007 0.009 114.531 114.536 velocity_verlet 10 3.0 0.001 0.002 86.239 86.325 scf_env_do_scf_inner_loop 99 6.5 0.057 0.072 86.097 86.109 dbcsr_multiply_generic 2055 12.4 0.256 0.299 43.519 43.700 qs_scf_new_mos 99 7.5 0.009 0.009 41.769 41.858 qs_scf_loop_do_ot 99 8.5 0.010 0.010 41.760 41.849 ot_scf_mini 99 9.5 0.022 0.022 38.844 38.901 init_scf_loop 11 6.9 0.104 0.108 28.251 28.257 rebuild_ks_matrix 110 8.3 0.001 0.001 27.221 27.286 qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.015 27.220 27.285 multiply_cannon 2055 13.4 0.508 0.547 24.056 27.176 qs_rho_update_rho_low 110 7.6 0.001 0.002 25.474 25.488 calculate_rho_elec 110 8.6 0.489 0.489 25.472 25.487 qs_ks_update_qs_env 110 7.6 0.001 0.001 24.774 24.833 prepare_preconditioner 11 7.9 0.000 0.000 23.291 23.317 make_preconditioner 11 8.9 0.002 0.002 23.291 23.317 make_full_inverse_cholesky 11 9.9 0.022 0.029 21.232 21.614 ot_mini 99 10.5 0.010 0.010 21.081 21.135 multiply_cannon_loop 2055 14.4 0.140 0.145 20.497 20.695 grid_collocate_task_list 110 9.6 17.801 17.932 17.801 17.932 make_m2s 4110 13.4 0.074 0.075 14.953 16.457 multiply_cannon_multrec 4110 15.4 14.746 15.073 14.815 15.146 sum_up_and_integrate 110 10.3 0.002 0.002 14.915 14.926 integrate_v_rspace 110 11.3 0.003 0.003 14.862 14.872 mp_waitall_1 67234 16.8 10.623 14.196 10.623 14.196 make_images 4110 14.4 1.987 2.060 10.414 11.894 qs_ot_get_derivative 99 11.5 0.003 0.004 10.661 10.713 cp_fm_cholesky_invert 11 10.9 10.416 10.425 10.416 10.425 ot_diis_step 99 11.5 0.066 0.075 10.396 10.396 init_scf_run 11 5.9 0.001 0.004 10.365 10.367 scf_env_initial_rho_setup 11 6.9 0.096 0.207 10.364 10.367 apply_preconditioner_dbcsr 110 12.6 0.009 0.009 9.937 10.033 grid_integrate_task_list 110 12.3 9.928 10.030 9.928 10.030 apply_single 110 13.6 0.000 0.000 9.928 10.025 qs_ot_get_p 110 10.4 0.001 0.001 9.593 9.752 hybrid_alltoall_any 4261 16.3 0.866 2.147 7.004 9.296 make_images_data 4110 15.4 0.052 0.057 6.288 8.602 fft_wrap_pw1pw2 1111 11.6 0.016 0.016 8.570 8.581 wfi_extrapolate 11 7.9 0.001 0.001 7.802 7.803 fft_wrap_pw1pw2_140 451 12.1 0.947 0.953 7.602 7.624 qs_energies_init_hamiltonians 11 5.9 0.008 0.008 6.615 6.617 fft3d_ps 1111 13.6 3.679 3.725 6.437 6.442 dbcsr_complete_redistribute 325 12.2 2.429 2.490 5.946 6.240 qs_ot_p2m_diag 48 11.0 0.241 0.252 6.088 6.101 density_rs2pw 110 9.6 0.005 0.005 5.861 5.992 cp_fm_cholesky_decompose 22 10.9 5.838 5.874 5.838 5.874 cp_dbcsr_syevd 48 12.0 0.004 0.004 5.550 5.551 multiply_cannon_metrocomm3 4110 15.4 0.008 0.009 4.109 5.297 mp_alltoall_d11v 2046 13.8 5.203 5.286 5.203 5.286 mp_allgather_i34 2055 14.4 2.147 5.075 2.147 5.075 copy_dbcsr_to_fm 151 11.3 0.003 0.003 4.972 5.015 cp_fm_diag_elpa 48 13.0 0.000 0.000 4.813 4.813 cp_fm_diag_elpa_base 48 14.0 4.471 4.531 4.810 4.810 dbcsr_make_dense_low 5207 15.5 0.054 0.054 4.589 4.724 make_dense_data 5207 16.5 4.160 4.323 4.516 4.650 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 4.564 4.614 dbcsr_make_images_dense 3552 14.7 0.027 0.027 4.159 4.331 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 4.113 4.262 qs_env_update_s_mstruct 11 6.9 0.001 0.001 3.883 3.907 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.711 3.713 copy_fm_to_dbcsr 174 11.2 0.001 0.001 3.380 3.635 transfer_dbcsr_to_fm 11 10.9 0.000 0.000 3.614 3.628 potential_pw2rs 110 12.3 0.045 0.045 3.458 3.461 calculate_dm_sparse 110 9.5 0.001 0.001 3.413 3.455 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.119 3.202 qs_create_task_list 11 7.9 0.000 0.000 3.169 3.181 generate_qs_task_list 11 8.9 1.796 1.860 3.169 3.181 qs_ot_get_derivative_diag 47 12.0 0.001 0.002 3.107 3.170 dbcsr_copy 1918 11.9 0.300 0.302 3.013 3.081 qs_ot_get_derivative_taylor 52 13.0 0.002 0.002 2.989 3.056 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="216", plot="h2o_128_md", label="(4n/1r/36t)", y=143.784000, yerr=0.000000 PlotPoint: name="217", plot="h2o_128_md_mem", label="(4n/1r/36t)", y=5899.636364, yerr=1526.510594 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/21/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 64 x 64 x 64 30366760960 0.0% 100.0% 0.0% flops 64 x 64 x 96 91100282880 0.0% 100.0% 0.0% flops 64 x 96 x 64 91100282880 0.0% 100.0% 0.0% flops 96 x 64 x 64 91100282880 0.0% 100.0% 0.0% flops 64 x 64 x 849 203420073984 0.0% 100.0% 0.0% flops 64 x 64 x 853 204378472448 0.0% 100.0% 0.0% flops 64 x 64 x 858 205576470528 0.0% 100.0% 0.0% flops 849 x 64 x 64 250602848256 0.0% 100.0% 0.0% flops 853 x 64 x 64 251783544832 0.0% 100.0% 0.0% flops 858 x 64 x 64 253259415552 0.0% 100.0% 0.0% flops 96 x 96 x 64 273300848640 0.0% 100.0% 0.0% flops 64 x 96 x 96 273300848640 0.0% 100.0% 0.0% flops 96 x 64 x 96 273300848640 0.0% 100.0% 0.0% flops 9 x 9 x 64 355059998208 0.0% 100.0% 0.0% flops 22 x 9 x 64 493014297600 0.0% 100.0% 0.0% flops 9 x 22 x 64 494442584064 0.0% 100.0% 0.0% flops 64 x 96 x 849 610260221952 0.0% 100.0% 0.0% flops 96 x 64 x 849 610260221952 0.0% 100.0% 0.0% flops 64 x 96 x 853 613135417344 0.0% 100.0% 0.0% flops 96 x 64 x 853 613135417344 0.0% 100.0% 0.0% flops 64 x 96 x 858 616729411584 0.0% 100.0% 0.0% flops 96 x 64 x 858 616729411584 0.0% 100.0% 0.0% flops 22 x 22 x 64 683571924992 0.0% 100.0% 0.0% flops 849 x 64 x 96 751808544768 0.0% 100.0% 0.0% flops 849 x 96 x 64 751808544768 0.0% 100.0% 0.0% flops 853 x 96 x 64 755350634496 0.0% 100.0% 0.0% flops 853 x 64 x 96 755350634496 0.0% 100.0% 0.0% flops 849 x 64 x 849 755814629376 0.0% 100.0% 0.0% flops 849 x 64 x 853 759375593472 0.0% 100.0% 0.0% flops 853 x 64 x 849 759375593472 0.0% 100.0% 0.0% flops 858 x 64 x 96 759778246656 0.0% 100.0% 0.0% flops 858 x 96 x 64 759778246656 0.0% 100.0% 0.0% flops 853 x 64 x 853 762953334784 0.0% 100.0% 0.0% flops 858 x 64 x 849 763826798592 0.0% 100.0% 0.0% flops 849 x 64 x 858 763826798592 0.0% 100.0% 0.0% flops 858 x 64 x 853 767425511424 0.0% 100.0% 0.0% flops 853 x 64 x 858 767425511424 0.0% 100.0% 0.0% flops 858 x 64 x 858 771923902464 0.0% 100.0% 0.0% flops 96 x 96 x 96 819902545920 0.0% 100.0% 0.0% flops 9 x 9 x 96 1065179994624 0.0% 100.0% 0.0% flops 22 x 9 x 96 1479042892800 0.0% 100.0% 0.0% flops 9 x 22 x 96 1483327752192 0.0% 100.0% 0.0% flops 96 x 96 x 849 1830780665856 0.0% 100.0% 0.0% flops 96 x 96 x 853 1839406252032 0.0% 100.0% 0.0% flops 96 x 96 x 858 1850188234752 0.0% 100.0% 0.0% flops 22 x 22 x 96 2050715774976 0.0% 100.0% 0.0% flops 849 x 96 x 96 2255425634304 0.0% 100.0% 0.0% flops 853 x 96 x 96 2266051903488 0.0% 100.0% 0.0% flops 849 x 96 x 849 2267443888128 0.0% 100.0% 0.0% flops 853 x 96 x 849 2278126780416 0.0% 100.0% 0.0% flops 849 x 96 x 853 2278126780416 0.0% 100.0% 0.0% flops 858 x 96 x 96 2279334739968 0.0% 100.0% 0.0% flops 853 x 96 x 853 2288860004352 0.0% 100.0% 0.0% flops 858 x 96 x 849 2291480395776 0.0% 100.0% 0.0% flops 849 x 96 x 858 2291480395776 0.0% 100.0% 0.0% flops 853 x 96 x 858 2302276534272 0.0% 100.0% 0.0% flops 858 x 96 x 853 2302276534272 0.0% 100.0% 0.0% flops 858 x 96 x 858 2315771707392 0.0% 100.0% 0.0% flops 9 x 32 x 9 5962613575680 0.0% 100.0% 0.0% flops 9 x 32 x 22 8325932617728 0.0% 100.0% 0.0% flops 22 x 32 x 9 8325932617728 0.0% 100.0% 0.0% flops 22 x 32 x 22 11452938371072 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 93.514751E+12 0.0% 100.0% 0.0% flops max/rank 743.796131E+09 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 3090542384 0.0% 100.0% 0.0% number of processed stacks 8143364 0.0% 100.0% 0.0% average stack size 0.0 379.5 0.0 marketing flops 144.582793E+12 ------------------------------------------------------------------------------- # multiplications 2507 max memory usage/rank 302.399488E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 7942176 MPI messages size (bytes): total size 4.882017E+12 min size 0.000000E+00 max size 5.889312E+06 average size 614.695188E+03 MPI breakdown and total messages size (bytes): size <= 128 50820 0 128 < size <= 8192 0 0 8192 < size <= 32768 419100 13717209088 32768 < size <= 131072 3341184 205282344960 131072 < size <= 4194304 3928320 3481881758608 4194304 < size <= 16777216 202752 1181116006400 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 65 12. MP_Allreduce 13450 37. MP_Alltoall 9654 122761. MP_ISend 120292 295011. MP_IRecv 120292 293276. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3999 67268. MP_Allreduce 11060 761. MP_Sync 87 MP_Alltoall 2483 1586374. MP_SendRecv 36608 12928. MP_ISendRecv 36608 12928. MP_Wait 53492 MP_ISend 14748 95649. MP_IRecv 14748 95649. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.051 0.052 173.493 173.496 qs_mol_dyn_low 1 2.0 0.004 0.005 173.001 173.018 qs_forces 11 3.9 0.005 0.006 172.858 172.869 qs_energies 11 4.9 0.001 0.002 167.019 167.082 scf_env_do_scf 11 5.9 0.001 0.001 153.180 153.198 scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 130.898 130.902 velocity_verlet 10 3.0 0.001 0.002 109.828 109.831 dbcsr_multiply_generic 2507 12.6 0.181 0.192 83.728 85.428 qs_scf_new_mos 117 7.6 0.001 0.001 77.762 79.269 qs_scf_loop_do_ot 117 8.6 0.001 0.001 77.762 79.268 ot_scf_mini 117 9.6 0.003 0.003 71.975 73.027 multiply_cannon 2507 13.6 0.228 0.272 61.918 66.860 multiply_cannon_loop 2507 14.6 0.300 0.329 57.610 61.815 multiply_cannon_multrec 30084 15.6 38.399 45.684 38.416 45.702 ot_mini 117 10.6 0.001 0.001 40.406 41.529 rebuild_ks_matrix 128 8.3 0.001 0.001 40.021 40.608 qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.019 40.020 40.607 qs_ks_update_qs_env 128 7.6 0.001 0.001 36.188 36.724 mp_waitall_1 319528 16.5 26.426 34.219 26.426 34.219 sum_up_and_integrate 128 10.3 0.002 0.004 22.686 22.740 integrate_v_rspace 128 11.3 0.004 0.005 22.632 22.696 qs_rho_update_rho_low 128 7.7 0.001 0.001 22.197 22.404 calculate_rho_elec 128 8.7 0.038 0.054 22.197 22.403 qs_ot_get_derivative 117 11.6 0.001 0.002 21.335 22.397 init_scf_loop 11 6.9 0.000 0.000 22.176 22.181 apply_preconditioner_dbcsr 128 12.6 0.000 0.001 18.781 19.790 apply_single 128 13.6 0.001 0.001 18.780 19.790 ot_diis_step 117 11.6 0.007 0.008 18.895 18.896 make_m2s 5014 13.6 0.113 0.121 16.801 17.849 qs_ot_get_p 128 10.4 0.001 0.001 16.200 17.568 prepare_preconditioner 11 7.9 0.000 0.000 17.156 17.290 make_preconditioner 11 8.9 0.000 0.000 17.156 17.290 multiply_cannon_metrocomm3 30084 15.6 0.105 0.118 5.221 16.462 make_images 5014 14.6 0.302 0.322 14.784 15.864 make_full_inverse_cholesky 11 9.9 0.000 0.000 15.042 15.407 multiply_cannon_metrocomm1 30084 15.6 0.132 0.146 10.750 15.045 grid_integrate_task_list 128 12.3 12.549 12.966 12.549 12.966 make_images_data 5014 15.6 0.068 0.077 9.253 11.558 hybrid_alltoall_any 5200 16.5 0.250 2.541 7.536 11.230 grid_collocate_task_list 128 9.7 10.755 11.196 10.755 11.196 qs_ot_p2m_diag 83 11.4 0.044 0.058 10.997 11.024 init_scf_run 11 5.9 0.000 0.001 10.512 10.514 scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.512 10.514 qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.574 10.348 cp_dbcsr_syevd 83 12.4 0.005 0.006 10.317 10.325 wfi_extrapolate 11 7.9 0.001 0.001 9.341 9.341 density_rs2pw 128 9.7 0.006 0.007 8.653 9.121 fft_wrap_pw1pw2 1291 11.7 0.016 0.020 8.566 8.892 mp_alltoall_d11v 2415 14.1 8.226 8.691 8.226 8.691 fft3d_ps 1291 13.7 1.752 2.210 7.668 7.950 fft_wrap_pw1pw2_140 523 12.2 0.195 0.210 7.044 7.410 potential_pw2rs 128 12.3 0.007 0.008 7.077 7.144 calculate_dm_sparse 128 9.5 0.001 0.001 6.427 7.071 mp_sum_l 12367 13.3 3.412 7.028 3.412 7.028 cp_fm_cholesky_invert 11 10.9 6.899 6.913 6.899 6.913 cp_fm_diag_elpa 83 13.4 0.000 0.001 6.706 6.710 cp_fm_diag_elpa_base 83 14.4 6.645 6.668 6.694 6.694 cp_fm_cholesky_decompose 22 10.9 6.467 6.533 6.467 6.533 mp_irecv_dv 71972 16.2 2.512 6.130 2.512 6.130 multiply_cannon_metrocomm4 27577 15.6 0.100 0.114 2.224 5.927 mp_allgather_i34 2507 14.6 2.007 5.740 2.007 5.740 make_images_sizes 5014 15.6 0.007 0.007 2.854 5.578 mp_alltoall_i44 5014 16.6 2.847 5.571 2.847 5.571 mp_waitany 14748 13.8 4.139 5.268 4.139 5.268 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 4.874 4.914 transfer_rs2pw 523 10.6 0.006 0.008 4.449 4.899 dbcsr_complete_redistribute 395 12.7 0.453 0.631 4.422 4.851 transfer_pw2rs 523 13.3 0.006 0.008 4.627 4.696 mp_alltoall_z22v 1291 15.7 4.351 4.647 4.351 4.647 mp_sum_d 4469 12.1 2.562 4.393 2.562 4.393 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 4.175 4.224 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.908 4.089 qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.501 3.794 dbcsr_dot_sd 1318 12.0 0.479 0.533 1.938 3.757 copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.213 3.563 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="400", plot="h2o_256_md", label="(4n/36r/1t)", y=173.496000, yerr=0.000000 PlotPoint: name="401", plot="h2o_256_md_mem", label="(4n/36r/1t)", y=285.181818, yerr=3.185712 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/22/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 64 x 64 x 64 30366760960 0.0% 100.0% 0.0% flops 778 x 64 x 64 57411371008 0.0% 100.0% 0.0% flops 782 x 64 x 64 57706545152 0.0% 100.0% 0.0% flops 800 x 64 x 64 59034828800 0.0% 100.0% 0.0% flops 907 x 64 x 64 66930737152 0.0% 100.0% 0.0% flops 920 x 64 x 64 67890053120 0.0% 100.0% 0.0% flops 929 x 64 x 64 68554194944 0.0% 100.0% 0.0% flops 942 x 64 x 64 69513510912 0.0% 100.0% 0.0% flops 778 x 64 x 840 85658173440 0.0% 100.0% 0.0% flops 782 x 64 x 840 86098575360 0.0% 100.0% 0.0% flops 800 x 64 x 840 88080384000 0.0% 100.0% 0.0% flops 64 x 64 x 96 91100282880 0.0% 100.0% 0.0% flops 96 x 64 x 64 91100282880 0.0% 100.0% 0.0% flops 64 x 96 x 64 91100282880 0.0% 100.0% 0.0% flops 907 x 64 x 840 99861135360 0.0% 100.0% 0.0% flops 64 x 64 x 840 100631838720 0.0% 100.0% 0.0% flops 920 x 64 x 840 101292441600 0.0% 100.0% 0.0% flops 929 x 64 x 840 102283345920 0.0% 100.0% 0.0% flops 942 x 64 x 840 103714652160 0.0% 100.0% 0.0% flops 951 x 64 x 64 140355305472 0.0% 100.0% 0.0% flops 760 x 64 x 64 168249262080 0.0% 100.0% 0.0% flops 778 x 64 x 96 172234113024 0.0% 100.0% 0.0% flops 778 x 96 x 64 172234113024 0.0% 100.0% 0.0% flops 782 x 64 x 96 173119635456 0.0% 100.0% 0.0% flops 782 x 96 x 64 173119635456 0.0% 100.0% 0.0% flops 778 x 64 x 849 173151879168 0.0% 100.0% 0.0% flops 782 x 64 x 849 174042120192 0.0% 100.0% 0.0% flops 778 x 64 x 862 175803203584 0.0% 100.0% 0.0% flops 782 x 64 x 862 176707076096 0.0% 100.0% 0.0% flops 800 x 64 x 96 177104486400 0.0% 100.0% 0.0% flops 800 x 96 x 64 177104486400 0.0% 100.0% 0.0% flops 800 x 64 x 849 178048204800 0.0% 100.0% 0.0% flops 800 x 64 x 862 180774502400 0.0% 100.0% 0.0% flops 907 x 64 x 96 200792211456 0.0% 100.0% 0.0% flops 907 x 96 x 64 200792211456 0.0% 100.0% 0.0% flops 907 x 64 x 849 201862152192 0.0% 100.0% 0.0% flops 64 x 64 x 849 203420073984 0.0% 100.0% 0.0% flops 920 x 64 x 96 203670159360 0.0% 100.0% 0.0% flops 920 x 96 x 64 203670159360 0.0% 100.0% 0.0% flops 920 x 64 x 849 204755435520 0.0% 100.0% 0.0% flops 907 x 64 x 862 204953092096 0.0% 100.0% 0.0% flops 929 x 64 x 96 205662584832 0.0% 100.0% 0.0% flops 929 x 96 x 64 205662584832 0.0% 100.0% 0.0% flops 64 x 64 x 862 206534868992 0.0% 100.0% 0.0% flops 929 x 64 x 849 206758477824 0.0% 100.0% 0.0% flops 920 x 64 x 862 207890677760 0.0% 100.0% 0.0% flops 942 x 96 x 64 208540532736 0.0% 100.0% 0.0% flops 942 x 64 x 96 208540532736 0.0% 100.0% 0.0% flops 951 x 64 x 840 209411112960 0.0% 100.0% 0.0% flops 942 x 64 x 849 209651761152 0.0% 100.0% 0.0% flops 929 x 64 x 862 209924390912 0.0% 100.0% 0.0% flops 942 x 64 x 862 212861976576 0.0% 100.0% 0.0% flops 760 x 64 x 840 251029094400 0.0% 100.0% 0.0% flops 778 x 96 x 840 256974520320 0.0% 100.0% 0.0% flops 782 x 96 x 840 258295726080 0.0% 100.0% 0.0% flops 800 x 96 x 840 264241152000 0.0% 100.0% 0.0% flops 96 x 96 x 64 273300848640 0.0% 100.0% 0.0% flops 96 x 64 x 96 273300848640 0.0% 100.0% 0.0% flops 64 x 96 x 96 273300848640 0.0% 100.0% 0.0% flops 907 x 96 x 840 299583406080 0.0% 100.0% 0.0% flops 96 x 64 x 840 301895516160 0.0% 100.0% 0.0% flops 64 x 96 x 840 301895516160 0.0% 100.0% 0.0% flops 920 x 96 x 840 303877324800 0.0% 100.0% 0.0% flops 929 x 96 x 840 306850037760 0.0% 100.0% 0.0% flops 942 x 96 x 840 311143956480 0.0% 100.0% 0.0% flops 9 x 9 x 64 355059998208 0.0% 100.0% 0.0% flops 951 x 64 x 96 421065916416 0.0% 100.0% 0.0% flops 951 x 96 x 64 421065916416 0.0% 100.0% 0.0% flops 951 x 64 x 849 423309606912 0.0% 100.0% 0.0% flops 951 x 64 x 862 429791379456 0.0% 100.0% 0.0% flops 22 x 9 x 64 493014297600 0.0% 100.0% 0.0% flops 9 x 22 x 64 494442584064 0.0% 100.0% 0.0% flops 760 x 96 x 64 504747786240 0.0% 100.0% 0.0% flops 760 x 64 x 96 504747786240 0.0% 100.0% 0.0% flops 760 x 64 x 849 507437383680 0.0% 100.0% 0.0% flops 760 x 64 x 862 515207331840 0.0% 100.0% 0.0% flops 778 x 96 x 96 516702339072 0.0% 100.0% 0.0% flops 782 x 96 x 96 519358906368 0.0% 100.0% 0.0% flops 778 x 96 x 849 519455637504 0.0% 100.0% 0.0% flops 782 x 96 x 849 522126360576 0.0% 100.0% 0.0% flops 778 x 96 x 862 527409610752 0.0% 100.0% 0.0% flops 782 x 96 x 862 530121228288 0.0% 100.0% 0.0% flops 800 x 96 x 96 531313459200 0.0% 100.0% 0.0% flops 800 x 96 x 849 534144614400 0.0% 100.0% 0.0% flops 800 x 96 x 862 542323507200 0.0% 100.0% 0.0% flops 907 x 96 x 96 602376634368 0.0% 100.0% 0.0% flops 907 x 96 x 849 605586456576 0.0% 100.0% 0.0% flops 96 x 64 x 849 610260221952 0.0% 100.0% 0.0% flops 64 x 96 x 849 610260221952 0.0% 100.0% 0.0% flops 920 x 96 x 96 611010478080 0.0% 100.0% 0.0% flops 920 x 96 x 849 614266306560 0.0% 100.0% 0.0% flops 907 x 96 x 862 614859276288 0.0% 100.0% 0.0% flops 929 x 96 x 96 616987754496 0.0% 100.0% 0.0% flops 96 x 64 x 862 619604606976 0.0% 100.0% 0.0% flops 64 x 96 x 862 619604606976 0.0% 100.0% 0.0% flops 929 x 96 x 849 620275433472 0.0% 100.0% 0.0% flops 920 x 96 x 862 623672033280 0.0% 100.0% 0.0% flops 942 x 96 x 96 625621598208 0.0% 100.0% 0.0% flops 951 x 96 x 840 628233338880 0.0% 100.0% 0.0% flops 942 x 96 x 849 628955283456 0.0% 100.0% 0.0% flops 929 x 96 x 862 629773172736 0.0% 100.0% 0.0% flops 942 x 96 x 862 638585929728 0.0% 100.0% 0.0% flops 22 x 22 x 64 683571924992 0.0% 100.0% 0.0% flops 760 x 96 x 840 753087283200 0.0% 100.0% 0.0% flops 96 x 96 x 96 819902545920 0.0% 100.0% 0.0% flops 96 x 96 x 840 905686548480 0.0% 100.0% 0.0% flops 9 x 9 x 96 1065179994624 0.0% 100.0% 0.0% flops 951 x 96 x 96 1263197749248 0.0% 100.0% 0.0% flops 951 x 96 x 849 1269928820736 0.0% 100.0% 0.0% flops 951 x 96 x 862 1289374138368 0.0% 100.0% 0.0% flops 22 x 9 x 96 1479042892800 0.0% 100.0% 0.0% flops 9 x 22 x 96 1483327752192 0.0% 100.0% 0.0% flops 760 x 96 x 96 1514243358720 0.0% 100.0% 0.0% flops 760 x 96 x 849 1522312151040 0.0% 100.0% 0.0% flops 760 x 96 x 862 1545621995520 0.0% 100.0% 0.0% flops 96 x 96 x 849 1830780665856 0.0% 100.0% 0.0% flops 96 x 96 x 862 1858813820928 0.0% 100.0% 0.0% flops 22 x 22 x 96 2050715774976 0.0% 100.0% 0.0% flops 9 x 32 x 9 5962613575680 0.0% 100.0% 0.0% flops 22 x 32 x 9 8325932617728 0.0% 100.0% 0.0% flops 9 x 32 x 22 8325932617728 0.0% 100.0% 0.0% flops 22 x 32 x 22 11452938371072 0.0% 100.0% 0.0% flops inhomo. stacks 6920499888128 100.0% 0.0% 0.0% flops total 94.184287E+12 7.3% 92.7% 0.0% flops max/rank 1.486220E+12 7.4% 92.6% 0.0% matmuls inhomo. stacks 168480 100.0% 0.0% 0.0% matmuls total 3090542384 0.0% 100.0% 0.0% number of processed stacks 8217740 2.1% 97.9% 0.0% average stack size 1.0 383.9 0.0 marketing flops 144.582793E+12 ------------------------------------------------------------------------------- # multiplications 2507 max memory usage/rank 411.529216E+06 # max total images/rank 2 # max 3D layers 1 # MPI messages exchanged 3790584 MPI messages size (bytes): total size 4.296164E+12 min size 0.000000E+00 max size 11.799056E+06 average size 1.133378E+06 MPI breakdown and total messages size (bytes): size <= 128 23892 0 128 < size <= 8192 0 0 8192 < size <= 32768 118100 3862691840 32768 < size <= 131072 1497000 117745909760 131072 < size <= 4194304 1818160 1529661358080 4194304 < size <= 16777216 333432 2644950338448 16777216 < size 0 0 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3999 67595. MP_Allreduce 11058 879. MP_Sync 87 MP_Alltoall 1969 4195840. MP_SendRecv 18176 24736. MP_ISendRecv 18176 24736. MP_Wait 36332 MP_ISend 16020 128768. MP_IRecv 16020 128768. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.042 0.073 270.656 270.667 qs_mol_dyn_low 1 2.0 0.102 0.169 269.947 269.981 qs_forces 11 3.9 0.014 0.040 269.536 269.546 qs_energies 11 4.9 0.052 0.204 260.129 260.193 scf_env_do_scf 11 5.9 0.003 0.008 239.319 239.391 scf_env_do_scf_inner_loop 117 6.6 0.004 0.039 208.866 208.871 velocity_verlet 10 3.0 0.017 0.025 169.631 169.648 dbcsr_multiply_generic 2507 12.6 0.251 0.278 143.137 146.811 qs_scf_new_mos 117 7.6 0.001 0.002 128.461 130.346 qs_scf_loop_do_ot 117 8.6 0.001 0.002 128.460 130.344 ot_scf_mini 117 9.6 0.005 0.007 119.397 121.052 multiply_cannon 2507 13.6 0.335 0.432 108.602 115.556 multiply_cannon_loop 2507 14.6 0.595 0.674 101.987 111.429 multiply_cannon_multrec 30084 15.6 82.884 96.053 82.923 96.093 ot_mini 117 10.6 0.001 0.002 71.443 73.527 rebuild_ks_matrix 128 8.3 0.001 0.001 60.571 61.944 qs_ks_build_kohn_sham_matrix 128 9.3 0.022 0.030 60.570 61.944 qs_ks_update_qs_env 128 7.6 0.001 0.002 54.668 55.928 mp_waitall_1 240928 16.6 22.835 54.411 22.835 54.411 multiply_cannon_metrocomm3 30084 15.6 0.156 0.177 9.842 41.813 apply_preconditioner_dbcsr 128 12.6 0.001 0.001 36.796 39.132 apply_single 128 13.6 0.001 0.002 36.795 39.131 ot_diis_step 117 11.6 0.081 0.274 37.919 37.924 qs_ot_get_derivative 117 11.6 0.002 0.003 33.178 34.909 qs_rho_update_rho_low 128 7.7 0.001 0.001 33.769 33.998 calculate_rho_elec 128 8.7 0.075 0.081 33.768 33.997 sum_up_and_integrate 128 10.3 0.004 0.008 31.071 31.123 integrate_v_rspace 128 11.3 0.005 0.006 30.965 31.025 init_scf_loop 11 6.9 0.001 0.004 30.308 30.350 make_m2s 5014 13.6 0.158 0.177 24.092 25.039 grid_collocate_task_list 128 9.7 21.813 22.674 21.813 22.674 qs_ot_get_p 128 10.4 0.001 0.003 20.039 22.666 prepare_preconditioner 11 7.9 0.000 0.001 22.385 22.607 make_preconditioner 11 8.9 0.000 0.002 22.385 22.606 grid_integrate_task_list 128 12.3 20.819 21.656 20.819 21.656 make_images 5014 14.6 1.362 1.693 20.353 21.623 make_full_inverse_cholesky 11 9.9 0.000 0.001 20.998 21.403 multiply_cannon_metrocomm4 27577 15.6 0.168 0.190 6.061 15.809 qs_ot_get_derivative_diag 77 12.4 0.003 0.004 14.354 15.639 mp_irecv_dv 74558 16.2 6.140 15.519 6.140 15.519 init_scf_run 11 5.9 0.000 0.009 15.496 15.498 scf_env_initial_rho_setup 11 6.9 0.001 0.008 15.496 15.497 mp_sum_l 12367 13.3 6.977 14.569 6.977 14.569 make_images_data 5014 15.6 0.090 0.108 11.699 13.692 wfi_extrapolate 11 7.9 0.002 0.003 13.569 13.570 hybrid_alltoall_any 5200 16.5 0.461 2.824 10.112 13.142 qs_ot_p2m_diag 83 11.4 0.114 0.151 12.106 12.230 fft_wrap_pw1pw2 1291 11.7 0.029 0.038 11.574 11.838 cp_fm_cholesky_invert 11 10.9 11.387 11.415 11.387 11.415 cp_dbcsr_syevd 83 12.4 0.007 0.007 11.275 11.335 calculate_dm_sparse 128 9.5 0.001 0.002 10.038 10.954 fft_wrap_pw1pw2_140 523 12.2 0.525 0.546 10.368 10.589 density_rs2pw 128 9.7 0.009 0.010 9.484 10.341 fft3d_ps 1291 13.7 4.233 4.489 9.194 9.402 cp_dbcsr_sm_fm_multiply 37 9.5 0.003 0.004 8.063 8.086 mp_sum_d 4469 12.1 4.503 8.041 4.503 8.041 dbcsr_dot_sd 1318 12.0 1.035 1.121 4.536 8.005 mp_alltoall_d11v 2415 14.1 6.729 7.635 6.729 7.635 cp_fm_diag_elpa 83 13.4 0.001 0.001 7.584 7.592 cp_fm_diag_elpa_base 83 14.4 7.432 7.454 7.570 7.570 qs_ot_get_orbitals 117 10.6 0.001 0.002 7.300 7.542 potential_pw2rs 128 12.3 0.017 0.020 7.279 7.344 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 6.540 6.930 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 6.537 6.709 cp_fm_cholesky_decompose 22 10.9 6.254 6.308 6.254 6.308 qs_ot_get_derivative_taylor 40 13.0 0.002 0.002 5.350 5.854 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="402", plot="h2o_256_md", label="(4n/18r/2t)", y=270.667000, yerr=0.000000 PlotPoint: name="403", plot="h2o_256_md_mem", label="(4n/18r/2t)", y=389.090909, yerr=5.648082 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/23/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 128 x 64 x 64 15300820992 0.0% 100.0% 0.0% flops 64 x 64 x 64 38252052480 0.0% 100.0% 0.0% flops 128 x 96 x 64 45902462976 0.0% 100.0% 0.0% flops 128 x 64 x 96 45902462976 0.0% 100.0% 0.0% flops 96 x 64 x 64 68853694464 0.0% 100.0% 0.0% flops 128 x 64 x 862 104171307008 0.0% 100.0% 0.0% flops 64 x 96 x 64 114756157440 0.0% 100.0% 0.0% flops 64 x 64 x 96 114756157440 0.0% 100.0% 0.0% flops 876 x 64 x 64 130434465792 0.0% 100.0% 0.0% flops 911 x 64 x 64 135645888512 0.0% 100.0% 0.0% flops 128 x 96 x 96 137707388928 0.0% 100.0% 0.0% flops 929 x 64 x 64 138326048768 0.0% 100.0% 0.0% flops 964 x 64 x 64 143537471488 0.0% 100.0% 0.0% flops 128 x 64 x 849 205200556032 0.0% 100.0% 0.0% flops 96 x 64 x 96 206561083392 0.0% 100.0% 0.0% flops 96 x 96 x 64 206561083392 0.0% 100.0% 0.0% flops 720 x 64 x 64 214412820480 0.0% 100.0% 0.0% flops 64 x 64 x 862 260428267520 0.0% 100.0% 0.0% flops 128 x 96 x 862 312513921024 0.0% 100.0% 0.0% flops 64 x 96 x 96 344268472320 0.0% 100.0% 0.0% flops 9 x 9 x 64 357615005184 0.0% 100.0% 0.0% flops 876 x 64 x 96 391303397376 0.0% 100.0% 0.0% flops 876 x 96 x 64 391303397376 0.0% 100.0% 0.0% flops 876 x 64 x 862 398989099008 0.0% 100.0% 0.0% flops 911 x 96 x 64 406937665536 0.0% 100.0% 0.0% flops 911 x 64 x 96 406937665536 0.0% 100.0% 0.0% flops 911 x 64 x 862 414930444288 0.0% 100.0% 0.0% flops 929 x 96 x 64 414978146304 0.0% 100.0% 0.0% flops 929 x 64 x 96 414978146304 0.0% 100.0% 0.0% flops 929 x 64 x 862 423128850432 0.0% 100.0% 0.0% flops 964 x 64 x 96 430612414464 0.0% 100.0% 0.0% flops 964 x 96 x 64 430612414464 0.0% 100.0% 0.0% flops 964 x 64 x 862 439070195712 0.0% 100.0% 0.0% flops 96 x 64 x 862 468770881536 0.0% 100.0% 0.0% flops 22 x 9 x 64 496561241088 0.0% 100.0% 0.0% flops 9 x 22 x 64 498000070656 0.0% 100.0% 0.0% flops 64 x 64 x 849 513001390080 0.0% 100.0% 0.0% flops 128 x 96 x 849 615601668096 0.0% 100.0% 0.0% flops 96 x 96 x 96 619683250176 0.0% 100.0% 0.0% flops 720 x 96 x 64 643238461440 0.0% 100.0% 0.0% flops 720 x 64 x 96 643238461440 0.0% 100.0% 0.0% flops 720 x 64 x 862 655872491520 0.0% 100.0% 0.0% flops 22 x 22 x 64 688489179136 0.0% 100.0% 0.0% flops 64 x 96 x 862 781284802560 0.0% 100.0% 0.0% flops 876 x 64 x 849 785943724032 0.0% 100.0% 0.0% flops 911 x 64 x 849 817345585152 0.0% 100.0% 0.0% flops 929 x 64 x 849 833495113728 0.0% 100.0% 0.0% flops 964 x 64 x 849 864896974848 0.0% 100.0% 0.0% flops 96 x 64 x 849 923402502144 0.0% 100.0% 0.0% flops 9 x 9 x 96 1072845015552 0.0% 100.0% 0.0% flops 876 x 96 x 96 1173910192128 0.0% 100.0% 0.0% flops 876 x 96 x 862 1196967297024 0.0% 100.0% 0.0% flops 911 x 96 x 96 1220812996608 0.0% 100.0% 0.0% flops 911 x 96 x 862 1244791332864 0.0% 100.0% 0.0% flops 929 x 96 x 96 1244934438912 0.0% 100.0% 0.0% flops 929 x 96 x 862 1269386551296 0.0% 100.0% 0.0% flops 964 x 96 x 96 1291837243392 0.0% 100.0% 0.0% flops 720 x 64 x 849 1291962286080 0.0% 100.0% 0.0% flops 964 x 96 x 862 1317210587136 0.0% 100.0% 0.0% flops 96 x 96 x 862 1406312644608 0.0% 100.0% 0.0% flops 22 x 9 x 96 1489683723264 0.0% 100.0% 0.0% flops 9 x 22 x 96 1494000211968 0.0% 100.0% 0.0% flops 64 x 96 x 849 1539004170240 0.0% 100.0% 0.0% flops 720 x 96 x 96 1929715384320 0.0% 100.0% 0.0% flops 720 x 96 x 862 1967617474560 0.0% 100.0% 0.0% flops 22 x 22 x 96 2065467537408 0.0% 100.0% 0.0% flops 876 x 96 x 849 2357831172096 0.0% 100.0% 0.0% flops 911 x 96 x 849 2452036755456 0.0% 100.0% 0.0% flops 929 x 96 x 849 2500485341184 0.0% 100.0% 0.0% flops 964 x 96 x 849 2594690924544 0.0% 100.0% 0.0% flops 96 x 96 x 849 2770207506432 0.0% 100.0% 0.0% flops 720 x 96 x 849 3875886858240 0.0% 100.0% 0.0% flops 9 x 32 x 9 6003325476864 0.0% 100.0% 0.0% flops 22 x 32 x 9 8382768058368 0.0% 100.0% 0.0% flops 9 x 32 x 22 8382768058368 0.0% 100.0% 0.0% flops 22 x 32 x 22 11531106926592 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 94.228663E+12 0.0% 100.0% 0.0% flops max/rank 2.199914E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 3111737268 0.0% 100.0% 0.0% number of processed stacks 8336396 0.0% 100.0% 0.0% average stack size 0.0 373.3 0.0 marketing flops 145.647559E+12 ------------------------------------------------------------------------------- # multiplications 2527 max memory usage/rank 526.196736E+06 # max total images/rank 3 # max 3D layers 1 # MPI messages exchanged 2425920 MPI messages size (bytes): total size 4.132350E+12 min size 0.000000E+00 max size 17.653760E+06 average size 1.703416E+06 MPI breakdown and total messages size (bytes): size <= 128 14916 0 128 < size <= 8192 0 0 8192 < size <= 32768 71436 2336489472 32768 < size <= 131072 728832 55956209664 131072 < size <= 4194304 1386864 1409906900992 4194304 < size <= 16777216 155760 1473826772352 16777216 < size 68112 1190343475200 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4031 67293. MP_Allreduce 11138 958. MP_Sync 87 MP_Alltoall 1983 4911023. MP_SendRecv 12126 47072. MP_ISendRecv 12126 47072. MP_Wait 26114 MP_ISend 11836 212447. MP_IRecv 11836 212447. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.041 0.065 249.964 249.967 qs_mol_dyn_low 1 2.0 0.092 0.125 249.344 249.361 qs_forces 11 3.9 0.004 0.005 248.639 248.651 qs_energies 11 4.9 0.001 0.002 240.607 240.664 scf_env_do_scf 11 5.9 0.001 0.001 221.306 221.332 scf_env_do_scf_inner_loop 118 6.6 0.003 0.011 189.374 189.379 velocity_verlet 10 3.0 0.001 0.001 159.052 159.081 dbcsr_multiply_generic 2527 12.6 0.240 0.315 134.610 138.480 qs_scf_new_mos 118 7.6 0.001 0.001 122.022 123.472 qs_scf_loop_do_ot 118 8.6 0.001 0.002 122.021 123.471 ot_scf_mini 118 9.6 0.004 0.005 113.736 115.480 multiply_cannon 2527 13.6 0.321 0.373 100.908 106.489 multiply_cannon_loop 2527 14.6 0.577 0.717 95.867 103.048 multiply_cannon_multrec 30324 15.6 75.517 93.655 75.553 93.692 ot_mini 118 10.6 0.001 0.001 66.796 68.732 rebuild_ks_matrix 129 8.3 0.001 0.001 51.708 53.459 qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.020 51.707 53.459 mp_waitall_1 216438 16.6 23.497 50.849 23.497 50.849 qs_ks_update_qs_env 129 7.6 0.001 0.002 46.751 48.388 multiply_cannon_metrocomm3 30324 15.6 0.137 0.254 12.300 39.534 apply_preconditioner_dbcsr 129 12.6 0.000 0.001 35.780 38.603 apply_single 129 13.6 0.001 0.001 35.780 38.603 ot_diis_step 118 11.6 0.021 0.024 36.825 36.828 init_scf_loop 11 6.9 0.000 0.000 31.805 31.808 qs_ot_get_derivative 118 11.6 0.002 0.002 29.672 31.461 qs_rho_update_rho_low 129 7.7 0.001 0.001 27.083 27.310 calculate_rho_elec 129 8.7 0.110 0.133 27.082 27.309 sum_up_and_integrate 129 10.3 0.002 0.004 25.392 25.485 integrate_v_rspace 129 11.3 0.004 0.005 25.322 25.449 prepare_preconditioner 11 7.9 0.000 0.000 24.861 25.017 make_preconditioner 11 8.9 0.000 0.000 24.861 25.017 make_full_inverse_cholesky 11 9.9 0.000 0.001 23.128 23.762 make_m2s 5054 13.6 0.152 0.191 22.601 23.692 qs_ot_get_p 129 10.4 0.001 0.002 19.888 22.483 make_images 5054 14.6 1.396 2.007 19.139 20.374 grid_integrate_task_list 129 12.3 17.004 17.699 17.004 17.699 grid_collocate_task_list 129 9.7 16.601 17.671 16.601 17.671 multiply_cannon_metrocomm4 27797 15.6 0.147 0.268 6.879 17.497 mp_irecv_dv 70031 16.3 6.674 17.236 6.674 17.236 init_scf_run 11 5.9 0.000 0.001 14.673 14.676 scf_env_initial_rho_setup 11 6.9 0.000 0.001 14.673 14.675 qs_ot_get_derivative_diag 77 12.4 0.003 0.004 12.798 14.063 mp_sum_l 12463 13.3 7.525 14.030 7.525 14.030 make_images_data 5054 15.6 0.085 0.136 11.368 12.689 wfi_extrapolate 11 7.9 0.001 0.001 12.326 12.326 qs_ot_p2m_diag 83 11.4 0.163 0.195 12.113 12.186 cp_fm_cholesky_invert 11 10.9 11.983 12.010 11.983 12.010 hybrid_alltoall_any 5240 16.5 0.335 1.430 10.021 11.596 cp_dbcsr_syevd 83 12.4 0.006 0.007 11.127 11.140 fft_wrap_pw1pw2 1301 11.7 0.020 0.026 10.024 10.514 density_rs2pw 129 9.7 0.007 0.011 8.145 10.282 calculate_dm_sparse 129 9.5 0.001 0.002 9.167 9.899 fft3d_ps 1301 13.7 3.017 3.860 8.390 9.424 mp_sum_d 4503 12.1 4.837 9.236 4.837 9.236 fft_wrap_pw1pw2_140 527 12.2 0.331 0.393 8.747 8.920 mp_alltoall_d11v 2423 14.1 6.420 7.777 6.420 7.777 dbcsr_dot_sd 1330 12.0 0.927 1.073 4.541 7.736 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 7.664 7.681 cp_fm_diag_elpa 83 13.4 0.001 0.001 7.378 7.386 cp_fm_diag_elpa_base 83 14.4 7.133 7.175 7.362 7.363 mp_alltoall_z22v 1301 15.7 4.697 7.200 4.697 7.200 qs_ot_get_orbitals 118 10.6 0.001 0.001 6.385 6.590 cp_fm_cholesky_decompose 22 10.9 6.374 6.563 6.374 6.563 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 6.135 6.544 mp_waitany 11836 13.9 3.116 6.125 3.116 6.125 potential_pw2rs 129 12.3 0.020 0.026 5.849 6.042 transfer_rs2pw 527 10.6 0.007 0.009 3.309 5.759 dbcsr_complete_redistribute 395 12.7 0.978 1.178 4.594 5.705 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 5.500 5.660 make_images_sizes 5054 15.6 0.008 0.017 2.770 5.606 mp_alltoall_i44 5054 16.6 2.761 5.599 2.761 5.599 qs_ot_get_derivative_taylor 41 13.0 0.001 0.002 4.961 5.483 mp_allgather_i34 2527 14.6 2.092 5.301 2.092 5.301 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="404", plot="h2o_256_md", label="(4n/12r/3t)", y=249.967000, yerr=0.000000 PlotPoint: name="405", plot="h2o_256_md_mem", label="(4n/12r/3t)", y=483.545455, yerr=2.807899 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/24/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 96 x 192 x 192 25621954560 0.0% 100.0% 0.0% flops 96 x 160 x 192 42703257600 0.0% 100.0% 0.0% flops 96 x 192 x 160 42703257600 0.0% 100.0% 0.0% flops 347 x 192 x 192 57614303232 0.0% 100.0% 0.0% flops 369 x 192 x 192 61267083264 0.0% 100.0% 0.0% flops 96 x 160 x 160 71172096000 0.0% 100.0% 0.0% flops 347 x 160 x 192 96023838720 0.0% 100.0% 0.0% flops 347 x 192 x 160 96023838720 0.0% 100.0% 0.0% flops 369 x 160 x 192 102111805440 0.0% 100.0% 0.0% flops 369 x 192 x 160 102111805440 0.0% 100.0% 0.0% flops 64 x 192 x 192 102487818240 0.0% 100.0% 0.0% flops 320 x 192 x 192 106262691840 0.0% 100.0% 0.0% flops 342 x 192 x 192 113568251904 0.0% 100.0% 0.0% flops 96 x 192 x 1702 114693341184 0.0% 100.0% 0.0% flops 96 x 192 x 1707 115030278144 0.0% 100.0% 0.0% flops 96 x 192 x 1711 115299827712 0.0% 100.0% 0.0% flops 347 x 192 x 1702 116115505152 0.0% 100.0% 0.0% flops 347 x 192 x 1707 116456620032 0.0% 100.0% 0.0% flops 347 x 192 x 1711 116729511936 0.0% 100.0% 0.0% flops 369 x 192 x 1702 123477295104 0.0% 100.0% 0.0% flops 369 x 192 x 1707 123840036864 0.0% 100.0% 0.0% flops 369 x 192 x 1711 124130230272 0.0% 100.0% 0.0% flops 32 x 192 x 192 145191075840 0.0% 100.0% 0.0% flops 440 x 192 x 192 146111201280 0.0% 100.0% 0.0% flops 462 x 192 x 192 153416761344 0.0% 100.0% 0.0% flops 471 x 192 x 192 156405399552 0.0% 100.0% 0.0% flops 347 x 160 x 160 160039731200 0.0% 100.0% 0.0% flops 369 x 160 x 160 170186342400 0.0% 100.0% 0.0% flops 64 x 192 x 160 170813030400 0.0% 100.0% 0.0% flops 64 x 160 x 192 170813030400 0.0% 100.0% 0.0% flops 320 x 192 x 160 177104486400 0.0% 100.0% 0.0% flops 320 x 160 x 192 177104486400 0.0% 100.0% 0.0% flops 342 x 160 x 192 189280419840 0.0% 100.0% 0.0% flops 342 x 192 x 160 189280419840 0.0% 100.0% 0.0% flops 96 x 160 x 1702 191155568640 0.0% 100.0% 0.0% flops 96 x 160 x 1707 191717130240 0.0% 100.0% 0.0% flops 96 x 160 x 1711 192166379520 0.0% 100.0% 0.0% flops 347 x 160 x 1702 193525841920 0.0% 100.0% 0.0% flops 347 x 160 x 1707 194094366720 0.0% 100.0% 0.0% flops 347 x 160 x 1711 194549186560 0.0% 100.0% 0.0% flops 369 x 160 x 1702 205795491840 0.0% 100.0% 0.0% flops 369 x 160 x 1707 206400061440 0.0% 100.0% 0.0% flops 369 x 160 x 1711 206883717120 0.0% 100.0% 0.0% flops 320 x 192 x 1702 214161162240 0.0% 100.0% 0.0% flops 320 x 192 x 1707 214790307840 0.0% 100.0% 0.0% flops 320 x 192 x 1711 215293624320 0.0% 100.0% 0.0% flops 342 x 192 x 1702 228884742144 0.0% 100.0% 0.0% flops 342 x 192 x 1707 229557141504 0.0% 100.0% 0.0% flops 342 x 192 x 1711 230095060992 0.0% 100.0% 0.0% flops 32 x 192 x 160 241985126400 0.0% 100.0% 0.0% flops 32 x 160 x 192 241985126400 0.0% 100.0% 0.0% flops 440 x 160 x 192 243518668800 0.0% 100.0% 0.0% flops 440 x 192 x 160 243518668800 0.0% 100.0% 0.0% flops 462 x 192 x 160 255694602240 0.0% 100.0% 0.0% flops 462 x 160 x 192 255694602240 0.0% 100.0% 0.0% flops 471 x 160 x 192 260675665920 0.0% 100.0% 0.0% flops 471 x 192 x 160 260675665920 0.0% 100.0% 0.0% flops 64 x 160 x 160 284688384000 0.0% 100.0% 0.0% flops 440 x 192 x 1702 294471598080 0.0% 100.0% 0.0% flops 320 x 160 x 160 295174144000 0.0% 100.0% 0.0% flops 440 x 192 x 1707 295336673280 0.0% 100.0% 0.0% flops 440 x 192 x 1711 296028733440 0.0% 100.0% 0.0% flops 462 x 192 x 1702 309195177984 0.0% 100.0% 0.0% flops 462 x 192 x 1707 310103506944 0.0% 100.0% 0.0% flops 462 x 192 x 1711 310830170112 0.0% 100.0% 0.0% flops 471 x 192 x 1702 315218460672 0.0% 100.0% 0.0% flops 342 x 160 x 160 315467366400 0.0% 100.0% 0.0% flops 471 x 192 x 1707 316144484352 0.0% 100.0% 0.0% flops 471 x 192 x 1711 316885303296 0.0% 100.0% 0.0% flops 320 x 160 x 1702 356935270400 0.0% 100.0% 0.0% flops 320 x 160 x 1707 357983846400 0.0% 100.0% 0.0% flops 320 x 160 x 1711 358822707200 0.0% 100.0% 0.0% flops 342 x 160 x 1702 381474570240 0.0% 100.0% 0.0% flops 342 x 160 x 1707 382595235840 0.0% 100.0% 0.0% flops 342 x 160 x 1711 383491768320 0.0% 100.0% 0.0% flops 32 x 160 x 160 403308544000 0.0% 100.0% 0.0% flops 440 x 160 x 160 405864448000 0.0% 100.0% 0.0% flops 462 x 160 x 160 426157670400 0.0% 100.0% 0.0% flops 471 x 160 x 160 434459443200 0.0% 100.0% 0.0% flops 64 x 192 x 1702 458773364736 0.0% 100.0% 0.0% flops 64 x 192 x 1707 460121112576 0.0% 100.0% 0.0% flops 64 x 192 x 1711 461199310848 0.0% 100.0% 0.0% flops 440 x 160 x 1702 490785996800 0.0% 100.0% 0.0% flops 440 x 160 x 1707 492227788800 0.0% 100.0% 0.0% flops 440 x 160 x 1711 493381222400 0.0% 100.0% 0.0% flops 462 x 160 x 1702 515325296640 0.0% 100.0% 0.0% flops 462 x 160 x 1707 516839178240 0.0% 100.0% 0.0% flops 462 x 160 x 1711 518050283520 0.0% 100.0% 0.0% flops 471 x 160 x 1702 525364101120 0.0% 100.0% 0.0% flops 471 x 160 x 1707 526907473920 0.0% 100.0% 0.0% flops 471 x 160 x 1711 528142172160 0.0% 100.0% 0.0% flops 9 x 9 x 192 532590992640 0.0% 100.0% 0.0% flops 32 x 192 x 1702 649928933376 0.0% 100.0% 0.0% flops 32 x 192 x 1707 651838242816 0.0% 100.0% 0.0% flops 32 x 192 x 1711 653365690368 0.0% 100.0% 0.0% flops 22 x 9 x 192 739521446400 0.0% 100.0% 0.0% flops 9 x 22 x 192 741663876096 0.0% 100.0% 0.0% flops 449 x 192 x 192 745499197440 0.0% 100.0% 0.0% flops 64 x 160 x 1702 764622274560 0.0% 100.0% 0.0% flops 64 x 160 x 1707 766868520960 0.0% 100.0% 0.0% flops 64 x 160 x 1711 768665518080 0.0% 100.0% 0.0% flops 9 x 9 x 160 887651654400 0.0% 100.0% 0.0% flops 22 x 22 x 192 1025357887488 0.0% 100.0% 0.0% flops 32 x 160 x 1702 1083214888960 0.0% 100.0% 0.0% flops 32 x 160 x 1707 1086397071360 0.0% 100.0% 0.0% flops 32 x 160 x 1711 1088942817280 0.0% 100.0% 0.0% flops 22 x 9 x 160 1232535744000 0.0% 100.0% 0.0% flops 9 x 22 x 160 1236106460160 0.0% 100.0% 0.0% flops 449 x 192 x 160 1242498662400 0.0% 100.0% 0.0% flops 449 x 160 x 192 1242498662400 0.0% 100.0% 0.0% flops 449 x 192 x 1702 1502474403840 0.0% 100.0% 0.0% flops 449 x 192 x 1707 1506888253440 0.0% 100.0% 0.0% flops 449 x 192 x 1711 1510419333120 0.0% 100.0% 0.0% flops 22 x 22 x 160 1708929812480 0.0% 100.0% 0.0% flops 449 x 160 x 160 2070831104000 0.0% 100.0% 0.0% flops 449 x 160 x 1702 2504124006400 0.0% 100.0% 0.0% flops 449 x 160 x 1707 2511480422400 0.0% 100.0% 0.0% flops 449 x 160 x 1711 2517365555200 0.0% 100.0% 0.0% flops 9 x 32 x 9 5962625519616 0.0% 100.0% 0.0% flops 9 x 32 x 22 8325932617728 0.0% 100.0% 0.0% flops 22 x 32 x 9 8325932617728 0.0% 100.0% 0.0% flops 22 x 32 x 22 11452938371072 0.0% 100.0% 0.0% flops inhomo. stacks 3874753609728 100.0% 0.0% 0.0% flops total 93.663610E+12 4.1% 95.9% 0.0% flops max/rank 3.319256E+12 11.2% 88.8% 0.0% matmuls inhomo. stacks 49752 100.0% 0.0% 0.0% matmuls total 2962408280 0.0% 100.0% 0.0% number of processed stacks 5617614 0.9% 99.1% 0.0% average stack size 1.0 532.0 0.0 marketing flops 144.582793E+12 ------------------------------------------------------------------------------- # multiplications 2507 max memory usage/rank 662.884352E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 902520 MPI messages size (bytes): total size 2.219095E+12 min size 0.000000E+00 max size 23.420168E+06 average size 2.458777E+06 MPI breakdown and total messages size (bytes): size <= 128 5610 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 330 14417920 131072 < size <= 4194304 820800 1016825118720 4194304 < size <= 16777216 52740 665379475120 16777216 < size 23040 536870912000 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 125 12. MP_Allreduce 13570 37. MP_Alltoall 9654 365242. MP_ISend 60124 1071152. MP_IRecv 60124 1064037. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4009 67418. MP_Allreduce 11082 1000. MP_Sync 87 MP_Alltoall 1712 8598479. MP_SendRecv 8960 54272. MP_ISendRecv 8960 54272. MP_Wait 20708 MP_ISend 9612 307844. MP_IRecv 9612 307844. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.435 0.502 277.770 277.785 qs_mol_dyn_low 1 2.0 0.153 0.603 276.127 276.147 qs_forces 11 3.9 0.094 0.380 275.357 275.722 qs_energies 11 4.9 0.005 0.023 265.339 265.735 scf_env_do_scf 11 5.9 0.001 0.004 243.936 243.955 scf_env_do_scf_inner_loop 117 6.6 0.004 0.012 210.526 210.539 velocity_verlet 10 3.0 0.022 0.086 175.493 175.507 dbcsr_multiply_generic 2507 12.6 0.241 0.297 134.201 137.036 qs_scf_new_mos 117 7.6 0.001 0.002 129.983 131.476 qs_scf_loop_do_ot 117 8.6 0.001 0.002 129.982 131.474 ot_scf_mini 117 9.6 0.004 0.006 120.848 122.056 multiply_cannon 2507 13.6 0.329 0.362 95.896 105.093 multiply_cannon_loop 2507 14.6 0.301 0.362 87.818 95.732 multiply_cannon_multrec 15042 15.6 73.734 83.412 73.769 83.446 ot_mini 117 10.6 0.001 0.002 69.434 70.763 rebuild_ks_matrix 128 8.3 0.001 0.001 57.228 58.984 qs_ks_build_kohn_sham_matrix 128 9.3 0.019 0.023 57.227 58.983 qs_ks_update_qs_env 128 7.6 0.001 0.001 51.818 53.495 mp_waitall_1 171544 16.6 22.761 39.855 22.761 39.855 qs_rho_update_rho_low 128 7.7 0.001 0.001 36.554 36.886 calculate_rho_elec 128 8.7 0.148 0.157 36.553 36.886 qs_ot_get_derivative 117 11.6 0.002 0.004 35.059 36.281 apply_preconditioner_dbcsr 128 12.6 0.001 0.001 32.614 35.199 apply_single 128 13.6 0.001 0.001 32.613 35.198 ot_diis_step 117 11.6 0.031 0.146 34.099 34.101 init_scf_loop 11 6.9 0.001 0.004 33.222 33.228 sum_up_and_integrate 128 10.3 0.004 0.007 30.794 30.873 integrate_v_rspace 128 11.3 0.004 0.006 30.689 30.766 qs_ot_get_p 128 10.4 0.001 0.002 26.243 28.017 make_m2s 5014 13.6 0.133 0.160 26.146 27.397 prepare_preconditioner 11 7.9 0.000 0.001 25.698 25.837 make_preconditioner 11 8.9 0.000 0.001 25.698 25.837 grid_collocate_task_list 128 9.7 24.273 24.831 24.273 24.831 make_full_inverse_cholesky 11 9.9 0.000 0.002 23.982 24.266 make_images 5014 14.6 2.193 2.827 22.501 23.873 grid_integrate_task_list 128 12.3 20.876 21.466 20.876 21.466 multiply_cannon_metrocomm3 15042 15.6 0.060 0.116 4.081 19.520 qs_ot_p2m_diag 83 11.4 0.216 0.279 17.384 17.499 cp_dbcsr_syevd 83 12.4 0.006 0.007 16.245 16.253 qs_ot_get_derivative_diag 77 12.4 0.003 0.004 15.235 16.138 make_images_data 5014 15.6 0.077 0.111 13.485 15.873 init_scf_run 11 5.9 0.001 0.006 15.187 15.188 scf_env_initial_rho_setup 11 6.9 0.001 0.003 15.186 15.188 hybrid_alltoall_any 5200 16.5 0.757 4.820 11.585 15.036 cp_fm_cholesky_invert 11 10.9 13.682 13.703 13.682 13.703 wfi_extrapolate 11 7.9 0.001 0.002 13.460 13.460 fft_wrap_pw1pw2 1291 11.7 0.026 0.038 12.150 12.385 cp_fm_diag_elpa 83 13.4 0.001 0.001 11.973 11.985 cp_fm_diag_elpa_base 83 14.4 11.802 11.847 11.963 11.963 mp_sum_l 12367 13.3 8.094 11.841 8.094 11.841 calculate_dm_sparse 128 9.5 0.001 0.002 10.149 11.227 fft_wrap_pw1pw2_140 523 12.2 0.560 0.617 10.661 10.928 mp_irecv_dv 36752 16.2 4.178 10.869 4.178 10.869 density_rs2pw 128 9.7 0.009 0.012 9.584 10.112 fft3d_ps 1291 13.7 4.441 4.825 9.697 9.873 multiply_cannon_metrocomm1 15042 15.6 0.077 0.132 5.137 9.546 multiply_cannon_metrocomm4 12535 15.6 0.063 0.113 3.845 9.544 mp_alltoall_d11v 2415 14.1 7.387 7.950 7.387 7.950 cp_dbcsr_sm_fm_multiply 37 9.5 0.003 0.004 7.547 7.565 cp_fm_cholesky_decompose 22 10.9 7.319 7.419 7.319 7.419 potential_pw2rs 128 12.3 0.026 0.031 7.282 7.349 mp_allgather_i34 2507 14.6 2.952 7.171 2.952 7.171 mp_sum_d 4472 12.1 4.716 7.151 4.716 7.151 dbcsr_dot_sd 1318 12.0 1.294 1.384 4.708 7.097 qs_ot_get_orbitals 117 10.6 0.001 0.001 6.462 6.648 qs_ot_get_derivative_taylor 40 13.0 0.001 0.002 6.039 6.366 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 6.157 6.337 make_images_sizes 5014 15.6 0.009 0.017 2.683 6.199 mp_alltoall_i44 5014 16.6 2.674 6.191 2.674 6.191 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.704 6.178 dbcsr_complete_redistribute 395 12.7 1.330 1.410 5.401 5.860 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="406", plot="h2o_256_md", label="(4n/9r/4t)", y=277.785000, yerr=0.000000 PlotPoint: name="407", plot="h2o_256_md_mem", label="(4n/9r/4t)", y=622.636364, yerr=9.966058 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/25/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 96 x 192 x 64 16911433728 0.0% 100.0% 0.0% flops 96 x 160 x 64 28185722880 0.0% 100.0% 0.0% flops 267 x 192 x 64 29291839488 0.0% 100.0% 0.0% flops 289 x 192 x 64 31705399296 0.0% 100.0% 0.0% flops 311 x 192 x 64 34118959104 0.0% 100.0% 0.0% flops 342 x 192 x 64 37519884288 0.0% 100.0% 0.0% flops 409 x 192 x 64 44870270976 0.0% 100.0% 0.0% flops 418 x 192 x 64 45857636352 0.0% 100.0% 0.0% flops 427 x 192 x 64 46845001728 0.0% 100.0% 0.0% flops 436 x 192 x 64 47832367104 0.0% 100.0% 0.0% flops 267 x 160 x 64 48819732480 0.0% 100.0% 0.0% flops 96 x 192 x 96 50734301184 0.0% 100.0% 0.0% flops 289 x 160 x 64 52842332160 0.0% 100.0% 0.0% flops 64 x 192 x 64 56371445760 0.0% 100.0% 0.0% flops 311 x 160 x 64 56864931840 0.0% 100.0% 0.0% flops 342 x 160 x 64 62533140480 0.0% 100.0% 0.0% flops 409 x 160 x 64 74783784960 0.0% 100.0% 0.0% flops 418 x 160 x 64 76429393920 0.0% 100.0% 0.0% flops 427 x 160 x 64 78075002880 0.0% 100.0% 0.0% flops 436 x 160 x 64 79720611840 0.0% 100.0% 0.0% flops 96 x 160 x 96 84557168640 0.0% 100.0% 0.0% flops 267 x 192 x 96 87875518464 0.0% 100.0% 0.0% flops 267 x 192 x 862 89793202176 0.0% 100.0% 0.0% flops 32 x 192 x 64 90194313216 0.0% 100.0% 0.0% flops 64 x 160 x 64 93952409600 0.0% 100.0% 0.0% flops 289 x 192 x 96 95116197888 0.0% 100.0% 0.0% flops 289 x 192 x 862 97191892992 0.0% 100.0% 0.0% flops 453 x 192 x 64 99394781184 0.0% 100.0% 0.0% flops 311 x 192 x 96 102356877312 0.0% 100.0% 0.0% flops 311 x 192 x 862 104590583808 0.0% 100.0% 0.0% flops 342 x 192 x 96 112559652864 0.0% 100.0% 0.0% flops 342 x 192 x 862 115016011776 0.0% 100.0% 0.0% flops 96 x 192 x 862 115159007232 0.0% 100.0% 0.0% flops 409 x 192 x 96 134610812928 0.0% 100.0% 0.0% flops 409 x 192 x 862 137548388352 0.0% 100.0% 0.0% flops 418 x 192 x 96 137572909056 0.0% 100.0% 0.0% flops 427 x 192 x 96 140535005184 0.0% 100.0% 0.0% flops 418 x 192 x 862 140575125504 0.0% 100.0% 0.0% flops 436 x 192 x 96 143497101312 0.0% 100.0% 0.0% flops 427 x 192 x 862 143601862656 0.0% 100.0% 0.0% flops 440 x 192 x 64 144813588480 0.0% 100.0% 0.0% flops 267 x 160 x 96 146459197440 0.0% 100.0% 0.0% flops 436 x 192 x 862 146628599808 0.0% 100.0% 0.0% flops 267 x 160 x 862 149655336960 0.0% 100.0% 0.0% flops 32 x 160 x 64 150323855360 0.0% 100.0% 0.0% flops 289 x 160 x 96 158526996480 0.0% 100.0% 0.0% flops 289 x 160 x 862 161986488320 0.0% 100.0% 0.0% flops 453 x 160 x 64 165657968640 0.0% 100.0% 0.0% flops 64 x 192 x 96 169114337280 0.0% 100.0% 0.0% flops 311 x 160 x 96 170594795520 0.0% 100.0% 0.0% flops 311 x 160 x 862 174317639680 0.0% 100.0% 0.0% flops 267 x 192 x 849 176878024704 0.0% 100.0% 0.0% flops 342 x 160 x 96 187599421440 0.0% 100.0% 0.0% flops 289 x 192 x 849 191452243968 0.0% 100.0% 0.0% flops 342 x 160 x 862 191693352960 0.0% 100.0% 0.0% flops 96 x 160 x 862 191931678720 0.0% 100.0% 0.0% flops 311 x 192 x 849 206026463232 0.0% 100.0% 0.0% flops 409 x 160 x 96 224351354880 0.0% 100.0% 0.0% flops 342 x 192 x 849 226562863104 0.0% 100.0% 0.0% flops 96 x 192 x 849 226844540928 0.0% 100.0% 0.0% flops 409 x 160 x 862 229247313920 0.0% 100.0% 0.0% flops 418 x 160 x 96 229288181760 0.0% 100.0% 0.0% flops 427 x 160 x 96 234225008640 0.0% 100.0% 0.0% flops 418 x 160 x 862 234291875840 0.0% 100.0% 0.0% flops 436 x 160 x 96 239161835520 0.0% 100.0% 0.0% flops 427 x 160 x 862 239336437760 0.0% 100.0% 0.0% flops 440 x 160 x 64 241355980800 0.0% 100.0% 0.0% flops 436 x 160 x 862 244380999680 0.0% 100.0% 0.0% flops 32 x 192 x 96 270582939648 0.0% 100.0% 0.0% flops 409 x 192 x 849 270947985408 0.0% 100.0% 0.0% flops 418 x 192 x 849 276910166016 0.0% 100.0% 0.0% flops 64 x 160 x 96 281857228800 0.0% 100.0% 0.0% flops 427 x 192 x 849 282872346624 0.0% 100.0% 0.0% flops 436 x 192 x 849 288834527232 0.0% 100.0% 0.0% flops 267 x 160 x 849 294796707840 0.0% 100.0% 0.0% flops 453 x 192 x 96 298184343552 0.0% 100.0% 0.0% flops 453 x 192 x 862 304691539968 0.0% 100.0% 0.0% flops 289 x 160 x 849 319087073280 0.0% 100.0% 0.0% flops 311 x 160 x 849 343377438720 0.0% 100.0% 0.0% flops 9 x 9 x 64 352505737728 0.0% 100.0% 0.0% flops 342 x 160 x 849 377604771840 0.0% 100.0% 0.0% flops 96 x 160 x 849 378074234880 0.0% 100.0% 0.0% flops 64 x 192 x 862 383863357440 0.0% 100.0% 0.0% flops 440 x 192 x 96 434440765440 0.0% 100.0% 0.0% flops 440 x 192 x 862 443921448960 0.0% 100.0% 0.0% flops 32 x 160 x 96 450971566080 0.0% 100.0% 0.0% flops 409 x 160 x 849 451579975680 0.0% 100.0% 0.0% flops 418 x 160 x 849 461516943360 0.0% 100.0% 0.0% flops 427 x 160 x 849 471453911040 0.0% 100.0% 0.0% flops 436 x 160 x 849 481390878720 0.0% 100.0% 0.0% flops 22 x 9 x 64 489467860992 0.0% 100.0% 0.0% flops 9 x 22 x 64 490886212608 0.0% 100.0% 0.0% flops 453 x 160 x 96 496973905920 0.0% 100.0% 0.0% flops 453 x 160 x 862 507819233280 0.0% 100.0% 0.0% flops 453 x 192 x 849 600192847872 0.0% 100.0% 0.0% flops 32 x 192 x 862 614181371904 0.0% 100.0% 0.0% flops 64 x 160 x 862 639772262400 0.0% 100.0% 0.0% flops 22 x 22 x 64 678653927424 0.0% 100.0% 0.0% flops 440 x 160 x 96 724067942400 0.0% 100.0% 0.0% flops 440 x 160 x 862 739869081600 0.0% 100.0% 0.0% flops 64 x 192 x 849 756148469760 0.0% 100.0% 0.0% flops 440 x 192 x 849 874453155840 0.0% 100.0% 0.0% flops 453 x 160 x 849 1000321413120 0.0% 100.0% 0.0% flops 32 x 160 x 862 1023635619840 0.0% 100.0% 0.0% flops 9 x 9 x 96 1057517213184 0.0% 100.0% 0.0% flops 32 x 192 x 849 1209837551616 0.0% 100.0% 0.0% flops 64 x 160 x 849 1260247449600 0.0% 100.0% 0.0% flops 440 x 160 x 849 1457421926400 0.0% 100.0% 0.0% flops 22 x 9 x 96 1468403582976 0.0% 100.0% 0.0% flops 9 x 22 x 96 1472658637824 0.0% 100.0% 0.0% flops 32 x 160 x 849 2016395919360 0.0% 100.0% 0.0% flops 22 x 22 x 96 2035961782272 0.0% 100.0% 0.0% flops 9 x 32 x 9 5921915277312 0.0% 100.0% 0.0% flops 22 x 32 x 9 8269109747712 0.0% 100.0% 0.0% flops 9 x 32 x 22 8269109747712 0.0% 100.0% 0.0% flops 22 x 32 x 22 11374757920768 0.0% 100.0% 0.0% flops inhomo. stacks 22196602552320 100.0% 0.0% 0.0% flops total 94.291949E+12 23.5% 76.5% 0.0% flops max/rank 4.991778E+12 27.1% 72.9% 0.0% matmuls inhomo. stacks 639648 100.0% 0.0% 0.0% matmuls total 3069251512 0.0% 100.0% 0.0% number of processed stacks 8216872 7.8% 92.2% 0.0% average stack size 1.0 405.0 0.0 marketing flops 143.508480E+12 ------------------------------------------------------------------------------- # multiplications 2485 max memory usage/rank 844.009472E+06 # 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 660837747920 16777216 < size 30480 532676608000 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 3958 68502. MP_Allreduce 10955 1087. MP_Sync 86 MP_Alltoall 1700 12496378. 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 0.114 0.146 318.093 318.747 qs_mol_dyn_low 1 2.0 0.109 0.149 317.169 317.865 qs_forces 11 3.9 0.012 0.026 316.340 316.980 qs_energies 11 4.9 0.143 0.294 306.265 306.892 scf_env_do_scf 11 5.9 0.003 0.010 281.177 281.777 scf_env_do_scf_inner_loop 116 6.6 0.011 0.033 224.125 224.592 velocity_verlet 10 3.0 0.005 0.010 205.243 205.662 dbcsr_multiply_generic 2485 12.5 0.278 0.323 155.571 158.534 qs_scf_new_mos 116 7.6 0.007 0.026 141.182 142.123 qs_scf_loop_do_ot 116 8.6 0.001 0.003 141.175 142.122 ot_scf_mini 116 9.6 0.008 0.016 131.805 132.818 multiply_cannon 2485 13.5 0.349 0.390 102.194 113.847 multiply_cannon_loop 2485 14.5 0.524 0.628 95.340 105.658 multiply_cannon_multrec 29820 15.5 85.028 97.463 85.069 97.505 ot_mini 116 10.6 0.122 0.244 79.748 80.894 rebuild_ks_matrix 127 8.3 0.001 0.001 60.854 62.882 qs_ks_build_kohn_sham_matrix 127 9.3 0.021 0.028 60.854 62.881 qs_ks_update_qs_env 127 7.6 0.001 0.001 55.211 57.048 init_scf_loop 11 6.9 0.001 0.004 56.864 56.975 prepare_preconditioner 11 7.9 0.000 0.001 48.501 48.629 make_preconditioner 11 8.9 0.000 0.001 48.501 48.629 make_full_inverse_cholesky 11 9.9 0.023 0.033 40.047 46.721 apply_preconditioner_dbcsr 127 12.6 0.001 0.001 43.526 45.254 apply_single 127 13.6 0.001 0.001 43.525 45.253 ot_diis_step 116 11.6 0.039 0.053 43.386 43.505 make_m2s 4970 13.5 0.163 0.200 37.822 39.196 qs_ot_get_derivative 116 11.6 0.002 0.003 35.995 37.059 qs_rho_update_rho_low 127 7.7 0.001 0.001 36.096 36.214 calculate_rho_elec 127 8.7 0.211 0.229 36.095 36.212 make_images 4970 14.5 3.132 3.729 31.193 32.126 sum_up_and_integrate 127 10.3 0.003 0.006 30.120 30.177 integrate_v_rspace 127 11.3 0.004 0.005 30.011 30.072 mp_waitall_1 146592 16.7 20.361 27.625 20.361 27.625 grid_collocate_task_list 127 9.7 24.836 25.277 24.836 25.277 qs_ot_get_p 127 10.4 0.001 0.002 23.059 24.271 cp_fm_upper_to_full 104 14.8 15.463 22.336 15.463 22.336 grid_integrate_task_list 127 12.3 21.300 21.855 21.300 21.855 make_images_data 4970 15.5 0.090 0.142 19.281 20.921 hybrid_alltoall_any 5155 16.4 0.685 2.441 17.825 19.726 init_scf_run 11 5.9 0.000 0.001 18.088 18.126 scf_env_initial_rho_setup 11 6.9 0.001 0.003 18.088 18.126 dbcsr_complete_redistribute 393 12.7 1.836 1.981 12.385 17.524 qs_ot_get_derivative_diag 76 12.4 0.003 0.004 16.033 17.001 multiply_cannon_metrocomm4 24850 15.5 0.124 0.234 6.034 15.944 mp_irecv_dv 75445 16.2 5.891 15.627 5.891 15.627 copy_fm_to_dbcsr 208 11.6 0.002 0.003 10.332 15.455 cp_fm_cholesky_invert 11 10.9 15.064 15.086 15.064 15.086 wfi_extrapolate 11 7.9 0.002 0.003 15.023 15.055 qs_ot_p2m_diag 82 11.4 0.304 0.346 13.706 13.779 mp_alltoall_i22 712 14.1 8.112 13.599 8.112 13.599 transfer_fm_to_dbcsr 11 9.9 0.033 0.039 8.416 13.480 cp_dbcsr_syevd 82 12.4 0.006 0.007 12.468 12.495 fft_wrap_pw1pw2 1281 11.7 0.023 0.029 11.955 11.995 mp_sum_l 12261 13.2 8.421 11.742 8.421 11.742 calculate_dm_sparse 127 9.5 0.001 0.002 10.460 10.910 fft_wrap_pw1pw2_140 519 12.2 0.587 0.605 10.781 10.834 cp_dbcsr_sm_fm_multiply 37 9.5 0.003 0.003 9.281 9.404 fft3d_ps 1281 13.7 4.535 4.630 9.336 9.381 density_rs2pw 127 9.7 0.009 0.013 8.563 9.082 cp_fm_diag_elpa 82 13.4 0.001 0.001 8.921 8.937 cp_fm_diag_elpa_base 82 14.4 8.075 8.357 8.911 8.930 multiply_cannon_metrocomm3 29820 15.5 0.082 0.140 3.074 8.097 mp_alltoall_d11v 2401 14.1 7.533 8.033 7.533 8.033 cp_fm_cholesky_decompose 22 10.9 7.601 7.670 7.601 7.670 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 7.006 7.637 dbcsr_make_dense_low 12894 15.7 0.135 0.188 6.692 7.222 qs_ot_get_orbitals 116 10.6 0.001 0.001 6.903 7.065 make_dense_data 12894 16.7 6.146 6.511 6.495 6.920 dbcsr_make_images_dense 4344 14.8 0.094 0.129 6.200 6.795 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 6.470 6.635 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="408", plot="h2o_256_md", label="(4n/6r/6t)", y=318.747000, yerr=0.000000 PlotPoint: name="409", plot="h2o_256_md_mem", label="(4n/6r/6t)", y=745.636364, yerr=14.008262 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/26/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 160 x 256 x 256 190589173760 0.0% 100.0% 0.0% flops 182 x 256 x 256 216795185152 0.0% 100.0% 0.0% flops 267 x 256 x 256 318045683712 0.0% 100.0% 0.0% flops 169 x 256 x 256 402619629568 0.0% 100.0% 0.0% flops 160 x 256 x 2560 432852172800 0.0% 100.0% 0.0% flops 182 x 256 x 2560 492369346560 0.0% 100.0% 0.0% flops 64 x 256 x 256 612032839680 0.0% 100.0% 0.0% flops 302 x 256 x 256 719474130944 0.0% 100.0% 0.0% flops 267 x 256 x 2560 722322063360 0.0% 100.0% 0.0% flops 324 x 256 x 256 771886153728 0.0% 100.0% 0.0% flops 169 x 256 x 2560 914400215040 0.0% 100.0% 0.0% flops 32 x 256 x 256 1346472247296 0.0% 100.0% 0.0% flops 9 x 9 x 256 1430460020736 0.0% 100.0% 0.0% flops 302 x 256 x 2560 1634016952320 0.0% 100.0% 0.0% flops 324 x 256 x 2560 1753051299840 0.0% 100.0% 0.0% flops 22 x 9 x 256 1986244964352 0.0% 100.0% 0.0% flops 9 x 22 x 256 1992000282624 0.0% 100.0% 0.0% flops 311 x 256 x 256 2222746238976 0.0% 100.0% 0.0% flops 22 x 22 x 256 2753956716544 0.0% 100.0% 0.0% flops 64 x 256 x 2560 3093718630400 0.0% 100.0% 0.0% flops 289 x 256 x 256 3442516951040 0.0% 100.0% 0.0% flops 311 x 256 x 2560 5048138465280 0.0% 100.0% 0.0% flops 9 x 32 x 9 6003325476864 0.0% 100.0% 0.0% flops 32 x 256 x 2560 6806180986880 0.0% 100.0% 0.0% flops 289 x 256 x 2560 7818392371200 0.0% 100.0% 0.0% flops 9 x 32 x 22 8382768058368 0.0% 100.0% 0.0% flops 22 x 32 x 9 8382768058368 0.0% 100.0% 0.0% flops 22 x 32 x 22 11531106926592 0.0% 100.0% 0.0% flops inhomo. stacks 13914431553536 100.0% 0.0% 0.0% flops total 95.343065E+12 14.6% 85.4% 0.0% flops max/rank 7.230009E+12 22.4% 77.6% 0.0% matmuls inhomo. stacks 122672 100.0% 0.0% 0.0% matmuls total 2939459052 0.0% 100.0% 0.0% number of processed stacks 4567844 2.7% 97.3% 0.0% average stack size 1.0 661.2 0.0 marketing flops 145.646636E+12 ------------------------------------------------------------------------------- # multiplications 2527 max memory usage/rank 1.195626E+09 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 242592 MPI messages size (bytes): total size 1.341705E+12 min size 0.000000E+00 max size 52.428800E+06 average size 5.530704E+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 114816 60196651008 4194304 < size <= 16777216 105840 554906419200 16777216 < size 20352 726591885696 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 31 12. MP_Allreduce 13490 37. MP_Alltoall 9708 751306. MP_ISend 40388 2095057. MP_IRecv 40388 2094161. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4031 67282. MP_Allreduce 11137 1164. MP_Sync 87 MP_Alltoall 1724 18848449. MP_SendRecv 3870 122880. MP_ISendRecv 3870 122880. MP_Wait 16244 MP_ISend 10760 423501. MP_IRecv 10760 423501. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.126 0.301 312.311 312.334 qs_mol_dyn_low 1 2.0 0.009 0.025 311.268 311.287 qs_forces 11 3.9 0.005 0.010 310.573 310.611 qs_energies 11 4.9 0.003 0.008 300.604 300.652 scf_env_do_scf 11 5.9 0.003 0.011 274.251 274.293 scf_env_do_scf_inner_loop 118 6.6 0.010 0.030 232.130 232.146 velocity_verlet 10 3.0 0.001 0.002 199.072 199.106 dbcsr_multiply_generic 2527 12.6 0.268 0.284 162.421 166.952 qs_scf_new_mos 118 7.6 0.002 0.003 147.042 148.655 qs_scf_loop_do_ot 118 8.6 0.002 0.002 147.040 148.654 ot_scf_mini 118 9.6 0.009 0.022 137.104 138.681 multiply_cannon 2527 13.6 0.352 0.397 111.542 120.971 multiply_cannon_loop 2527 14.6 0.339 0.450 101.731 109.058 multiply_cannon_multrec 10108 15.6 83.984 96.442 84.024 96.483 ot_mini 118 10.6 0.005 0.017 82.552 84.176 rebuild_ks_matrix 129 8.3 0.001 0.001 62.575 64.701 qs_ks_build_kohn_sham_matrix 129 9.3 0.021 0.023 62.574 64.700 qs_ks_update_qs_env 129 7.6 0.001 0.001 57.244 59.178 apply_preconditioner_dbcsr 129 12.6 0.000 0.001 46.942 49.360 apply_single 129 13.6 0.001 0.001 46.941 49.359 ot_diis_step 118 11.6 0.032 0.039 47.517 47.527 make_m2s 5054 13.6 0.127 0.135 37.108 42.570 init_scf_loop 11 6.9 0.001 0.005 41.902 41.911 mp_waitall_1 126780 16.7 31.111 40.750 31.111 40.750 qs_rho_update_rho_low 129 7.7 0.003 0.010 36.546 36.691 calculate_rho_elec 129 8.7 0.318 0.330 36.543 36.691 qs_ot_get_derivative 118 11.6 0.002 0.003 34.927 36.497 make_images 5054 14.6 3.448 4.209 30.046 34.607 prepare_preconditioner 11 7.9 0.000 0.001 33.185 33.340 make_preconditioner 11 8.9 0.001 0.013 33.185 33.340 make_full_inverse_cholesky 11 9.9 0.018 0.033 30.447 30.876 sum_up_and_integrate 129 10.3 0.002 0.003 29.502 29.549 integrate_v_rspace 129 11.3 0.004 0.004 29.401 29.449 qs_ot_get_p 129 10.4 0.001 0.001 24.103 26.056 make_images_data 5054 15.6 0.071 0.089 19.726 25.455 grid_collocate_task_list 129 9.7 23.561 24.888 23.561 24.888 hybrid_alltoall_any 5240 16.5 1.206 5.401 18.565 22.783 grid_integrate_task_list 129 12.3 20.076 20.703 20.076 20.703 mp_allgather_i34 2527 14.6 6.735 20.658 6.735 20.658 multiply_cannon_metrocomm1 10108 15.6 0.045 0.048 9.997 20.602 init_scf_run 11 5.9 0.000 0.002 18.753 18.755 scf_env_initial_rho_setup 11 6.9 0.001 0.008 18.753 18.755 cp_fm_cholesky_invert 11 10.9 17.313 17.329 17.313 17.329 qs_ot_get_derivative_diag 77 12.4 0.003 0.003 15.449 16.606 wfi_extrapolate 11 7.9 0.001 0.001 15.639 15.640 qs_ot_p2m_diag 83 11.4 0.429 0.438 14.814 14.829 fft_wrap_pw1pw2 1301 11.7 0.021 0.025 13.436 13.740 cp_dbcsr_syevd 83 12.4 0.006 0.007 13.556 13.559 density_rs2pw 129 9.7 0.007 0.008 10.112 13.152 fft_wrap_pw1pw2_140 527 12.2 0.711 0.767 11.942 12.385 mp_sum_l 12463 13.3 6.815 11.649 6.815 11.649 calculate_dm_sparse 129 9.5 0.002 0.002 11.099 11.590 fft3d_ps 1301 13.7 4.934 5.073 10.478 10.582 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 10.169 10.307 cp_fm_diag_elpa 83 13.4 0.000 0.001 10.081 10.083 cp_fm_diag_elpa_base 83 14.4 9.721 9.844 10.075 10.075 multiply_cannon_metrocomm4 7581 15.6 0.033 0.037 4.574 10.062 mp_irecv_dv 29086 15.9 4.560 10.023 4.560 10.023 mp_alltoall_d11v 2423 14.1 8.274 9.408 8.274 9.408 dbcsr_make_dense_low 6490 15.6 0.075 0.079 7.542 9.370 make_dense_data 6490 16.6 7.234 8.972 7.440 9.264 dbcsr_make_images_dense 4420 14.8 0.046 0.048 6.696 8.570 multiply_cannon_metrocomm3 10108 15.6 0.035 0.037 2.606 8.528 cp_fm_cholesky_decompose 22 10.9 8.169 8.394 8.169 8.394 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 7.720 8.343 dbcsr_complete_redistribute 395 12.7 2.616 2.742 7.738 8.227 dbcsr_data_release 190776 16.1 4.483 8.055 4.483 8.055 dbcsr_dot_sd 1330 12.0 2.323 2.352 5.058 7.420 qs_ot_get_orbitals 118 10.6 0.001 0.001 6.654 6.719 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 6.470 6.660 dbcsr_destroy 27283 14.2 0.077 0.094 3.405 6.273 copy_dbcsr_to_fm 186 11.8 0.004 0.004 5.757 6.268 potential_pw2rs 129 12.3 0.035 0.037 6.239 6.250 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="410", plot="h2o_256_md", label="(4n/4r/9t)", y=312.334000, yerr=0.000000 PlotPoint: name="411", plot="h2o_256_md_mem", label="(4n/4r/9t)", y=1062.272727, yerr=8.046353 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/27/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 32 x 192 x 192 121550929920 0.0% 100.0% 0.0% flops 462 x 192 x 192 154779254784 0.0% 100.0% 0.0% flops 462 x 192 x 1702 155805382656 0.0% 100.0% 0.0% flops 64 x 192 x 192 156279767040 0.0% 100.0% 0.0% flops 32 x 192 x 160 202584883200 0.0% 100.0% 0.0% flops 32 x 160 x 192 202584883200 0.0% 100.0% 0.0% flops 440 x 192 x 192 221113221120 0.0% 100.0% 0.0% flops 440 x 192 x 1702 222579118080 0.0% 100.0% 0.0% flops 462 x 160 x 192 257965424640 0.0% 100.0% 0.0% flops 462 x 192 x 160 257965424640 0.0% 100.0% 0.0% flops 462 x 160 x 1702 259675637760 0.0% 100.0% 0.0% flops 64 x 192 x 160 260466278400 0.0% 100.0% 0.0% flops 64 x 160 x 192 260466278400 0.0% 100.0% 0.0% flops 32 x 192 x 1702 269960183808 0.0% 100.0% 0.0% flops 453 x 192 x 192 303528148992 0.0% 100.0% 0.0% flops 453 x 192 x 1702 305540425728 0.0% 100.0% 0.0% flops 462 x 192 x 1698 310878425088 0.0% 100.0% 0.0% flops 462 x 192 x 1711 313258530816 0.0% 100.0% 0.0% flops 32 x 160 x 160 337641472000 0.0% 100.0% 0.0% flops 64 x 192 x 1702 347091664896 0.0% 100.0% 0.0% flops 440 x 192 x 160 368522035200 0.0% 100.0% 0.0% flops 440 x 160 x 192 368522035200 0.0% 100.0% 0.0% flops 440 x 160 x 1702 370965196800 0.0% 100.0% 0.0% flops 449 x 192 x 192 376059985920 0.0% 100.0% 0.0% flops 449 x 192 x 1702 378553121280 0.0% 100.0% 0.0% flops 462 x 160 x 160 429942374400 0.0% 100.0% 0.0% flops 64 x 160 x 160 434110464000 0.0% 100.0% 0.0% flops 440 x 192 x 1698 444112035840 0.0% 100.0% 0.0% flops 440 x 192 x 1711 447512186880 0.0% 100.0% 0.0% flops 32 x 160 x 1702 449933639680 0.0% 100.0% 0.0% flops 453 x 192 x 160 505880248320 0.0% 100.0% 0.0% flops 453 x 160 x 192 505880248320 0.0% 100.0% 0.0% flops 453 x 160 x 1702 509234042880 0.0% 100.0% 0.0% flops 462 x 160 x 1698 518130708480 0.0% 100.0% 0.0% flops 462 x 160 x 1711 522097551360 0.0% 100.0% 0.0% flops 9 x 9 x 192 536420454912 0.0% 100.0% 0.0% flops 32 x 192 x 1698 538651459584 0.0% 100.0% 0.0% flops 32 x 192 x 1711 542775410688 0.0% 100.0% 0.0% flops 64 x 160 x 1702 578486108160 0.0% 100.0% 0.0% flops 453 x 192 x 1698 609644703744 0.0% 100.0% 0.0% flops 440 x 160 x 160 614203392000 0.0% 100.0% 0.0% flops 453 x 192 x 1711 614312183808 0.0% 100.0% 0.0% flops 449 x 192 x 160 626766643200 0.0% 100.0% 0.0% flops 449 x 160 x 192 626766643200 0.0% 100.0% 0.0% flops 449 x 160 x 1702 630921868800 0.0% 100.0% 0.0% flops 64 x 192 x 1698 692551876608 0.0% 100.0% 0.0% flops 64 x 192 x 1711 697854099456 0.0% 100.0% 0.0% flops 440 x 160 x 1698 740186726400 0.0% 100.0% 0.0% flops 22 x 9 x 192 744845967360 0.0% 100.0% 0.0% flops 440 x 160 x 1711 745853644800 0.0% 100.0% 0.0% flops 9 x 22 x 192 747002539008 0.0% 100.0% 0.0% flops 449 x 192 x 1698 755326909440 0.0% 100.0% 0.0% flops 449 x 192 x 1711 761109742080 0.0% 100.0% 0.0% flops 453 x 160 x 160 843133747200 0.0% 100.0% 0.0% flops 9 x 9 x 160 894034091520 0.0% 100.0% 0.0% flops 32 x 160 x 1698 897752432640 0.0% 100.0% 0.0% flops 32 x 160 x 1711 904625684480 0.0% 100.0% 0.0% flops 453 x 160 x 1698 1016074506240 0.0% 100.0% 0.0% flops 453 x 160 x 1711 1023853639680 0.0% 100.0% 0.0% flops 22 x 22 x 192 1032734512128 0.0% 100.0% 0.0% flops 449 x 160 x 160 1044611072000 0.0% 100.0% 0.0% flops 64 x 160 x 1698 1154253127680 0.0% 100.0% 0.0% flops 64 x 160 x 1711 1163090165760 0.0% 100.0% 0.0% flops 22 x 9 x 160 1241409945600 0.0% 100.0% 0.0% flops 9 x 22 x 160 1245004231680 0.0% 100.0% 0.0% flops 449 x 160 x 1698 1258878182400 0.0% 100.0% 0.0% flops 449 x 160 x 1711 1268516236800 0.0% 100.0% 0.0% flops 22 x 22 x 160 1721224186880 0.0% 100.0% 0.0% flops 9 x 32 x 9 6003301257216 0.0% 100.0% 0.0% flops 22 x 32 x 9 8382804148224 0.0% 100.0% 0.0% flops 9 x 32 x 22 8382804148224 0.0% 100.0% 0.0% flops 22 x 32 x 22 11531114856448 0.0% 100.0% 0.0% flops inhomo. stacks 24540172943360 100.0% 0.0% 0.0% flops total 98.911627E+12 24.8% 75.2% 0.0% flops max/rank 9.800266E+12 26.8% 73.2% 0.0% matmuls inhomo. stacks 325374 100.0% 0.0% 0.0% matmuls total 2982635438 0.0% 100.0% 0.0% number of processed stacks 5605318 5.8% 94.2% 0.0% average stack size 1.0 564.8 0.0 marketing flops 145.664755E+12 ------------------------------------------------------------------------------- # multiplications 2535 max memory usage/rank 1.525096E+09 # max total images/rank 3 # max 3D layers 1 # MPI messages exchanged 243360 MPI messages size (bytes): total size 1.842975E+12 min size 0.000000E+00 max size 70.506240E+06 average size 7.573040E+06 MPI breakdown and total messages size (bytes): size <= 128 1386 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 66 2883584 131072 < size <= 4194304 145668 137053077504 4194304 < size <= 16777216 70800 494927872000 16777216 < size 25440 1210987567760 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4048 67936. MP_Allreduce 11179 1416. MP_Sync 88 MP_Alltoall 1724 25011495. MP_SendRecv 2838 150016. MP_ISendRecv 2838 150016. MP_Wait 13060 MP_ISend 8608 623442. MP_IRecv 8608 623442. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.072 0.141 420.958 420.971 qs_mol_dyn_low 1 2.0 0.011 0.034 420.039 420.080 qs_forces 11 3.9 0.073 0.292 419.506 419.520 qs_energies 11 4.9 0.003 0.007 407.313 407.375 scf_env_do_scf 11 5.9 0.001 0.004 376.069 376.113 scf_env_do_scf_inner_loop 118 6.6 0.011 0.028 294.266 294.273 velocity_verlet 10 3.0 0.005 0.017 277.144 277.157 dbcsr_multiply_generic 2535 12.6 0.314 0.354 216.858 220.664 qs_scf_new_mos 118 7.6 0.010 0.018 194.544 196.581 qs_scf_loop_do_ot 118 8.6 0.001 0.002 194.535 196.564 ot_scf_mini 118 9.6 0.004 0.006 184.498 185.917 multiply_cannon 2535 13.6 0.387 0.416 154.694 162.209 multiply_cannon_loop 2535 14.6 0.371 0.440 144.188 151.425 ot_mini 118 10.6 0.001 0.002 111.648 113.310 multiply_cannon_multrec 15210 15.6 85.185 108.233 85.238 108.294 mp_waitall_1 126132 16.7 64.120 101.653 64.120 101.653 init_scf_loop 11 6.9 0.016 0.073 81.510 81.526 multiply_cannon_metrocomm3 15210 15.6 0.068 0.119 42.470 81.330 rebuild_ks_matrix 129 8.3 0.001 0.001 74.530 76.555 qs_ks_build_kohn_sham_matrix 129 9.3 0.020 0.021 74.529 76.554 apply_preconditioner_dbcsr 129 12.6 0.001 0.001 68.628 71.716 apply_single 129 13.6 0.001 0.001 68.627 71.715 prepare_preconditioner 11 7.9 0.000 0.000 70.582 70.722 make_preconditioner 11 8.9 0.001 0.002 70.582 70.722 qs_ks_update_qs_env 129 7.6 0.001 0.001 68.322 70.152 ot_diis_step 118 11.6 0.032 0.035 68.341 68.345 make_full_inverse_cholesky 11 9.9 0.038 0.045 58.363 67.770 make_m2s 5070 13.6 0.151 0.185 46.086 47.820 qs_ot_get_derivative 118 11.6 0.002 0.003 43.007 44.451 qs_rho_update_rho_low 129 7.7 0.001 0.001 41.423 41.737 calculate_rho_elec 129 8.7 0.414 0.426 41.422 41.737 make_images 5070 14.6 4.529 5.143 35.762 37.550 multiply_cannon_metrocomm4 12675 15.6 0.073 0.124 14.229 36.665 mp_irecv_dv 35523 16.2 14.112 36.504 14.112 36.504 cp_fm_upper_to_full 106 14.8 25.550 35.567 25.550 35.567 qs_ot_get_p 129 10.4 0.002 0.002 30.218 32.306 sum_up_and_integrate 129 10.3 0.003 0.005 31.261 31.389 integrate_v_rspace 129 11.3 0.004 0.005 31.152 31.285 grid_collocate_task_list 129 9.7 28.858 29.301 28.858 29.301 hybrid_alltoall_any 5257 16.5 1.358 3.762 22.835 26.177 make_images_data 5070 15.6 0.083 0.131 23.555 25.960 dbcsr_complete_redistribute 397 12.7 3.245 3.479 18.417 25.007 init_scf_run 11 5.9 0.000 0.003 22.449 22.460 scf_env_initial_rho_setup 11 6.9 0.001 0.003 22.449 22.460 grid_integrate_task_list 129 12.3 22.166 22.378 22.166 22.378 cp_fm_cholesky_invert 11 10.9 21.396 21.429 21.396 21.429 copy_fm_to_dbcsr 210 11.7 0.002 0.003 14.846 21.264 qs_ot_p2m_diag 84 11.4 0.541 0.601 19.703 19.797 qs_ot_get_derivative_diag 78 12.4 0.003 0.004 17.994 19.535 wfi_extrapolate 11 7.9 0.001 0.001 19.297 19.298 transfer_fm_to_dbcsr 11 9.9 0.038 0.041 12.175 18.493 cp_dbcsr_syevd 84 12.4 0.007 0.007 18.126 18.131 mp_alltoall_i22 720 14.1 10.814 17.713 10.814 17.713 mp_sum_l 12503 13.3 9.616 14.642 9.616 14.642 cp_fm_diag_elpa 84 13.4 0.001 0.001 14.484 14.503 cp_fm_diag_elpa_base 84 14.4 12.424 12.975 14.468 14.468 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 13.249 13.468 fft_wrap_pw1pw2 1301 11.7 0.023 0.030 13.289 13.453 calculate_dm_sparse 129 9.5 0.001 0.002 11.338 12.104 fft_wrap_pw1pw2_140 527 12.2 0.829 0.843 11.914 12.080 dbcsr_make_dense_low 10950 15.7 0.117 0.166 10.637 11.332 make_dense_data 10950 16.7 10.103 10.666 10.471 11.066 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 10.345 10.855 fft3d_ps 1301 13.7 5.217 5.701 10.219 10.471 dbcsr_dot_sd 1331 12.0 2.926 3.130 6.623 10.466 dbcsr_make_images_dense 4436 14.8 0.073 0.105 9.877 10.346 density_rs2pw 129 9.7 0.008 0.011 9.350 10.273 mp_alltoall_d11v 2429 14.1 9.106 9.586 9.106 9.586 qs_ot_get_orbitals 118 10.6 0.001 0.001 8.975 9.234 cp_fm_cholesky_decompose 22 10.9 8.640 8.801 8.640 8.801 mp_sum_d 4510 12.1 4.470 8.420 4.470 8.420 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="412", plot="h2o_256_md", label="(4n/3r/12t)", y=420.971000, yerr=0.000000 PlotPoint: name="413", plot="h2o_256_md_mem", label="(4n/3r/12t)", y=1306.545455, yerr=29.724075 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/28/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 258 x 256 x 256 603828781056 0.0% 100.0% 0.0% flops 32 x 256 x 256 959925190656 0.0% 100.0% 0.0% flops 64 x 256 x 256 959925190656 0.0% 100.0% 0.0% flops 280 x 256 x 256 982977085440 0.0% 100.0% 0.0% flops 258 x 256 x 2560 1374305648640 0.0% 100.0% 0.0% flops 9 x 9 x 256 1410026932224 0.0% 100.0% 0.0% flops 22 x 9 x 256 1957874282496 0.0% 100.0% 0.0% flops 9 x 22 x 256 1963544850432 0.0% 100.0% 0.0% flops 280 x 256 x 2560 2237241753600 0.0% 100.0% 0.0% flops 289 x 256 x 256 2367336480768 0.0% 100.0% 0.0% flops 22 x 22 x 256 2714615709696 0.0% 100.0% 0.0% flops 311 x 256 x 256 3275420073984 0.0% 100.0% 0.0% flops 32 x 256 x 2560 4864050462720 0.0% 100.0% 0.0% flops 64 x 256 x 2560 4864050462720 0.0% 100.0% 0.0% flops 289 x 256 x 2560 5388023889920 0.0% 100.0% 0.0% flops 9 x 32 x 9 5921932197888 0.0% 100.0% 0.0% flops 311 x 256 x 2560 7454809128960 0.0% 100.0% 0.0% flops 22 x 32 x 9 8269116235776 0.0% 100.0% 0.0% flops 9 x 32 x 22 8269116235776 0.0% 100.0% 0.0% flops 22 x 32 x 22 11374757920768 0.0% 100.0% 0.0% flops inhomo. stacks 18907284897792 100.0% 0.0% 0.0% flops total 96.127545E+12 19.7% 80.3% 0.0% flops max/rank 12.381166E+12 20.6% 79.4% 0.0% matmuls inhomo. stacks 164400 100.0% 0.0% 0.0% matmuls total 2899478492 0.0% 100.0% 0.0% number of processed stacks 4421644 3.7% 96.3% 0.0% average stack size 1.0 681.0 0.0 marketing flops 143.503447E+12 ------------------------------------------------------------------------------- # multiplications 2483 max memory usage/rank 4.084429E+09 # max total images/rank 2 # max 3D layers 1 # MPI messages exchanged 99320 MPI messages size (bytes): total size 1.127355E+12 min size 0.000000E+00 max size 104.857600E+06 average size 11.350736E+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 44688 34678505472 4194304 < size <= 16777216 43984 376564613120 16777216 < size 10032 716108963904 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4057 67787. MP_Allreduce 11202 1501. MP_Sync 87 MP_Alltoall 1700 36954327. MP_SendRecv 1778 218624. MP_ISendRecv 1778 218624. MP_Wait 9728 MP_ISend 6360 1080477. MP_IRecv 6360 1080477. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.174 0.221 463.631 463.646 qs_mol_dyn_low 1 2.0 0.078 0.101 462.059 462.071 qs_forces 11 3.9 0.017 0.025 460.789 460.804 qs_energies 11 4.9 0.008 0.011 447.982 448.028 scf_env_do_scf 11 5.9 0.027 0.036 414.083 414.115 velocity_verlet 10 3.0 0.010 0.014 311.856 311.880 scf_env_do_scf_inner_loop 116 6.6 0.008 0.027 282.179 282.193 dbcsr_multiply_generic 2483 12.5 0.306 0.318 195.157 195.741 qs_scf_new_mos 116 7.6 0.002 0.005 179.997 180.648 qs_scf_loop_do_ot 116 8.6 0.002 0.003 179.995 180.646 ot_scf_mini 116 9.6 0.006 0.006 169.697 170.429 multiply_cannon 2483 13.5 0.383 0.410 133.389 137.113 init_scf_loop 11 6.9 0.019 0.021 131.489 131.505 multiply_cannon_loop 2483 14.5 0.287 0.320 127.510 129.076 prepare_preconditioner 11 7.9 0.000 0.000 120.163 120.308 make_preconditioner 11 8.9 0.001 0.002 120.163 120.308 make_full_inverse_cholesky 11 9.9 0.048 0.054 96.007 116.298 multiply_cannon_multrec 9932 15.5 82.540 97.845 82.596 97.899 ot_mini 116 10.6 0.027 0.030 97.070 97.778 mp_waitall_1 102688 16.8 57.625 80.709 57.625 80.709 rebuild_ks_matrix 127 8.3 0.001 0.001 74.013 74.727 qs_ks_build_kohn_sham_matrix 127 9.3 0.018 0.020 74.012 74.727 cp_fm_upper_to_full 105 14.8 51.220 73.341 51.220 73.341 qs_ks_update_qs_env 127 7.6 0.001 0.001 67.837 68.678 apply_preconditioner_dbcsr 127 12.6 0.001 0.001 59.914 60.445 apply_single 127 13.6 0.001 0.001 59.913 60.444 ot_diis_step 116 11.6 0.041 0.054 58.670 58.675 multiply_cannon_metrocomm3 9932 15.5 0.037 0.039 36.771 57.795 make_m2s 4966 13.5 0.131 0.136 48.377 49.465 qs_rho_update_rho_low 127 7.7 0.001 0.001 45.469 45.714 calculate_rho_elec 127 8.7 0.569 0.572 45.468 45.713 dbcsr_complete_redistribute 395 12.7 4.535 4.622 31.906 44.492 copy_fm_to_dbcsr 209 11.7 0.002 0.002 27.296 39.828 qs_ot_get_derivative 116 11.6 0.002 0.003 38.291 39.012 make_images 4966 14.5 5.983 6.128 35.472 36.593 transfer_fm_to_dbcsr 11 9.9 0.040 0.044 24.106 36.505 mp_alltoall_i22 714 14.1 21.891 34.528 21.891 34.528 qs_ot_get_p 127 10.4 0.001 0.001 33.294 34.123 sum_up_and_integrate 127 10.3 0.002 0.002 32.659 32.762 integrate_v_rspace 127 11.3 0.004 0.004 32.547 32.651 grid_collocate_task_list 127 9.7 31.344 31.657 31.344 31.657 cp_fm_cholesky_invert 11 10.9 30.084 30.098 30.084 30.098 make_images_data 4966 15.5 0.073 0.082 23.217 26.687 hybrid_alltoall_any 5152 16.4 2.026 4.586 23.516 26.568 qs_ot_p2m_diag 83 11.4 0.705 0.712 24.046 24.050 init_scf_run 11 5.9 0.001 0.002 23.185 23.196 scf_env_initial_rho_setup 11 6.9 0.020 0.023 23.184 23.195 grid_integrate_task_list 127 12.3 22.254 22.638 22.254 22.638 cp_dbcsr_syevd 83 12.4 0.006 0.007 22.474 22.476 wfi_extrapolate 11 7.9 0.001 0.001 19.647 19.647 cp_fm_diag_elpa 83 13.4 0.001 0.001 18.969 18.979 cp_fm_diag_elpa_base 83 14.4 14.642 16.067 18.964 18.971 qs_ot_get_derivative_diag 77 12.4 0.003 0.003 15.869 16.389 multiply_cannon_metrocomm4 7449 15.5 0.038 0.040 6.742 15.522 mp_irecv_dv 23587 16.2 6.679 15.411 6.679 15.411 fft_wrap_pw1pw2 1281 11.7 0.021 0.024 14.546 14.607 dbcsr_make_dense_low 8542 15.6 0.095 0.096 13.539 13.686 make_dense_data 8542 16.6 13.063 13.172 13.408 13.555 fft_wrap_pw1pw2_140 519 12.2 1.150 1.176 13.061 13.224 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 12.960 13.020 cp_fm_cholesky_decompose 22 10.9 12.875 12.931 12.875 12.931 dbcsr_make_images_dense 4340 14.8 0.054 0.056 12.425 12.575 calculate_dm_sparse 127 9.5 0.001 0.002 11.661 12.264 mp_alltoall_d11v 2407 14.1 11.374 11.772 11.374 11.772 fft3d_ps 1281 13.7 5.681 5.713 10.812 10.971 density_rs2pw 127 9.7 0.007 0.008 10.043 10.882 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 9.393 9.886 copy_dbcsr_to_fm 186 11.8 0.004 0.004 9.368 9.739 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="414", plot="h2o_256_md", label="(4n/2r/18t)", y=463.646000, yerr=0.000000 PlotPoint: name="415", plot="h2o_256_md_mem", label="(4n/2r/18t)", y=3160.545455, yerr=460.359220 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/29/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 9 x 32 x 32 3321888768 0.0% 100.0% 0.0% flops 22 x 32 x 32 4060086272 0.0% 100.0% 0.0% flops 138 x 512 x 512 986298974208 0.0% 100.0% 0.0% flops 160 x 512 x 512 1334124216320 0.0% 100.0% 0.0% flops 9 x 9 x 512 1430454546432 0.0% 100.0% 0.0% flops 32 x 512 x 512 1975684956160 0.0% 100.0% 0.0% flops 22 x 9 x 512 1986255912960 0.0% 100.0% 0.0% flops 9 x 22 x 512 1992006770688 0.0% 100.0% 0.0% flops 138 x 512 x 5120 2240009994240 0.0% 100.0% 0.0% flops 129 x 512 x 512 2612262862848 0.0% 100.0% 0.0% flops 22 x 22 x 512 2753958699008 0.0% 100.0% 0.0% flops 160 x 512 x 5120 3029965209600 0.0% 100.0% 0.0% flops 151 x 512 x 512 5036318916608 0.0% 100.0% 0.0% flops 129 x 512 x 5120 5932780093440 0.0% 100.0% 0.0% flops 9 x 32 x 9 6003301257216 0.0% 100.0% 0.0% flops 22 x 32 x 9 8382804148224 0.0% 100.0% 0.0% flops 9 x 32 x 22 8382804148224 0.0% 100.0% 0.0% flops 32 x 512 x 5120 9899899617280 0.0% 100.0% 0.0% flops 151 x 512 x 5120 11438118666240 0.0% 100.0% 0.0% flops 22 x 32 x 22 11531114856448 0.0% 100.0% 0.0% flops inhomo. stacks 7886533033984 100.0% 0.0% 0.0% flops total 94.842079E+12 8.3% 91.7% 0.0% flops max/rank 25.191160E+12 10.7% 89.3% 0.0% matmuls inhomo. stacks 33456 100.0% 0.0% 0.0% matmuls total 2896413710 0.0% 100.0% 0.0% number of processed stacks 3533312 0.9% 99.1% 0.0% average stack size 1.0 827.6 0.0 marketing flops 145.663816E+12 ------------------------------------------------------------------------------- # multiplications 2535 max memory usage/rank 15.193997E+09 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 20280 MPI messages size (bytes): total size 447.369052E+09 min size 0.000000E+00 max size 209.715200E+06 average size 22.059618E+06 MPI breakdown and total messages size (bytes): size <= 128 110 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 22 2883584 131072 < size <= 4194304 9632 20199768064 4194304 < size <= 16777216 0 0 16777216 < size 10516 427166319952 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 21 12. MP_Allreduce 13511 37. MP_Alltoall 9736 3268137. MP_ISend 20236 5653666. MP_IRecv 20236 5648332. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4037 68119. MP_Allreduce 11157 1841. MP_Sync 88 MP_Alltoall 1702 67413288. MP_SendRecv 1161 1443499. MP_ISendRecv 1161 1443499. MP_Wait 3401 MP_ISend 1680 4586667. MP_IRecv 1680 4586667. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.122 0.154 423.915 423.916 qs_mol_dyn_low 1 2.0 0.032 0.033 421.940 421.946 qs_forces 11 3.9 0.027 0.028 421.341 421.349 qs_energies 11 4.9 0.048 0.049 402.588 402.607 scf_env_do_scf 11 5.9 0.015 0.047 359.204 359.218 velocity_verlet 10 3.0 0.001 0.002 283.364 283.394 scf_env_do_scf_inner_loop 118 6.6 0.013 0.037 270.400 270.420 qs_scf_new_mos 118 7.6 0.002 0.003 162.754 163.463 qs_scf_loop_do_ot 118 8.6 0.002 0.002 162.752 163.461 dbcsr_multiply_generic 2535 12.6 0.378 0.390 153.996 154.476 ot_scf_mini 118 9.6 0.024 0.043 151.051 151.634 multiply_cannon 2535 13.6 0.954 1.016 90.799 98.618 init_scf_loop 11 6.9 0.003 0.004 88.127 88.135 multiply_cannon_loop 2535 14.6 0.169 0.185 82.240 82.412 ot_mini 118 10.6 0.002 0.002 81.516 82.172 multiply_cannon_multrec 5070 15.6 75.038 76.727 75.312 77.003 prepare_preconditioner 11 7.9 0.000 0.000 76.218 76.273 make_preconditioner 11 8.9 0.002 0.002 76.218 76.273 rebuild_ks_matrix 129 8.3 0.001 0.001 69.394 70.036 qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 69.393 70.035 make_full_inverse_cholesky 11 9.9 0.072 0.076 68.462 69.497 qs_ks_update_qs_env 129 7.6 0.001 0.001 64.205 64.908 qs_rho_update_rho_low 129 7.7 0.001 0.002 55.901 56.091 calculate_rho_elec 129 8.7 1.033 1.034 55.899 56.090 make_m2s 5070 13.6 0.132 0.134 48.986 53.714 qs_ot_get_p 129 10.4 0.001 0.001 41.897 42.522 apply_preconditioner_dbcsr 129 12.6 0.016 0.016 41.948 42.453 apply_single 129 13.6 0.001 0.001 41.932 42.437 ot_diis_step 118 11.6 0.052 0.070 42.039 42.040 qs_ot_get_derivative 118 11.6 0.003 0.003 39.391 39.976 grid_collocate_task_list 129 9.7 37.961 38.429 37.961 38.429 cp_fm_cholesky_invert 11 10.9 37.490 37.499 37.490 37.499 make_images 5070 14.6 6.917 7.199 31.871 36.615 sum_up_and_integrate 129 10.3 0.002 0.002 35.503 35.537 integrate_v_rspace 129 11.3 0.004 0.005 35.397 35.431 qs_ot_p2m_diag 84 11.4 1.250 1.260 34.178 34.187 cp_dbcsr_syevd 84 12.4 0.007 0.008 31.827 31.828 mp_waitall_1 82841 16.9 22.550 31.523 22.550 31.523 hybrid_alltoall_any 5257 16.5 3.365 7.680 20.686 28.024 cp_fm_diag_elpa 84 13.4 0.000 0.000 27.932 27.933 cp_fm_diag_elpa_base 84 14.4 26.655 26.879 27.926 27.926 make_images_data 5070 15.6 0.072 0.084 19.460 26.595 init_scf_run 11 5.9 0.001 0.003 24.880 24.880 scf_env_initial_rho_setup 11 6.9 0.005 0.008 24.879 24.880 grid_integrate_task_list 129 12.3 23.627 23.738 23.627 23.738 dbcsr_complete_redistribute 397 12.7 8.731 8.961 21.294 22.000 wfi_extrapolate 11 7.9 0.001 0.001 21.450 21.450 fft_wrap_pw1pw2 1301 11.7 0.021 0.021 19.333 19.344 dbcsr_make_dense_low 6514 15.6 0.087 0.087 18.806 18.885 make_dense_data 6514 16.6 18.070 18.244 18.689 18.769 fft_wrap_pw1pw2_140 527 12.2 2.149 2.156 17.226 17.253 dbcsr_make_images_dense 4436 14.8 0.044 0.044 16.361 16.402 qs_ot_get_derivative_diag 78 12.4 0.003 0.003 15.863 16.385 copy_dbcsr_to_fm 187 11.8 0.004 0.004 15.046 15.308 cp_fm_cholesky_decompose 22 10.9 14.933 15.054 14.933 15.054 mp_alltoall_d11v 2429 14.1 13.824 14.216 13.824 14.216 fft3d_ps 1301 13.7 7.846 7.862 14.077 14.090 mp_allgather_i34 2535 14.6 5.281 13.830 5.281 13.830 calculate_dm_sparse 129 9.5 0.001 0.001 13.436 13.736 qs_energies_init_hamiltonians 11 5.9 0.001 0.002 13.631 13.633 copy_fm_to_dbcsr 210 11.7 0.002 0.002 12.758 13.389 density_rs2pw 129 9.7 0.006 0.006 13.030 13.339 cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 13.091 13.123 transfer_dbcsr_to_fm 11 10.9 0.000 0.000 11.043 11.158 build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 9.002 9.403 qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 8.807 8.873 dbcsr_dot_sd 1331 12.0 7.852 7.889 8.451 8.652 cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 8.362 8.482 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="416", plot="h2o_256_md", label="(4n/1r/36t)", y=423.916000, yerr=0.000000 PlotPoint: name="417", plot="h2o_256_md_mem", label="(4n/1r/36t)", y=11616.181818, yerr=2375.040395 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/30/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 1.924147E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9697789 0.0% 100.0% 0.0% average stack size 0.0 993.4 0.0 marketing flops 1.742118E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 683.044864E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 351648 MPI messages size (bytes): total size 4.213128E+12 min size 0.000000E+00 max size 25.408928E+06 average size 11.981093E+06 MPI breakdown and total messages size (bytes): size <= 128 92928 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 0 0 131072 < size <= 4194304 14784 17360358048 4194304 < size <= 16777216 77033 1000974278216 16777216 < size 166903 3194793140616 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 2 12. MP_Allreduce 716 49. MP_Alltoall 310 1558616. MP_ISend 5328 5698210. MP_IRecv 5328 5713852. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 1033 288600. MP_Allreduce 3140 5695. MP_Sync 4 MP_Alltoall 61 4925786. MP_SendRecv 429 12000. MP_ISendRecv 429 12000. MP_Wait 1251 MP_ISend 726 139430. MP_IRecv 726 139314. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.023 0.046 201.213 201.216 qs_energies 1 2.0 0.000 0.000 199.883 199.897 ls_scf 1 3.0 0.000 0.000 198.694 198.708 dbcsr_multiply_generic 111 6.7 0.015 0.036 182.546 184.635 multiply_cannon 111 7.7 0.018 0.021 152.511 168.904 multiply_cannon_loop 111 8.7 0.049 0.057 149.448 165.455 multiply_cannon_multrec 1332 9.7 120.998 143.579 121.117 143.704 ls_scf_main 1 4.0 0.000 0.000 114.972 114.974 density_matrix_trs4 2 5.0 0.002 0.003 108.432 109.215 ls_scf_init_scf 1 4.0 0.000 0.000 75.661 75.662 ls_scf_init_matrix_S 1 5.0 0.000 0.000 73.490 74.006 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 68.443 68.471 mp_waitall_1 12957 10.9 28.433 43.913 28.433 43.913 mp_sum_l 898 5.1 17.481 29.464 17.481 29.464 multiply_cannon_metrocomm1 1332 9.7 0.010 0.015 15.692 25.430 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 12.847 22.832 multiply_cannon_metrocomm3 1332 9.7 0.007 0.011 3.967 21.504 make_m2s 222 7.7 0.007 0.008 11.702 12.053 make_images 222 8.7 0.062 0.070 11.686 12.038 mp_irecv_dv 3257 11.0 4.097 11.615 4.097 11.615 make_images_data 222 9.7 0.004 0.006 8.942 9.648 hybrid_alltoall_any 227 10.6 0.150 2.274 6.981 8.661 ls_scf_post 1 4.0 0.000 0.000 8.061 8.076 ls_scf_store_result 1 5.0 0.000 0.000 7.522 7.919 multiply_cannon_metrocomm4 1221 9.7 0.007 0.013 2.527 6.946 multiply_cannon_metrocomm2 1221 9.7 0.008 0.015 1.622 6.804 make_images_sizes 222 9.7 0.000 0.001 0.602 5.922 mp_alltoall_i44 222 10.7 0.602 5.921 0.602 5.921 calculate_norms 2376 9.8 4.474 5.500 4.474 5.500 apply_matrix_preconditioner 6 5.3 0.000 0.000 4.482 5.077 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(4n/36r/1t)", y=201.216000, yerr=0.000000 PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(4n/36r/1t)", y=633.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/31/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 3.668084E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9699679 0.0% 100.0% 0.0% average stack size 0.0 993.3 0.0 marketing flops 1.742118E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 1.140150E+09 # max total images/rank 2 # max 3D layers 1 # MPI messages exchanged 167832 MPI messages size (bytes): total size 3.077601E+12 min size 0.000000E+00 max size 46.966736E+06 average size 18.337394E+06 MPI breakdown and total messages size (bytes): size <= 128 40320 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 320 37241600 131072 < size <= 4194304 9250 5469089776 4194304 < size <= 16777216 22076 261636692280 16777216 < size 95866 2810458169672 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 1033 295119. MP_Allreduce 3139 7842. MP_Sync 4 MP_Alltoall 54 41432280. MP_SendRecv 213 26880. MP_ISendRecv 213 26880. MP_Wait 945 MP_ISend 642 238274. MP_IRecv 642 237962. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.039 0.071 343.693 343.755 qs_energies 1 2.0 0.000 0.000 342.780 342.843 ls_scf 1 3.0 0.000 0.000 341.682 341.743 dbcsr_multiply_generic 111 6.7 0.019 0.031 315.400 317.843 multiply_cannon 111 7.7 0.025 0.029 269.667 294.187 multiply_cannon_loop 111 8.7 0.087 0.106 263.875 289.708 multiply_cannon_multrec 1332 9.7 235.277 266.560 235.433 266.715 ls_scf_main 1 4.0 0.000 0.002 197.746 197.781 density_matrix_trs4 2 5.0 0.003 0.008 186.907 187.919 ls_scf_init_scf 1 4.0 0.000 0.000 130.177 130.199 ls_scf_init_matrix_S 1 5.0 0.000 0.000 126.655 127.434 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.005 118.058 118.084 mp_waitall_1 10071 10.9 25.384 51.929 25.384 51.929 mp_sum_l 898 5.1 24.458 43.046 24.458 43.046 multiply_cannon_metrocomm3 1332 9.7 0.009 0.012 8.002 36.764 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 19.659 35.312 multiply_cannon_metrocomm1 1332 9.7 0.006 0.007 6.043 23.582 make_m2s 222 7.7 0.009 0.012 19.666 20.391 make_images 222 8.7 1.621 2.078 19.635 20.364 mp_irecv_dv 3391 11.0 6.471 16.904 6.471 16.904 multiply_cannon_metrocomm4 1221 9.7 0.010 0.012 5.363 14.837 ls_scf_post 1 4.0 0.000 0.000 13.759 13.775 ls_scf_store_result 1 5.0 0.000 0.000 12.988 13.563 make_images_data 222 9.7 0.006 0.006 11.838 13.126 hybrid_alltoall_any 227 10.6 0.417 2.928 10.080 12.248 calculate_norms 2376 9.8 7.592 8.552 7.592 8.552 apply_matrix_preconditioner 6 5.3 0.000 0.000 6.840 8.094 arnoldi_extremal 4 6.8 0.000 0.000 7.411 7.439 arnoldi_normal_ev 4 7.8 0.001 0.004 7.411 7.439 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(4n/18r/2t)", y=343.755000, yerr=0.000000 PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(4n/18r/2t)", y=1026.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/32/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 7.107463E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9667097 0.0% 100.0% 0.0% average stack size 0.0 996.6 0.0 marketing flops 1.742118E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 2.017141E+09 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 39960 MPI messages size (bytes): total size 1.915058E+12 min size 0.000000E+00 max size 93.908080E+06 average size 47.924364E+06 MPI breakdown and total messages size (bytes): size <= 128 9600 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 0 0 131072 < size <= 4194304 1920 1170063360 4194304 < size <= 16777216 720 6721008480 16777216 < size 27720 1907167008560 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 2 12. MP_Allreduce 716 87. MP_Alltoall 310 5824960. MP_ISend 2664 20322846. MP_IRecv 2664 20175729. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 1033 294684. MP_Allreduce 3138 10459. MP_Sync 4 MP_Alltoall 47 20667983. MP_SendRecv 105 57600. MP_ISendRecv 105 57600. MP_Wait 567 MP_ISend 378 618054. MP_IRecv 378 618834. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.067 0.168 353.506 353.510 qs_energies 1 2.0 0.000 0.000 352.457 352.468 ls_scf 1 3.0 0.000 0.000 350.672 350.682 dbcsr_multiply_generic 111 6.7 0.019 0.022 322.331 326.082 multiply_cannon 111 7.7 0.026 0.030 267.009 295.109 multiply_cannon_loop 111 8.7 0.065 0.084 259.155 288.307 multiply_cannon_multrec 666 9.7 232.783 266.411 232.921 266.521 ls_scf_main 1 4.0 0.000 0.001 202.390 202.409 density_matrix_trs4 2 5.0 0.003 0.006 190.394 191.728 ls_scf_init_scf 1 4.0 0.000 0.000 134.024 134.025 ls_scf_init_matrix_S 1 5.0 0.000 0.000 130.082 131.067 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.005 121.238 121.258 mp_waitall_1 7293 11.0 29.324 54.749 29.324 54.749 mp_sum_l 898 5.1 27.314 46.027 27.314 46.027 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 22.192 38.357 multiply_cannon_metrocomm1 666 9.7 0.004 0.007 10.974 29.623 make_m2s 222 7.7 0.007 0.008 24.982 26.708 make_images 222 8.7 3.134 3.848 24.941 26.669 make_images_data 222 9.7 0.004 0.007 16.406 19.295 hybrid_alltoall_any 227 10.6 0.658 4.826 15.835 18.074 mp_irecv_dv 1601 11.0 6.650 17.779 6.650 17.779 multiply_cannon_metrocomm3 666 9.7 0.003 0.005 2.576 17.448 multiply_cannon_metrocomm4 555 9.7 0.003 0.007 4.090 14.580 ls_scf_post 1 4.0 0.000 0.000 14.258 14.277 ls_scf_store_result 1 5.0 0.000 0.000 13.361 13.983 multiply_cannon_metrocomm2 555 9.7 0.004 0.006 3.405 11.595 make_images_sizes 222 9.7 0.000 0.001 2.294 8.843 mp_alltoall_i44 222 10.7 2.293 8.843 2.293 8.843 apply_matrix_preconditioner 6 5.3 0.000 0.000 7.396 8.360 mp_allgather_i34 111 8.7 2.188 7.401 2.188 7.401 arnoldi_extremal 4 6.8 0.000 0.000 7.151 7.185 arnoldi_normal_ev 4 7.8 0.001 0.004 7.151 7.185 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(4n/9r/4t)", y=353.510000, yerr=0.000000 PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(4n/9r/4t)", y=1405.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/33/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 10.747127E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9703792 0.0% 100.0% 0.0% average stack size 0.0 992.8 0.0 marketing flops 1.742116E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 2.920604E+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 30.356988E+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 1033 301899. 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 0.049 0.070 352.747 352.752 qs_energies 1 2.0 0.000 0.000 350.850 350.859 ls_scf 1 3.0 0.000 0.000 349.242 349.252 dbcsr_multiply_generic 111 6.7 0.022 0.026 320.160 322.337 multiply_cannon 111 7.7 0.029 0.033 263.498 283.260 multiply_cannon_loop 111 8.7 0.101 0.110 255.367 275.465 multiply_cannon_multrec 1332 9.7 240.120 260.261 240.238 260.375 ls_scf_main 1 4.0 0.000 0.001 202.572 202.591 density_matrix_trs4 2 5.0 0.003 0.005 190.756 191.496 ls_scf_init_scf 1 4.0 0.000 0.000 132.464 132.465 ls_scf_init_matrix_S 1 5.0 0.000 0.000 128.723 129.278 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.003 119.726 119.742 mp_sum_l 898 5.1 20.107 38.875 20.107 38.875 make_m2s 222 7.7 0.009 0.011 32.635 33.637 make_images 222 8.7 3.718 4.152 32.575 33.579 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 16.615 32.587 mp_waitall_1 6369 11.0 23.098 28.987 23.098 28.987 make_images_data 222 9.7 0.005 0.008 18.929 20.724 hybrid_alltoall_any 227 10.6 0.931 4.025 17.977 20.456 multiply_cannon_metrocomm4 1110 9.7 0.007 0.011 4.606 15.977 mp_irecv_dv 3229 10.9 4.600 15.949 4.600 15.949 ls_scf_post 1 4.0 0.000 0.000 14.207 14.218 ls_scf_store_result 1 5.0 0.000 0.000 13.481 13.967 multiply_cannon_metrocomm1 1332 9.7 0.004 0.006 3.671 8.831 dbcsr_data_release 10900 10.7 7.262 8.075 7.262 8.075 apply_matrix_preconditioner 6 5.3 0.000 0.000 7.149 7.945 arnoldi_extremal 4 6.8 0.000 0.000 7.416 7.465 arnoldi_normal_ev 4 7.8 0.001 0.006 7.416 7.465 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(4n/6r/6t)", y=352.752000, yerr=0.000000 PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(4n/6r/6t)", y=1999.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/34/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 15.383312E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9657067 0.0% 100.0% 0.0% average stack size 0.0 997.6 0.0 marketing flops 1.742118E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 4.242809E+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 716 126. MP_Alltoall 310 12920694. MP_ISend 1776 40180426. MP_IRecv 1776 40465032. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 1033 300769. 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 0.101 0.262 357.224 357.238 qs_energies 1 2.0 0.001 0.006 355.423 355.431 ls_scf 1 3.0 0.001 0.004 353.280 353.304 dbcsr_multiply_generic 111 6.7 0.022 0.025 321.063 322.768 multiply_cannon 111 7.7 0.030 0.032 267.934 281.137 multiply_cannon_loop 111 8.7 0.074 0.085 257.379 266.502 multiply_cannon_multrec 444 9.7 237.894 249.434 238.007 249.544 ls_scf_main 1 4.0 0.052 0.214 206.629 206.638 density_matrix_trs4 2 5.0 0.009 0.048 193.214 193.981 ls_scf_init_scf 1 4.0 0.000 0.002 132.499 132.502 ls_scf_init_matrix_S 1 5.0 0.000 0.001 129.051 129.538 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.004 119.889 119.898 mp_waitall_1 5436 11.0 34.002 39.752 34.002 39.752 make_m2s 222 7.7 0.006 0.007 34.423 37.914 make_images 222 8.7 3.690 4.437 34.352 37.838 make_images_data 222 9.7 0.004 0.006 24.572 29.195 hybrid_alltoall_any 227 10.6 1.179 5.585 22.724 28.101 mp_sum_l 898 5.1 13.538 25.244 13.538 25.244 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 10.386 20.125 ls_scf_post 1 4.0 0.006 0.028 14.151 14.182 mp_allgather_i34 111 8.7 3.554 14.055 3.554 14.055 ls_scf_store_result 1 5.0 0.000 0.000 13.561 13.892 multiply_cannon_metrocomm1 444 9.7 0.003 0.004 9.098 11.906 make_images_sizes 222 9.7 0.000 0.001 2.401 11.582 mp_alltoall_i44 222 10.7 2.401 11.582 2.401 11.582 dbcsr_data_release 10924 10.7 8.606 9.672 8.606 9.672 apply_matrix_preconditioner 6 5.3 0.000 0.000 7.969 8.783 arnoldi_extremal 4 6.8 0.007 0.026 8.635 8.659 arnoldi_normal_ev 4 7.8 0.013 0.051 8.628 8.659 multiply_cannon_metrocomm4 333 9.7 0.002 0.003 2.958 7.455 mp_irecv_dv 1241 11.2 2.960 7.445 2.960 7.445 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(4n/4r/9t)", y=357.238000, yerr=0.000000 PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(4n/4r/9t)", y=2777.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/35/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 23 x 23 x 23 234439235724792 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 234.439236E+12 0.0% 100.0% 0.0% flops max/rank 20.557908E+12 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 9634225188 0.0% 100.0% 0.0% number of processed stacks 9673530 0.0% 100.0% 0.0% average stack size 0.0 995.9 0.0 marketing flops 1.742116E+15 ------------------------------------------------------------------------------- # multiplications 111 max memory usage/rank 5.584736E+09 # max total images/rank 3 # max 3D layers 1 # MPI messages exchanged 10656 MPI messages size (bytes): total size 1.158041E+12 min size 0.000000E+00 max size 265.321008E+06 average size 108.674984E+06 MPI breakdown and total messages size (bytes): size <= 128 1536 0 128 < size <= 8192 0 0 8192 < size <= 32768 0 0 32768 < size <= 131072 0 0 131072 < size <= 4194304 1536 702038016 4194304 < size <= 16777216 72 672100848 16777216 < size 7512 1156666219168 ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 1033 319979. MP_Allreduce 3129 18839. MP_Sync 4 MP_Alltoall 47 60354741. MP_SendRecv 33 144000. MP_ISendRecv 33 144000. MP_Wait 432 MP_ISend 336 1403879. MP_IRecv 336 1404443. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.140 0.179 367.544 367.556 qs_energies 1 2.0 0.006 0.019 365.913 365.931 ls_scf 1 3.0 0.003 0.004 363.214 363.236 dbcsr_multiply_generic 111 6.7 0.024 0.026 327.925 328.736 multiply_cannon 111 7.7 0.032 0.033 270.050 286.164 multiply_cannon_loop 111 8.7 0.089 0.095 259.668 274.678 multiply_cannon_multrec 666 9.7 227.286 234.577 227.374 234.671 ls_scf_main 1 4.0 0.025 0.049 217.278 217.298 density_matrix_trs4 2 5.0 0.083 0.325 201.934 202.382 ls_scf_init_scf 1 4.0 0.001 0.002 131.639 131.642 ls_scf_init_matrix_S 1 5.0 0.001 0.001 127.994 128.311 matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.004 119.008 119.021 mp_waitall_1 5424 11.0 40.097 57.698 40.097 57.698 make_m2s 222 7.7 0.007 0.008 37.496 40.590 make_images 222 8.7 4.559 5.063 37.407 40.498 multiply_cannon_metrocomm3 666 9.7 0.004 0.005 16.583 35.689 multiply_cannon_metrocomm4 555 9.7 0.003 0.006 8.580 28.451 mp_irecv_dv 1779 11.1 8.540 28.248 8.540 28.248 hybrid_alltoall_any 227 10.6 1.690 3.755 23.872 28.010 make_images_data 222 9.7 0.005 0.007 24.156 27.416 mp_sum_l 898 5.1 16.916 25.722 16.916 25.722 dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 13.921 22.199 ls_scf_post 1 4.0 0.016 0.023 14.295 14.317 ls_scf_store_result 1 5.0 0.000 0.000 13.732 13.995 dbcsr_data_release 12835 10.6 7.084 12.336 7.084 12.336 apply_matrix_preconditioner 6 5.3 0.000 0.000 9.446 9.745 arnoldi_extremal 4 6.8 0.001 0.004 8.339 8.349 arnoldi_normal_ev 4 7.8 0.004 0.011 8.338 8.349 ls_scf_dm_to_ks 2 5.0 0.000 0.000 7.429 7.471 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(4n/3r/12t)", y=367.556000, yerr=0.000000 PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(4n/3r/12t)", y=3579.000000, yerr=0.000000 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ~~~~~~~~~ RESULT ~~~~~~~~~ RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC40-gnu/2ed94b8b96b2a4590f5340063d84c5d639a6fefe_performance_tests/36/result.log @@@@@@@@@@ Run number: 1 @@@@@@@@@@ ------------------------------------------------------------------------------- - - - DBCSR STATISTICS - - - ------------------------------------------------------------------------------- COUNTER TOTAL BLAS SMM ACC flops 64 x 64 x 64 109521666048 0.0% 100.0% 0.0% flops 32 x 32 x 849 202529832960 0.0% 100.0% 0.0% flops 32 x 32 x 853 203484037120 0.0% 100.0% 0.0% flops 32 x 32 x 858 204676792320 0.0% 100.0% 0.0% flops 64 x 64 x 96 328564998144 0.0% 100.0% 0.0% flops 64 x 96 x 64 328564998144 0.0% 100.0% 0.0% flops 96 x 64 x 64 328564998144 0.0% 100.0% 0.0% flops 9 x 32 x 32 549621596160 0.0% 100.0% 0.0% flops 22 x 32 x 32 671759728640 0.0% 100.0% 0.0% flops 64 x 64 x 849 936533557248 0.0% 100.0% 0.0% flops 64 x 64 x 853 940945965056 0.0% 100.0% 0.0% flops 64 x 64 x 858 946461474816 0.0% 100.0% 0.0% flops 96 x 96 x 64 985694994432 0.0% 100.0% 0.0% flops 64 x 96 x 96 985694994432 0.0% 100.0% 0.0% flops 96 x 64 x 96 985694994432 0.0% 100.0% 0.0% flops 849 x 64 x 64 1285508038656 0.0% 100.0% 0.0% flops 853 x 64 x 64 1291564613632 0.0% 100.0% 0.0% flops 858 x 64 x 64 1299135332352 0.0% 100.0% 0.0% flops 9 x 9 x 64 1833777211392 0.0% 100.0% 0.0% flops 9 x 22 x 64 2466560397312 0.0% 100.0% 0.0% flops 22 x 9 x 64 2471027226624 0.0% 100.0% 0.0% flops 64 x 96 x 849 2809600671744 0.0% 100.0% 0.0% flops 96 x 64 x 849 2809600671744 0.0% 100.0% 0.0% flops 64 x 96 x 853 2822837895168 0.0% 100.0% 0.0% flops 96 x 64 x 853 2822837895168 0.0% 100.0% 0.0% flops 64 x 96 x 858 2839384424448 0.0% 100.0% 0.0% flops 96 x 64 x 858 2839384424448 0.0% 100.0% 0.0% flops 849 x 64 x 849 2928781688832 0.0% 100.0% 0.0% flops 849 x 64 x 853 2942580424704 0.0% 100.0% 0.0% flops 853 x 64 x 849 2942580424704 0.0% 100.0% 0.0% flops 853 x 64 x 853 2956444172288 0.0% 100.0% 0.0% flops 96 x 96 x 96 2957084983296 0.0% 100.0% 0.0% flops 849 x 64 x 858 2959828844544 0.0% 100.0% 0.0% flops 858 x 64 x 849 2959828844544 0.0% 100.0% 0.0% flops 853 x 64 x 858 2973773856768 0.0% 100.0% 0.0% flops 858 x 64 x 853 2973773856768 0.0% 100.0% 0.0% flops 858 x 64 x 858 2991205122048 0.0% 100.0% 0.0% flops 22 x 22 x 64 3338610130944 0.0% 100.0% 0.0% flops 849 x 64 x 96 3856524115968 0.0% 100.0% 0.0% flops 849 x 96 x 64 3856524115968 0.0% 100.0% 0.0% flops 853 x 64 x 96 3874693840896 0.0% 100.0% 0.0% flops 853 x 96 x 64 3874693840896 0.0% 100.0% 0.0% flops 858 x 96 x 64 3897405997056 0.0% 100.0% 0.0% flops 858 x 64 x 96 3897405997056 0.0% 100.0% 0.0% flops 9 x 9 x 96 5501331634176 0.0% 100.0% 0.0% flops 9 x 22 x 96 7399681191936 0.0% 100.0% 0.0% flops 22 x 9 x 96 7413081679872 0.0% 100.0% 0.0% flops 96 x 96 x 849 8428802015232 0.0% 100.0% 0.0% flops 96 x 96 x 853 8468513685504 0.0% 100.0% 0.0% flops 96 x 96 x 858 8518153273344 0.0% 100.0% 0.0% flops 849 x 96 x 849 8786345066496 0.0% 100.0% 0.0% flops 849 x 96 x 853 8827741274112 0.0% 100.0% 0.0% flops 853 x 96 x 849 8827741274112 0.0% 100.0% 0.0% flops 853 x 96 x 853 8869332516864 0.0% 100.0% 0.0% flops 858 x 96 x 849 8879486533632 0.0% 100.0% 0.0% flops 849 x 96 x 858 8879486533632 0.0% 100.0% 0.0% flops 858 x 96 x 853 8921321570304 0.0% 100.0% 0.0% flops 853 x 96 x 858 8921321570304 0.0% 100.0% 0.0% flops 858 x 96 x 858 8973615366144 0.0% 100.0% 0.0% flops 22 x 22 x 96 10015830392832 0.0% 100.0% 0.0% flops 849 x 96 x 96 11569572347904 0.0% 100.0% 0.0% flops 853 x 96 x 96 11624081522688 0.0% 100.0% 0.0% flops 858 x 96 x 96 11692217991168 0.0% 100.0% 0.0% flops 9 x 32 x 9 21312216612864 0.0% 100.0% 0.0% flops 22 x 32 x 9 29317892972544 0.0% 100.0% 0.0% flops 9 x 32 x 22 29317892972544 0.0% 100.0% 0.0% flops 22 x 32 x 22 40107728764928 0.0% 100.0% 0.0% flops inhomo. stacks 0 0.0% 0.0% 0.0% flops total 383.054662E+12 0.0% 100.0% 0.0% flops max/rank 769.048094E+09 0.0% 100.0% 0.0% matmuls inhomo. stacks 0 0.0% 0.0% 0.0% matmuls total 11370092824 0.0% 100.0% 0.0% number of processed stacks 36472128 0.0% 100.0% 0.0% average stack size 0.0 311.7 0.0 marketing flops 780.451392E+12 ------------------------------------------------------------------------------- # multiplications 1445 max memory usage/rank 362.881024E+06 # max total images/rank 1 # max 3D layers 1 # MPI messages exchanged 38286720 MPI messages size (bytes): total size 22.066386E+12 min size 0.000000E+00 max size 5.889312E+06 average size 576.345688E+03 MPI breakdown and total messages size (bytes): size <= 128 274344 0 128 < size <= 8192 0 0 8192 < size <= 32768 1746712 57194053632 32768 < size <= 131072 13942784 856644648960 131072 < size <= 4194304 21501504 16367441085440 4194304 < size <= 16777216 821376 4784862003200 16777216 < size 0 0 ------------------------------------------------------------------------------- - - - DBCSR MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Bcast 68 12. MP_Allreduce 7415 50. MP_Alltoall 5329 496642. MP_ISend 138692 290073. MP_IRecv 138692 288852. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - MESSAGE PASSING PERFORMANCE - - - ------------------------------------------------------------------------------- ROUTINE CALLS AVE VOLUME [Bytes] MP_Group 4 MP_Bcast 4640 77189. MP_Allreduce 13232 2300. MP_Sync 1064 MP_Alltoall 2588 4919617. MP_SendRecv 126500 14304. MP_ISendRecv 69000 14304. MP_Wait 78200 MP_comm_split 40 MP_ISend 24680 99620. MP_IRecv 36480 68885. ------------------------------------------------------------------------------- ------------------------------------------------------------------------------- - - - T I M I N G - - - ------------------------------------------------------------------------------- SUBROUTINE CALLS ASD SELF TIME TOTAL TIME MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM CP2K 1 1.0 0.181 0.745 307.426 307.619 qs_mol_dyn_low 1 2.0 0.072 0.316 303.087 303.279 qs_forces 5 3.8 0.005 0.006 302.089 302.300 qs_energies 5 4.8 0.003 0.019 296.899 297.122 scf_env_do_scf 5 5.8 0.000 0.001 283.266 283.438 scf_env_do_scf_inner_loop 105 6.6 0.002 0.008 233.956 234.088 qs_scf_new_mos 105 7.6 0.001 0.001 145.303 146.559 qs_scf_loop_do_ot 105 8.6 0.001 0.001 145.302 146.558 velocity_verlet 4 3.0 0.001 0.001 139.929 140.028 ot_scf_mini 105 9.6 0.004 0.006 133.745 134.587 dbcsr_multiply_generic 1445 12.2 0.146 0.183 107.614 110.793 multiply_cannon 1445 13.2 0.185 0.216 86.216 92.166 multiply_cannon_loop 1445 14.2 0.334 0.495 82.681 88.924 qs_ot_get_p 112 10.4 0.001 0.002 65.901 67.842 mp_waitall_1 372490 16.1 53.479 67.679 53.479 67.679 rebuild_ks_matrix 110 8.4 0.001 0.001 54.480 55.008 qs_ks_build_kohn_sham_matrix 110 9.4 0.013 0.016 54.479 55.007 qs_ot_p2m_diag 40 11.0 0.022 0.030 51.072 51.154 qs_ks_update_qs_env 112 7.6 0.001 0.001 50.253 50.734 cp_dbcsr_syevd 40 12.0 0.003 0.004 50.073 50.141 init_scf_loop 7 6.6 0.000 0.001 49.222 49.330 ot_mini 105 10.6 0.001 0.001 46.863 47.774 multiply_cannon_multrec 34680 15.2 39.239 47.085 39.250 47.095 qs_rho_update_rho_low 110 7.6 0.001 0.001 46.205 46.968 calculate_rho_elec 110 8.6 0.047 0.049 46.205 46.967 multiply_cannon_metrocomm3 34680 15.2 0.125 0.254 11.635 46.642 cp_fm_syevd 40 13.0 0.000 0.001 43.033 43.172 prepare_preconditioner 7 7.6 0.000 0.000 41.214 41.354 make_preconditioner 7 8.6 0.000 0.000 41.214 41.354 multiply_cannon_metrocomm1 34680 15.2 0.146 0.274 27.747 39.340 cp_fm_redistribute_end 40 14.0 18.395 36.683 18.405 36.694 cp_fm_syevd_base 40 14.0 18.273 36.568 18.273 36.568 make_full_inverse_cholesky 7 9.6 0.000 0.000 31.993 32.327 qs_ot_get_derivative 55 11.6 0.001 0.001 29.629 30.498 fft_wrap_pw1pw2 1425 12.5 0.019 0.022 25.368 25.927 density_rs2pw 110 9.6 0.006 0.009 22.798 24.822 fft_wrap_pw1pw2_240 915 14.0 0.428 0.474 22.659 23.377 sum_up_and_integrate 60 10.3 0.001 0.002 23.003 23.049 integrate_v_rspace 60 11.3 0.002 0.003 22.970 23.016 grid_collocate_task_list 110 9.6 20.400 22.037 20.400 22.037 fft3d_pb 915 15.0 4.971 5.397 20.539 21.307 qs_vxc_create 110 10.4 0.002 0.003 19.605 19.703 apply_preconditioner_dbcsr 62 12.6 0.000 0.000 17.777 18.914 apply_single 62 13.6 0.000 0.000 17.777 18.913 transfer_rs2pw 445 10.6 0.007 0.010 15.820 17.821 ot_new_cg_direction 55 11.6 0.001 0.001 16.973 16.984 cp_fm_cholesky_invert 7 10.6 16.915 16.926 16.915 16.926 make_m2s 2890 13.2 0.097 0.113 15.073 16.486 mp_alltoall_z22v 2340 16.7 14.365 15.643 14.365 15.643 make_images 2890 14.2 0.238 0.270 13.880 15.445 xc_pw_derive 510 13.4 0.006 0.008 13.602 13.914 cp_fm_cholesky_decompose 14 10.2 13.710 13.759 13.710 13.759 xc_vxc_pw_create 60 11.3 0.079 0.097 13.536 13.613 calculate_dm_sparse 110 9.5 0.001 0.001 12.183 13.102 xc_rho_set_and_dset_create 110 12.4 0.138 0.173 13.025 13.102 mp_waitany 6270 13.5 10.599 13.016 10.599 13.016 check_diag 80 13.5 10.239 10.490 12.495 12.630 make_images_data 2890 15.2 0.069 0.090 9.918 11.887 qs_ot_get_derivative_taylor 37 12.8 0.001 0.002 10.624 11.193 potential_pw2rs 60 12.3 0.003 0.005 10.738 10.813 grid_integrate_task_list 60 12.3 9.977 10.588 9.977 10.588 mp_sum_l 7231 12.6 5.370 10.296 5.370 10.296 hybrid_alltoall_any 2983 16.1 0.101 1.297 7.592 9.921 init_scf_run 5 5.8 0.000 0.000 9.902 9.910 scf_env_initial_rho_setup 5 6.8 0.000 0.001 9.902 9.909 mp_alltoall_d11v 1300 13.8 7.513 8.665 7.513 8.665 transfer_pw2rs 245 13.2 0.003 0.004 8.539 8.618 make_full_single_inverse 7 9.6 0.001 0.001 8.157 8.283 cube_transpose_3 560 16.1 0.518 0.664 7.317 8.065 mp_irecv_dv 72697 16.1 3.209 7.354 3.209 7.354 transfer_rs2pw_240 115 11.5 1.052 1.419 4.223 7.182 multiply_cannon_metrocomm4 33235 15.2 0.115 0.236 3.037 7.125 qs_ot_get_derivative_diag 18 12.0 0.001 0.001 6.749 7.058 make_images_sizes 2890 15.2 0.004 0.009 2.366 6.489 mp_alltoall_i44 2890 16.2 2.362 6.485 2.362 6.485 mp_allgather_i34 1445 14.2 2.239 6.480 2.239 6.480 mp_sum_d 3661 11.6 3.283 6.364 3.283 6.364 xc_pw_divergence 60 12.3 0.002 0.002 6.158 6.273 qs_ot_get_orbitals 105 10.6 0.001 0.001 5.863 6.155 ------------------------------------------------------------------------------- ~ ~ ~ ~ DATA POINTS ~ ~ ~ ~ PlotPoint: name="601", plot="h2o_512_md", label="(16n/36r/1t)", y=307.619000, yerr=0.000000 PlotPoint: name="602", plot="h2o_512_md_mem", label="(16n/36r/1t)", y=345.000000, yerr=2.097618 ~ ~ ~ ~ END DATA POINTS ~ ~ ~ ~~~~~~ END RESULT ~~~~~~~~ ========= END RESULTS =========== CommitSHA: 2ed94b8b96b2a4590f5340063d84c5d639a6fefe Summary: empty Status: OK