=== 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: 2216e2348d729b940ade5c6e00d35197f22708d2
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.2, ELPA 2022.11.001.rc2, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.0.0, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.1,
# SIRIUS 7.4.0, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Author: Matthias Krack (21.12.2022)
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2022.11.001.rc2
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.0.0
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.1
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.0
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
HDF5_VER := $(strip $(HDF5_VER))
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(HDF5_VER)/lib
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__HDF5
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta_prefixed_scalapack.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/01
job id: 43722886
--- Point ---
name: 10
plot: h2o_32_ri_rpa_mp2
regex: Total RI-RPA Time=
label: RI-RPA (8n/2r/6t)
--- Point ---
name: 11
plot: h2o_32_ri_rpa_mp2_mem
regex: Estimated peak process memory
label: RI-RPA (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-MP2.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-HF-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-MP2-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/02
job id: 43722887
--- Point ---
name: 20
plot: h2o_32_ri_rpa_mp2
regex: Total MP2 Time=
label: RI-MP2 (8n/6r/2t)
--- Point ---
name: 21
plot: h2o_32_ri_rpa_mp2_mem
regex: Estimated peak process memory
label: RI-MP2 (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/03
job id: 43722888
--- Point ---
name: 100
plot: h2o_64_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 101
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/04
job id: 43722889
--- Point ---
name: 102
plot: h2o_64_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 103
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/05
job id: 43722890
--- Point ---
name: 104
plot: h2o_64_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 105
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/06
job id: 43722891
--- Point ---
name: 106
plot: h2o_64_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 107
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/07
job id: 43722892
--- Point ---
name: 108
plot: h2o_64_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 109
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-64 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-64.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 5
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/08
job id: 43722893
--- Point ---
name: 110
plot: h2o_64_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 111
plot: h2o_64_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/09
job id: 43722894
--- Point ---
name: 200
plot: h2o_128_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 201
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/10
job id: 43722895
--- Point ---
name: 202
plot: h2o_128_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 203
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/11
job id: 43722896
--- Point ---
name: 204
plot: h2o_128_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 205
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/12
job id: 43722897
--- Point ---
name: 206
plot: h2o_128_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 207
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/13
job id: 43722898
--- Point ---
name: 208
plot: h2o_128_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 209
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-128 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-128.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/14
job id: 43722899
--- Point ---
name: 210
plot: h2o_128_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 211
plot: h2o_128_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/15
job id: 43722900
--- Point ---
name: 400
plot: h2o_256_md
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 401
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/16
job id: 43722901
--- Point ---
name: 402
plot: h2o_256_md
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 403
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/17
job id: 43722902
--- Point ---
name: 404
plot: h2o_256_md
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 405
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/18
job id: 43722903
--- Point ---
name: 406
plot: h2o_256_md
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 407
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/19
job id: 43722904
--- Point ---
name: 408
plot: h2o_256_md
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 409
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-256 test - DBCSR dominated (MPI/OMP)
input file: benchmarks/QS/H2O-256.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 30
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/20
job id: 43722905
--- Point ---
name: 410
plot: h2o_256_md
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 411
plot: h2o_256_md_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/21
job id: 43722906
--- Point ---
name: 500
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/12r/1t)
--- Point ---
name: 501
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 6
# threads/rank = 2
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/22
job id: 43722907
--- Point ---
name: 502
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/6r/2t)
--- Point ---
name: 503
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/6r/2t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 4
# threads/rank = 3
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/23
job id: 43722908
--- Point ---
name: 504
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/4r/3t)
--- Point ---
name: 505
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/4r/3t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 3
# threads/rank = 4
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/24
job id: 43722909
--- Point ---
name: 506
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/3r/4t)
--- Point ---
name: 507
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/3r/4t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/25
job id: 43722910
--- Point ---
name: 508
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/2r/6t)
--- Point ---
name: 509
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/2r/6t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 (NREP 3) linear scaling test (864 H2O)
input file: benchmarks/QS_DM_LS/H2O-dft-ls.inp
required files: []
output file: result.log
# nodes = 8
# ranks/node = 1
# threads/rank = 12
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/26
job id: 43722911
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
=== 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]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 344 9.
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 0.022 0.036 135.211 135.212
farming_run 1 2.0 134.754 134.758 135.180 135.184
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 4194304 0.0% 0.0% 100.0%
flops 14 x 32 x 32 154140672 0.0% 0.0% 100.0%
flops 29 x 32 x 32 159645696 0.0% 0.0% 100.0%
flops 14 x 14 x 32 208732160 0.0% 0.0% 100.0%
flops 29 x 14 x 32 212860928 0.0% 0.0% 100.0%
flops 14 x 29 x 32 212860928 0.0% 0.0% 100.0%
flops 29 x 29 x 32 227352576 0.0% 0.0% 100.0%
flops 14 x 32 x 14 896801644032 0.0% 0.0% 100.0%
flops 29 x 32 x 14 928925089792 0.0% 0.0% 100.0%
flops 14 x 32 x 29 928925089792 0.0% 0.0% 100.0%
flops 29 x 32 x 29 962100985856 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693169221632 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753639550976 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.164741E+12 0.0% 0.0% 100.0%
flops max/rank 447.801317E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249492158 0.0% 0.0% 100.0%
number of processed stacks 164328 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1518.3
marketing flops 7.165779E+12
-------------------------------------------------------------------------------
# multiplications 1160
max memory usage/rank 1.456452E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 228 1113141.
MP_Allreduce 485 2282278.
MP_Sync 27
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 8
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 0.011 0.024 116.167 116.167
qs_energies 1 2.0 0.000 0.000 115.932 115.933
mp2_main 1 3.0 0.000 0.000 113.612 113.613
mp2_gpw_main 1 4.0 0.019 0.025 112.472 112.473
mp2_ri_gpw_compute_in 1 5.0 0.171 0.172 93.254 93.720
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.004 55.354 55.819
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.165 41.656 47.015
get_2c_integrals 1 6.0 0.000 0.000 36.992 37.729
integrate_v_rspace 273 8.0 0.439 0.453 25.082 30.216
pw_transfer 6555 10.6 0.379 0.387 27.259 27.783
grid_integrate_task_list 273 9.0 20.907 26.493 20.907 26.493
fft_wrap_pw1pw2 5465 11.4 0.045 0.047 25.915 26.428
fft_wrap_pw1pw2_100 2178 12.4 1.163 1.229 23.472 23.987
compute_2c_integrals 1 7.0 0.002 0.002 19.324 19.325
rpa_ri_compute_en 1 5.0 0.000 0.000 19.110 19.324
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.824 18.980
mp2_eri_2c_integrate_gpw 1 9.0 2.393 2.437 18.821 18.977
cp_fm_cholesky_decompose 12 8.2 17.740 18.447 17.740 18.447
cholesky_decomp 1 7.0 0.000 0.000 16.504 17.245
fft3d_s 5443 13.4 16.100 16.512 16.123 16.534
ao_to_mo_and_store_B_mult_1 272 7.0 10.854 15.553 10.854 15.553
calculate_wavefunction 272 8.0 5.390 5.529 12.492 13.130
rpa_num_int 1 6.0 0.000 0.000 10.770 10.770
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.680 10.767
calc_mat_Q 8 8.0 0.000 0.000 9.403 9.522
contract_S_to_Q 8 9.0 0.000 0.000 8.824 8.944
calc_potential_gpw 544 9.5 0.005 0.006 8.230 8.646
parallel_gemm_fm 14 9.1 0.000 0.000 8.420 8.480
parallel_gemm_fm_cosma 14 10.1 8.420 8.480 8.420 8.480
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.186 8.416
potential_pw2rs 545 10.0 0.107 0.109 7.670 8.345
create_integ_mat 1 6.0 0.006 0.008 7.780 7.780
collocate_single_gaussian 272 10.0 0.041 0.043 7.427 7.631
array2fm 1 7.0 0.000 0.000 6.749 7.117
pw_scatter_s 2720 13.7 4.428 4.598 4.428 4.598
pw_gather_s 2722 13.2 3.850 4.263 3.850 4.263
array2fm_buffer_send 1 8.0 2.987 3.149 2.987 3.149
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=112.472639, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2727.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 344 10.
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 0.031 0.039 396.968 396.970
farming_run 1 2.0 396.148 396.154 396.920 396.923
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 16777216 0.0% 0.0% 100.0%
flops 14 x 32 x 32 565182464 0.0% 0.0% 100.0%
flops 29 x 32 x 32 585367552 0.0% 0.0% 100.0%
flops 14 x 14 x 32 626196480 0.0% 0.0% 100.0%
flops 29 x 14 x 32 638582784 0.0% 0.0% 100.0%
flops 14 x 29 x 32 638582784 0.0% 0.0% 100.0%
flops 29 x 29 x 32 682057728 0.0% 0.0% 100.0%
flops 14 x 32 x 14 897827141120 0.0% 0.0% 100.0%
flops 29 x 32 x 14 929989394432 0.0% 0.0% 100.0%
flops 14 x 32 x 29 929989394432 0.0% 0.0% 100.0%
flops 29 x 32 x 29 963203301376 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693481172992 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753962643456 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.172206E+12 0.0% 0.0% 100.0%
flops max/rank 150.696064E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249788822 0.0% 0.0% 100.0%
number of processed stacks 98736 0.0% 0.0% 100.0%
average stack size 0.0 0.0 2529.9
marketing flops 7.174951E+12
-------------------------------------------------------------------------------
# multiplications 1140
max memory usage/rank 1.222156E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 703 408373.
MP_Allreduce 1821 23730.
MP_Sync 38
MP_Alltoall 77 24488229.
MP_SendRecv 2876 2171486.
MP_ISendRecv 1034 172620.
MP_Wait 1346
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.047 209.471 209.471
qs_energies 1 2.0 0.001 0.005 209.240 209.252
scf_env_do_scf 1 3.0 0.000 0.000 106.338 106.338
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.397 105.406
rebuild_ks_matrix 4 6.0 0.000 0.000 105.395 105.404
qs_ks_build_kohn_sham_matrix 4 7.0 0.055 0.063 105.395 105.404
hfx_ks_matrix 4 8.0 0.001 0.001 105.010 105.013
integrate_four_center 4 9.0 0.143 0.454 105.009 105.012
mp2_main 1 3.0 0.001 0.015 102.601 102.608
mp2_gpw_main 1 4.0 0.038 0.065 101.616 101.625
integrate_four_center_main 4 10.0 0.134 0.601 96.608 99.303
integrate_four_center_bin 263 11.0 96.473 98.853 96.473 98.853
init_scf_loop 1 4.0 0.000 0.000 92.206 92.206
mp2_ri_gpw_compute_in 1 5.0 0.082 0.140 74.743 75.800
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.011 54.486 55.548
mp2_eri_3c_integrate_gpw 91 7.0 0.144 0.156 42.208 47.293
integrate_v_rspace 95 8.0 0.398 0.569 28.596 33.475
pw_transfer 2240 10.6 0.146 0.165 29.827 30.175
fft_wrap_pw1pw2 1868 11.4 0.018 0.020 28.859 29.227
grid_integrate_task_list 95 9.0 23.878 28.958 23.878 28.958
mp2_ri_gpw_compute_en 1 5.0 0.055 0.065 26.708 28.425
fft_wrap_pw1pw2_100 730 12.4 1.244 1.375 26.583 26.933
ao_to_mo_and_store_B_mult_1 91 7.0 10.593 26.927 10.593 26.927
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.833 1.875 25.013 25.025
get_2c_integrals 1 6.0 0.011 0.052 20.136 20.187
compute_2c_integrals 1 7.0 0.003 0.005 19.107 19.130
compute_2c_integrals_loop_lm 1 8.0 0.002 0.007 18.747 18.994
mp2_eri_2c_integrate_gpw 1 9.0 1.754 1.860 18.745 18.992
fft3d_s 1823 13.4 18.399 18.745 18.412 18.758
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.130 14.130
calculate_wavefunction 91 8.0 2.012 2.034 9.732 9.954
mp2_ri_gpw_compute_en_expansio 172 7.0 0.558 0.590 8.776 9.179
potential_pw2rs 186 10.0 0.033 0.035 8.607 9.063
local_gemm 172 8.0 8.217 8.601 8.217 8.601
mp2_ri_gpw_compute_en_comm 22 7.0 0.488 0.504 8.021 8.525
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.174 8.452
calc_potential_gpw 182 9.5 0.002 0.002 7.886 8.119
collocate_single_gaussian 91 10.0 0.018 0.044 7.820 8.088
mp_sendrecv_dm3 2068 8.0 6.084 6.562 6.084 6.562
mp2_ri_gpw_compute_en_ener 172 7.0 6.339 6.417 6.339 6.417
mp_sync 38 10.4 3.109 6.051 3.109 6.051
pw_gather_s 912 13.2 4.914 5.254 4.914 5.254
pw_scatter_s 910 13.7 3.927 4.198 3.927 4.198
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.610659, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1509.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/03/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 29.277748E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 5055360 0.0% 0.0% 100.0%
average stack size 0.0 0.0 29.1
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 451.465216E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62385.
MP_Allreduce 10249 271.
MP_Sync 580
MP_Alltoall 2083 944707.
MP_SendRecv 22610 5520.
MP_ISendRecv 22610 5520.
MP_Wait 37876
MP_comm_split 50
MP_ISend 20771 42672.
MP_IRecv 20771 42672.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.040 0.084 54.416 54.418
qs_mol_dyn_low 1 2.0 0.003 0.004 53.972 53.980
qs_forces 11 3.9 0.002 0.002 53.911 53.911
qs_energies 11 4.9 0.001 0.001 52.407 52.421
scf_env_do_scf 11 5.9 0.000 0.001 46.302 46.302
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 44.040 44.041
dbcsr_multiply_generic 2286 12.5 0.094 0.097 34.458 35.019
qs_scf_new_mos 108 7.5 0.000 0.001 33.377 33.659
qs_scf_loop_do_ot 108 8.5 0.000 0.001 33.377 33.659
ot_scf_mini 108 9.5 0.002 0.002 31.658 31.854
multiply_cannon 2286 13.5 0.187 0.196 26.218 28.245
multiply_cannon_loop 2286 14.5 1.492 1.560 25.380 27.360
velocity_verlet 10 3.0 0.001 0.002 26.480 26.480
ot_mini 108 10.5 0.001 0.001 20.136 20.381
qs_ot_get_derivative 108 11.5 0.001 0.001 17.150 17.358
mp_waitall_1 245248 16.5 9.033 15.648 9.033 15.648
multiply_cannon_metrocomm3 54864 15.5 0.067 0.073 6.117 13.310
multiply_cannon_multrec 54864 15.5 4.149 6.436 7.501 10.853
rebuild_ks_matrix 119 8.3 0.000 0.000 8.257 8.406
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.257 8.406
mp_sum_l 7207 12.9 5.898 7.732 5.898 7.732
multiply_cannon_sync_h2d 54864 15.5 5.803 7.730 5.803 7.730
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.297 7.438
qs_ot_get_p 119 10.4 0.001 0.001 6.789 7.098
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 6.035 6.173
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.522 5.960
init_scf_run 11 5.9 0.000 0.001 4.849 4.849
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.849 4.849
sum_up_and_integrate 119 10.3 0.012 0.014 4.792 4.799
integrate_v_rspace 119 11.3 0.002 0.002 4.780 4.788
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.470 4.614
calculate_rho_elec 119 8.7 0.011 0.016 4.470 4.613
dbcsr_mm_accdrv_process 76910 16.1 1.188 1.833 3.274 4.583
qs_ot_p2m_diag 50 11.0 0.004 0.006 3.842 3.922
multiply_cannon_metrocomm1 54864 15.5 0.051 0.056 2.004 3.872
rs_pw_transfer 974 11.9 0.011 0.012 3.093 3.200
calculate_dm_sparse 119 9.5 0.000 0.000 3.043 3.167
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.921 3.116
apply_single 119 13.6 0.000 0.000 2.921 3.116
cp_dbcsr_syevd 50 12.0 0.002 0.003 2.957 2.957
ot_diis_step 108 11.5 0.006 0.006 2.681 2.681
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.679 2.680
jit_kernel_multiply 13 15.8 2.024 2.657 2.024 2.657
cp_fm_redistribute_end 50 14.0 2.434 2.652 2.440 2.654
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.496 2.580
cp_fm_diag_elpa_base 50 14.0 0.211 2.558 0.212 2.566
density_rs2pw 119 9.7 0.004 0.004 2.473 2.552
calculate_first_density_matrix 1 7.0 0.001 0.001 2.539 2.544
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.282 2.284
acc_transpose_blocks 54864 15.5 0.232 0.252 1.773 2.254
init_scf_loop 11 6.9 0.000 0.000 2.237 2.237
wfi_extrapolate 11 7.9 0.001 0.001 2.231 2.231
potential_pw2rs 119 12.3 0.004 0.004 2.117 2.139
grid_integrate_task_list 119 12.3 2.027 2.119 2.027 2.119
make_m2s 4572 13.5 0.055 0.057 2.039 2.107
pw_transfer 1439 11.6 0.051 0.055 1.964 2.056
make_images 4572 14.5 0.134 0.140 1.956 2.022
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.949 1.998
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 1.889 1.984
mp_sum_d 4125 12.0 1.300 1.932 1.300 1.932
fft3d_ps 1201 14.6 0.360 0.466 1.669 1.758
mp_alltoall_d11v 2130 13.8 1.545 1.710 1.545 1.710
fft_wrap_pw1pw2_140 487 13.2 0.079 0.092 1.446 1.539
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.488 1.512
mp_waitany 12084 13.8 1.271 1.373 1.271 1.373
grid_collocate_task_list 119 9.7 1.289 1.349 1.289 1.349
dbcsr_dot_sd 1205 11.9 0.047 0.058 0.722 1.137
prepare_preconditioner 11 7.9 0.000 0.000 1.097 1.125
make_preconditioner 11 8.9 0.000 0.000 1.097 1.125
acc_transpose_blocks_kernels 54864 16.5 0.258 0.391 0.779 1.110
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=54.418000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=430.181818, yerr=1.113404
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 486.715392E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62664.
MP_Allreduce 10226 305.
MP_Sync 104
MP_Alltoall 2060 1329389.
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 0.025 0.074 39.341 39.346
qs_mol_dyn_low 1 2.0 0.005 0.017 38.983 38.990
qs_forces 11 3.9 0.002 0.008 38.856 38.857
qs_energies 11 4.9 0.006 0.034 37.167 37.171
scf_env_do_scf 11 5.9 0.001 0.002 31.873 31.873
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 29.322 29.322
dbcsr_multiply_generic 2286 12.5 0.100 0.104 21.424 21.785
qs_scf_new_mos 108 7.5 0.001 0.001 20.150 20.390
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.149 20.389
ot_scf_mini 108 9.5 0.003 0.004 19.244 19.420
velocity_verlet 10 3.0 0.005 0.023 18.466 18.469
multiply_cannon 2286 13.5 0.209 0.219 16.292 17.897
multiply_cannon_loop 2286 14.5 0.897 0.977 15.146 16.505
ot_mini 108 10.5 0.001 0.001 11.840 12.077
mp_waitall_1 200699 16.5 5.801 10.869 5.801 10.869
qs_ot_get_derivative 108 11.5 0.001 0.002 9.380 9.558
multiply_cannon_metrocomm3 27432 15.5 0.067 0.070 4.119 9.529
multiply_cannon_multrec 27432 15.5 1.975 4.281 5.884 8.810
rebuild_ks_matrix 119 8.3 0.000 0.000 7.254 7.394
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.016 7.253 7.393
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.418 6.545
dbcsr_mm_accdrv_process 47894 16.0 3.102 5.118 3.838 5.686
qs_ot_get_p 119 10.4 0.001 0.001 4.473 4.707
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.601 4.453
sum_up_and_integrate 119 10.3 0.024 0.030 4.276 4.283
integrate_v_rspace 119 11.3 0.002 0.004 4.252 4.261
apply_preconditioner_dbcsr 119 12.6 0.000 0.001 3.051 4.123
apply_single 119 13.6 0.000 0.000 3.051 4.123
mp_sum_l 7207 12.9 2.096 4.024 2.096 4.024
init_scf_run 11 5.9 0.000 0.001 3.966 3.967
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.966 3.967
qs_rho_update_rho_low 119 7.7 0.001 0.002 3.739 3.770
calculate_rho_elec 119 8.7 0.021 0.024 3.738 3.769
qs_ot_p2m_diag 50 11.0 0.009 0.013 2.970 2.988
rs_pw_transfer 974 11.9 0.010 0.011 2.770 2.887
make_m2s 4572 13.5 0.053 0.054 2.640 2.887
make_images 4572 14.5 0.201 0.237 2.551 2.795
multiply_cannon_sync_h2d 27432 15.5 2.187 2.776 2.187 2.776
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.533 2.534
init_scf_loop 11 6.9 0.000 0.002 2.526 2.526
calculate_first_density_matrix 1 7.0 0.001 0.003 2.421 2.424
ot_diis_step 108 11.5 0.011 0.013 2.411 2.412
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.185 2.276
jit_kernel_multiply 12 16.2 0.684 2.251 0.684 2.251
density_rs2pw 119 9.7 0.004 0.004 2.090 2.206
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.174 2.175
calculate_dm_sparse 119 9.5 0.000 0.000 2.075 2.150
cp_fm_redistribute_end 50 14.0 1.796 2.146 1.800 2.147
cp_fm_diag_elpa_base 50 14.0 0.332 2.038 0.345 2.090
potential_pw2rs 119 12.3 0.006 0.006 2.025 2.038
grid_integrate_task_list 119 12.3 1.827 1.934 1.827 1.934
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.004 1.922 1.924
pw_transfer 1439 11.6 0.063 0.066 1.829 1.861
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 1.740 1.773
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.660 1.700
make_images_data 4572 15.5 0.045 0.051 1.196 1.683
prepare_preconditioner 11 7.9 0.000 0.000 1.601 1.627
make_preconditioner 11 8.9 0.000 0.001 1.601 1.627
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.500 1.563
hybrid_alltoall_any 4725 16.4 0.051 0.112 1.038 1.555
acc_transpose_blocks 27432 15.5 0.108 0.113 1.191 1.492
wfi_extrapolate 11 7.9 0.001 0.003 1.485 1.485
fft3d_ps 1201 14.6 0.500 0.554 1.448 1.477
mp_alltoall_d11v 2130 13.8 1.248 1.413 1.248 1.413
fft_wrap_pw1pw2_140 487 13.2 0.075 0.082 1.346 1.381
mp_allgather_i34 2286 14.5 0.588 1.367 0.588 1.367
grid_collocate_task_list 119 9.7 1.243 1.328 1.243 1.328
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.279 1.287
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.195 1.244
mp_sum_d 4125 12.0 0.618 1.073 0.618 1.073
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 0.981 0.983
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.893 0.906
make_images_sizes 4572 15.5 0.005 0.005 0.625 0.896
mp_alltoall_i44 4572 16.5 0.621 0.892 0.621 0.892
rs_pw_transfer_PW2RS_50 119 14.3 0.585 0.607 0.796 0.868
acc_transpose_blocks_kernels 27432 16.5 0.184 0.272 0.659 0.868
mp_alltoall_z22v 1201 16.6 0.729 0.807 0.729 0.807
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=39.346000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=463.272727, yerr=1.762793
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 521.019392E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62660.
MP_Allreduce 10225 303.
MP_Sync 104
MP_Alltoall 1821 1607811.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_comm_split 50
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.020 0.044 34.291 34.292
qs_mol_dyn_low 1 2.0 0.003 0.004 33.998 34.005
qs_forces 11 3.9 0.002 0.002 33.929 33.930
qs_energies 11 4.9 0.003 0.011 32.286 32.290
scf_env_do_scf 11 5.9 0.001 0.008 27.454 27.454
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 24.587 24.596
dbcsr_multiply_generic 2286 12.5 0.095 0.097 17.187 17.372
velocity_verlet 10 3.0 0.002 0.004 16.242 16.243
qs_scf_new_mos 108 7.5 0.001 0.001 16.101 16.130
qs_scf_loop_do_ot 108 8.5 0.001 0.004 16.101 16.130
ot_scf_mini 108 9.5 0.002 0.003 15.330 15.352
multiply_cannon 2286 13.5 0.198 0.206 13.337 14.237
multiply_cannon_loop 2286 14.5 0.637 0.668 12.451 13.514
ot_mini 108 10.5 0.001 0.001 9.419 9.444
qs_ot_get_derivative 108 11.5 0.001 0.001 7.837 7.856
multiply_cannon_multrec 18288 15.5 1.916 2.894 6.806 7.177
rebuild_ks_matrix 119 8.3 0.000 0.000 6.696 6.724
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.015 6.696 6.724
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.922 5.948
dbcsr_mm_accdrv_process 38222 16.0 4.119 5.424 4.808 5.682
mp_waitall_1 158411 16.6 3.214 4.977 3.214 4.977
sum_up_and_integrate 119 10.3 0.031 0.032 4.276 4.282
integrate_v_rspace 119 11.3 0.002 0.003 4.245 4.255
qs_ot_get_p 119 10.4 0.001 0.002 3.673 3.712
init_scf_run 11 5.9 0.000 0.001 3.607 3.608
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.607 3.608
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.970 3.605
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.480 3.484
calculate_rho_elec 119 8.7 0.030 0.031 3.479 3.484
multiply_cannon_metrocomm3 18288 15.5 0.044 0.045 1.589 3.185
init_scf_loop 11 6.9 0.001 0.004 2.842 2.867
rs_pw_transfer 974 11.9 0.009 0.009 2.647 2.762
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.421 2.431
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.089 2.426
apply_single 119 13.6 0.000 0.000 2.089 2.426
calculate_first_density_matrix 1 7.0 0.001 0.004 2.326 2.327
make_m2s 4572 13.5 0.045 0.046 2.068 2.260
make_images 4572 14.5 0.192 0.202 1.982 2.173
density_rs2pw 119 9.7 0.004 0.004 2.055 2.162
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.134 2.135
potential_pw2rs 119 12.3 0.007 0.008 2.035 2.052
prepare_preconditioner 11 7.9 0.000 0.000 1.971 1.974
make_preconditioner 11 8.9 0.000 0.001 1.971 1.974
pw_transfer 1439 11.6 0.063 0.066 1.910 1.931
calculate_dm_sparse 119 9.5 0.000 0.000 1.900 1.915
grid_integrate_task_list 119 12.3 1.799 1.908 1.799 1.908
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.821 1.903
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.842 1.845
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.820 1.843
cp_fm_redistribute_end 50 14.0 1.368 1.812 1.369 1.813
mp_sum_l 7207 12.9 1.398 1.768 1.398 1.768
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.759 1.765
cp_fm_diag_elpa_base 50 14.0 0.423 1.720 0.439 1.763
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.746 1.748
jit_kernel_multiply 9 16.0 0.637 1.638 0.637 1.638
multiply_cannon_sync_h2d 18288 15.5 1.327 1.576 1.327 1.576
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.556 1.568
ot_diis_step 108 11.5 0.011 0.012 1.538 1.538
fft3d_ps 1201 14.6 0.510 0.528 1.508 1.532
fft_wrap_pw1pw2_140 487 13.2 0.085 0.089 1.351 1.370
grid_collocate_task_list 119 9.7 1.202 1.273 1.202 1.273
acc_transpose_blocks 18288 15.5 0.076 0.077 1.243 1.271
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.230 1.235
wfi_extrapolate 11 7.9 0.001 0.001 1.228 1.228
make_images_data 4572 15.5 0.045 0.049 0.940 1.146
hybrid_alltoall_any 4725 16.4 0.055 0.114 0.807 1.020
mp_alltoall_d11v 2130 13.8 0.827 0.982 0.827 0.982
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.955 0.971
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 0.960 0.962
mp_alltoall_z22v 1201 16.6 0.819 0.894 0.819 0.894
cp_fm_cholesky_invert 11 10.9 0.865 0.870 0.865 0.870
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.848 0.851
acc_transpose_blocks_kernels 18288 16.5 0.214 0.223 0.801 0.818
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.657 0.760
mp_allgather_i34 2286 14.5 0.347 0.753 0.347 0.753
dbcsr_complete_redistribute 329 12.2 0.095 0.116 0.578 0.689
mp_waitany 9880 13.7 0.572 0.689 0.572 0.689
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=34.292000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=495.363636, yerr=1.366663
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/06/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 114.044384E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3805952 0.0% 0.0% 100.0%
average stack size 0.0 0.0 38.6
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 542.597120E+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 3672 62659.
MP_Allreduce 10224 344.
MP_Sync 104
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 0.019 0.050 37.680 37.681
qs_mol_dyn_low 1 2.0 0.003 0.004 37.410 37.417
qs_forces 11 3.9 0.002 0.002 37.347 37.348
qs_energies 11 4.9 0.002 0.005 35.522 35.528
scf_env_do_scf 11 5.9 0.001 0.002 30.156 30.158
scf_env_do_scf_inner_loop 108 6.5 0.002 0.008 26.306 26.307
dbcsr_multiply_generic 2286 12.5 0.100 0.103 19.188 19.305
velocity_verlet 10 3.0 0.001 0.002 18.990 18.992
qs_scf_new_mos 108 7.5 0.001 0.001 17.510 17.559
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.510 17.559
ot_scf_mini 108 9.5 0.002 0.003 16.530 16.578
multiply_cannon 2286 13.5 0.227 0.269 15.013 15.458
multiply_cannon_loop 2286 14.5 0.937 0.980 13.891 14.303
ot_mini 108 10.5 0.001 0.001 9.981 10.048
multiply_cannon_multrec 27432 15.5 2.392 3.096 8.858 9.182
qs_ot_get_derivative 108 11.5 0.001 0.001 8.099 8.149
dbcsr_mm_accdrv_process 47916 15.9 5.462 7.463 6.371 7.674
rebuild_ks_matrix 119 8.3 0.000 0.000 6.917 6.967
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.917 6.966
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.097 6.139
sum_up_and_integrate 119 10.3 0.035 0.038 4.127 4.137
integrate_v_rspace 119 11.3 0.002 0.002 4.092 4.103
init_scf_run 11 5.9 0.000 0.001 3.978 3.978
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.978 3.978
qs_ot_get_p 119 10.4 0.004 0.027 3.877 3.956
init_scf_loop 11 6.9 0.001 0.005 3.824 3.827
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.645 3.677
calculate_rho_elec 119 8.7 0.040 0.046 3.644 3.676
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.957 3.349
mp_waitall_1 137007 16.6 2.405 3.203 2.405 3.203
prepare_preconditioner 11 7.9 0.000 0.000 2.879 2.887
make_preconditioner 11 8.9 0.000 0.002 2.879 2.887
make_m2s 4572 13.5 0.055 0.056 2.629 2.823
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.476 2.807
make_images 4572 14.5 0.273 0.335 2.521 2.714
rs_pw_transfer 974 11.9 0.009 0.009 2.463 2.565
calculate_first_density_matrix 1 7.0 0.001 0.004 2.498 2.500
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.351 2.361
calculate_dm_sparse 119 9.5 0.000 0.000 2.186 2.246
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.088 2.214
apply_single 119 13.6 0.000 0.000 2.088 2.214
density_rs2pw 119 9.7 0.004 0.004 2.113 2.211
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.135 2.159
pw_transfer 1439 11.6 0.063 0.066 2.005 2.047
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.009 2.010
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.976 1.978
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.915 1.961
potential_pw2rs 119 12.3 0.008 0.009 1.880 1.892
grid_integrate_task_list 119 12.3 1.834 1.888 1.834 1.888
jit_kernel_multiply 9 15.7 0.848 1.852 0.848 1.852
ot_diis_step 108 11.5 0.012 0.012 1.841 1.841
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.762 1.775
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.687 1.688
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.865 1.666
cp_fm_redistribute_end 50 14.0 1.110 1.646 1.112 1.648
fft3d_ps 1201 14.6 0.536 0.587 1.597 1.631
cp_fm_diag_elpa_base 50 14.0 0.511 1.563 0.531 1.604
fft_wrap_pw1pw2_140 487 13.2 0.084 0.093 1.502 1.551
acc_transpose_blocks 27432 15.5 0.112 0.115 1.519 1.540
mp_sum_l 7207 12.9 1.141 1.465 1.141 1.465
wfi_extrapolate 11 7.9 0.001 0.001 1.427 1.427
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.336 1.346
make_images_data 4572 15.5 0.045 0.048 1.081 1.306
grid_collocate_task_list 119 9.7 1.220 1.302 1.220 1.302
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.188 1.203
dbcsr_complete_redistribute 329 12.2 0.121 0.149 0.893 1.172
hybrid_alltoall_any 4725 16.4 0.062 0.152 0.933 1.171
cp_fm_upper_to_full 72 13.5 0.810 1.127 0.810 1.127
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.092 1.093
multiply_cannon_sync_h2d 27432 15.5 0.979 1.054 0.979 1.054
mp_alltoall_d11v 2130 13.8 0.902 1.004 0.902 1.004
mp_alltoall_z22v 1201 16.6 0.944 0.982 0.944 0.982
cp_fm_cholesky_invert 11 10.9 0.950 0.954 0.950 0.954
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.913 0.918
acc_transpose_blocks_kernels 27432 16.5 0.271 0.279 0.894 0.908
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.625 0.893
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.793 0.871
mp_alltoall_i22 627 13.8 0.457 0.756 0.457 0.756
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=37.681000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=516.000000, yerr=3.045115
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/07/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 117.977176E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1384136 0.0% 0.0% 100.0%
average stack size 0.0 0.0 106.2
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 598.999040E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10224 344.
MP_Sync 104
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_comm_split 50
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 0.015 0.030 30.265 30.265
qs_mol_dyn_low 1 2.0 0.003 0.004 30.086 30.094
qs_forces 11 3.9 0.002 0.002 30.027 30.029
qs_energies 11 4.9 0.002 0.002 28.270 28.274
scf_env_do_scf 11 5.9 0.001 0.002 23.176 23.176
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 20.316 20.316
velocity_verlet 10 3.0 0.002 0.002 15.511 15.513
dbcsr_multiply_generic 2286 12.5 0.093 0.095 13.116 13.214
qs_scf_new_mos 108 7.5 0.001 0.001 12.202 12.234
qs_scf_loop_do_ot 108 8.5 0.001 0.001 12.201 12.234
ot_scf_mini 108 9.5 0.002 0.002 11.478 11.508
multiply_cannon 2286 13.5 0.230 0.237 10.003 10.649
multiply_cannon_loop 2286 14.5 0.331 0.342 8.993 9.192
ot_mini 108 10.5 0.001 0.001 6.205 6.236
rebuild_ks_matrix 119 8.3 0.000 0.000 6.148 6.179
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.147 6.179
multiply_cannon_multrec 9144 15.5 1.605 1.849 5.869 6.121
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.483 5.512
qs_ot_get_derivative 108 11.5 0.001 0.001 4.837 4.866
dbcsr_mm_accdrv_process 12550 15.8 3.251 3.999 4.163 4.221
sum_up_and_integrate 119 10.3 0.037 0.041 3.828 3.833
integrate_v_rspace 119 11.3 0.003 0.003 3.791 3.796
init_scf_run 11 5.9 0.000 0.001 3.629 3.629
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.629 3.629
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.498 3.504
calculate_rho_elec 119 8.7 0.059 0.061 3.497 3.503
qs_ot_get_p 119 10.4 0.001 0.001 3.358 3.393
init_scf_loop 11 6.9 0.000 0.000 2.831 2.831
mp_waitall_1 115863 16.7 2.093 2.702 2.093 2.702
calculate_first_density_matrix 1 7.0 0.000 0.000 2.363 2.364
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.289 2.291
make_m2s 4572 13.5 0.035 0.036 2.166 2.287
make_images 4572 14.5 0.268 0.300 2.076 2.196
rs_pw_transfer 974 11.9 0.008 0.008 1.982 2.096
prepare_preconditioner 11 7.9 0.000 0.000 2.057 2.063
make_preconditioner 11 8.9 0.000 0.000 2.057 2.062
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.057 2.058
density_rs2pw 119 9.7 0.003 0.004 1.878 1.992
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.934 1.964
grid_integrate_task_list 119 12.3 1.838 1.948 1.838 1.948
pw_transfer 1439 11.6 0.063 0.065 1.909 1.918
calculate_dm_sparse 119 9.5 0.000 0.000 1.865 1.883
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.820 1.829
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.773 1.775
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.682 1.696
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.687 1.689
cp_fm_redistribute_end 50 14.0 0.837 1.649 0.838 1.650
cp_fm_diag_elpa_base 50 14.0 0.763 1.560 0.808 1.626
potential_pw2rs 119 12.3 0.010 0.010 1.566 1.572
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.556 1.566
jit_kernel_multiply 8 16.0 0.874 1.503 0.874 1.503
fft3d_ps 1201 14.6 0.542 0.552 1.489 1.499
fft_wrap_pw1pw2_140 487 13.2 0.080 0.085 1.426 1.437
grid_collocate_task_list 119 9.7 1.275 1.362 1.275 1.362
ot_diis_step 108 11.5 0.012 0.013 1.348 1.348
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.303 1.319
make_images_data 4572 15.5 0.039 0.042 1.041 1.264
hybrid_alltoall_any 4725 16.4 0.062 0.175 0.997 1.249
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.215 1.249
apply_single 119 13.6 0.000 0.000 1.215 1.249
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.238 1.243
qs_energies_init_hamiltonians 11 5.9 0.003 0.020 1.238 1.240
wfi_extrapolate 11 7.9 0.001 0.001 1.214 1.214
mp_alltoall_d11v 2130 13.8 1.003 1.118 1.003 1.118
cp_fm_cholesky_invert 11 10.9 1.074 1.078 1.074 1.078
acc_transpose_blocks 9144 15.5 0.038 0.040 0.924 0.936
mp_allgather_i34 2286 14.5 0.341 0.921 0.341 0.921
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.870 0.918
multiply_cannon_metrocomm3 9144 15.5 0.019 0.019 0.379 0.877
mp_alltoall_z22v 1201 16.6 0.821 0.856 0.821 0.856
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.802 0.808
multiply_cannon_sync_h2d 9144 15.5 0.718 0.795 0.718 0.795
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.791 0.794
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.675 0.722
dbcsr_complete_redistribute 329 12.2 0.168 0.182 0.666 0.698
acc_transpose_blocks_kernels 9144 16.5 0.118 0.121 0.683 0.692
multiply_cannon_metrocomm1 9144 15.5 0.022 0.023 0.508 0.683
mp_sum_l 7207 12.9 0.520 0.631 0.520 0.631
make_images_sizes 4572 15.5 0.005 0.005 0.428 0.616
mp_alltoall_i44 4572 16.5 0.423 0.611 0.423 0.611
copy_dbcsr_to_fm 153 11.3 0.002 0.002 0.592 0.606
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=30.265000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=566.909091, yerr=4.077757
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/08/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 235.585836E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1388964 0.0% 0.0% 100.0%
average stack size 0.0 0.0 105.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 751.136768E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63729.
MP_Allreduce 10074 433.
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 0.043 0.097 42.768 42.768
qs_mol_dyn_low 1 2.0 0.003 0.005 42.468 42.474
qs_forces 11 3.9 0.002 0.002 42.050 42.051
qs_energies 11 4.9 0.005 0.006 40.063 40.067
scf_env_do_scf 11 5.9 0.001 0.002 34.075 34.076
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 25.873 25.875
velocity_verlet 10 3.0 0.002 0.002 23.979 24.029
dbcsr_multiply_generic 2286 12.5 0.099 0.100 18.081 18.225
qs_scf_new_mos 108 7.5 0.001 0.001 16.653 16.742
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.653 16.741
ot_scf_mini 108 9.5 0.002 0.002 15.542 15.636
multiply_cannon 2286 13.5 0.306 0.315 13.890 14.809
multiply_cannon_loop 2286 14.5 0.342 0.346 12.551 13.451
ot_mini 108 10.5 0.001 0.001 9.185 9.301
multiply_cannon_multrec 9144 15.5 3.398 4.960 8.590 8.662
init_scf_loop 11 6.9 0.000 0.000 8.170 8.174
prepare_preconditioner 11 7.9 0.000 0.000 7.196 7.211
make_preconditioner 11 8.9 0.000 0.000 7.196 7.211
rebuild_ks_matrix 119 8.3 0.000 0.000 7.059 7.203
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.058 7.202
qs_ot_get_derivative 108 11.5 0.001 0.001 7.097 7.192
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.769 7.086
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.387 6.517
dbcsr_mm_accdrv_process 12550 15.8 4.074 5.660 5.069 6.400
cp_fm_upper_to_full 72 14.2 3.211 4.575 3.211 4.575
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.892 3.945
calculate_rho_elec 119 8.7 0.118 0.121 3.892 3.945
init_scf_run 11 5.9 0.000 0.001 3.900 3.900
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.900 3.900
sum_up_and_integrate 119 10.3 0.064 0.066 3.823 3.831
integrate_v_rspace 119 11.3 0.003 0.003 3.759 3.766
mp_waitall_1 94719 16.7 2.747 3.704 2.747 3.704
qs_ot_get_p 119 10.4 0.001 0.001 3.519 3.658
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.569 2.985
dbcsr_complete_redistribute 329 12.2 0.288 0.292 2.013 2.850
make_m2s 4572 13.5 0.038 0.039 2.503 2.743
make_images 4572 14.5 0.351 0.382 2.382 2.622
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.235 2.521
apply_single 119 13.6 0.000 0.000 2.235 2.521
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.669 2.505
calculate_first_density_matrix 1 7.0 0.001 0.001 2.409 2.464
multiply_cannon_metrocomm3 9144 15.5 0.020 0.020 1.507 2.351
calculate_dm_sparse 119 9.5 0.000 0.000 2.271 2.318
mp_alltoall_i22 627 13.8 1.465 2.271 1.465 2.271
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.422 2.251
qs_ot_p2m_diag 50 11.0 0.043 0.043 2.183 2.184
pw_transfer 1439 11.6 0.066 0.066 2.146 2.148
grid_integrate_task_list 119 12.3 2.050 2.069 2.050 2.069
ot_diis_step 108 11.5 0.014 0.014 2.056 2.057
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.052 2.056
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.945 1.945
density_rs2pw 119 9.7 0.003 0.003 1.883 1.904
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.893 1.894
mp_sum_l 7207 12.9 1.153 1.862 1.153 1.862
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.780 1.831
qs_energies_init_hamiltonians 11 5.9 0.012 0.053 1.803 1.804
cp_fm_cholesky_invert 11 10.9 1.700 1.704 1.700 1.704
fft_wrap_pw1pw2_140 487 13.2 0.085 0.088 1.695 1.698
fft3d_ps 1201 14.6 0.569 0.580 1.689 1.691
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.613 1.638
hybrid_alltoall_any 4725 16.4 0.087 0.147 1.257 1.574
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.569 1.570
cp_fm_diag_elpa_base 50 14.0 1.422 1.475 1.567 1.568
make_images_data 4572 15.5 0.043 0.046 1.233 1.536
rs_pw_transfer 974 11.9 0.009 0.009 1.491 1.518
grid_collocate_task_list 119 9.7 1.482 1.503 1.482 1.503
jit_kernel_multiply 8 15.4 0.968 1.439 0.968 1.439
wfi_extrapolate 11 7.9 0.001 0.001 1.381 1.381
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.315 1.364
mp_alltoall_d11v 2130 13.8 1.312 1.361 1.312 1.361
potential_pw2rs 119 12.3 0.014 0.015 1.297 1.301
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.202 1.231
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.088 1.102
multiply_cannon_sync_h2d 9144 15.5 1.042 1.046 1.042 1.046
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.991 1.026
mp_alltoall_z22v 1201 16.6 0.985 0.999 0.985 0.999
qs_create_task_list 11 7.9 0.001 0.001 0.941 0.953
generate_qs_task_list 11 8.9 0.372 0.391 0.941 0.952
acc_transpose_blocks 9144 15.5 0.038 0.038 0.916 0.918
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.885 0.899
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=42.768000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=701.181818, yerr=16.948146
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/09/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 198.287135E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 8410880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 117.0
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 501.538816E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65372.
MP_Allreduce 9840 486.
MP_Sync 100
MP_Alltoall 1938 2231252.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_comm_split 48
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.031 84.516 84.517
qs_mol_dyn_low 1 2.0 0.003 0.005 83.872 84.284
qs_forces 11 3.9 0.003 0.003 83.794 83.794
qs_energies 11 4.9 0.002 0.002 80.922 80.937
scf_env_do_scf 11 5.9 0.000 0.001 72.003 72.005
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 66.115 66.116
dbcsr_multiply_generic 2055 12.4 0.107 0.111 52.353 52.680
qs_scf_new_mos 99 7.5 0.000 0.001 48.644 48.779
qs_scf_loop_do_ot 99 8.5 0.001 0.001 48.644 48.778
ot_scf_mini 99 9.5 0.002 0.002 46.228 46.372
multiply_cannon 2055 13.4 0.181 0.187 42.572 43.734
multiply_cannon_loop 2055 14.4 1.550 1.607 41.545 42.685
velocity_verlet 10 3.0 0.002 0.002 42.231 42.232
ot_mini 99 10.5 0.001 0.001 27.885 28.028
qs_ot_get_derivative 99 11.5 0.001 0.001 21.053 21.198
multiply_cannon_multrec 49320 15.4 12.137 12.958 17.194 18.025
rebuild_ks_matrix 110 8.3 0.000 0.001 14.822 14.963
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.822 14.962
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.045 13.170
mp_waitall_1 220248 16.4 11.700 12.816 11.700 12.816
multiply_cannon_sync_h2d 49320 15.4 9.978 10.656 9.978 10.656
qs_ot_get_p 110 10.4 0.001 0.001 9.798 9.897
multiply_cannon_metrocomm3 49320 15.4 0.079 0.084 6.832 8.122
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.170 7.780
apply_single 110 13.6 0.000 0.001 7.170 7.780
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 6.904 7.431
sum_up_and_integrate 110 10.3 0.036 0.042 7.298 7.310
integrate_v_rspace 110 11.3 0.002 0.003 7.261 7.282
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 7.171 7.248
init_scf_run 11 5.9 0.000 0.001 6.841 6.841
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.841 6.841
qs_rho_update_rho_low 110 7.6 0.000 0.001 6.486 6.622
calculate_rho_elec 110 8.6 0.020 0.024 6.486 6.621
ot_diis_step 99 11.5 0.006 0.006 6.559 6.560
qs_ot_p2m_diag 48 11.0 0.012 0.019 6.498 6.534
mp_sum_l 6514 12.8 5.122 6.075 5.122 6.075
init_scf_loop 11 6.9 0.000 0.000 5.858 5.858
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.462 5.463
dbcsr_mm_accdrv_process 87628 16.1 2.102 2.221 4.935 5.231
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.933 4.934
cp_fm_redistribute_end 48 14.0 4.300 4.900 4.304 4.902
cp_fm_diag_elpa_base 48 14.0 0.591 4.744 0.595 4.776
make_m2s 4110 13.4 0.062 0.067 4.206 4.323
rs_pw_transfer 902 11.9 0.011 0.013 4.103 4.249
make_images 4110 14.4 0.179 0.194 4.109 4.229
wfi_extrapolate 11 7.9 0.001 0.001 4.088 4.088
calculate_dm_sparse 110 9.5 0.000 0.001 3.796 3.892
multiply_cannon_metrocomm1 49320 15.4 0.060 0.063 2.715 3.821
density_rs2pw 110 9.6 0.004 0.004 3.582 3.737
prepare_preconditioner 11 7.9 0.000 0.000 3.618 3.638
make_preconditioner 11 8.9 0.000 0.000 3.618 3.638
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 3.589 3.593
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.393 3.437
grid_integrate_task_list 110 12.3 3.258 3.401 3.258 3.401
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.267 3.340
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.220 3.264
pw_transfer 1331 11.6 0.054 0.063 3.127 3.203
fft_wrap_pw1pw2 1111 12.6 0.007 0.008 3.040 3.119
potential_pw2rs 110 12.3 0.006 0.007 2.847 2.888
mp_alltoall_d11v 2046 13.8 2.194 2.837 2.194 2.837
calculate_first_density_matrix 1 7.0 0.001 0.002 2.643 2.645
fft3d_ps 1111 14.6 0.753 0.838 2.578 2.644
fft_wrap_pw1pw2_140 451 13.1 0.168 0.188 2.500 2.578
jit_kernel_multiply 13 15.9 2.547 2.564 2.547 2.564
acc_transpose_blocks 49320 15.4 0.226 0.239 2.213 2.291
mp_waitany 14300 13.8 1.886 2.203 1.886 2.203
grid_collocate_task_list 110 9.6 2.088 2.180 2.088 2.180
mp_sum_d 3879 11.9 1.478 2.062 1.478 2.062
make_images_data 4110 15.4 0.043 0.045 1.886 2.017
cp_fm_cholesky_invert 11 10.9 1.948 1.953 1.948 1.953
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.920 1.939
hybrid_alltoall_any 4261 16.3 0.081 0.479 1.648 1.912
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.666 1.694
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=84.517000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=476.636364, yerr=2.226809
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/10/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 390.715586E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 5019072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 196.1
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 587.325440E+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 3521 65587.
MP_Allreduce 9839 562.
MP_Sync 100
MP_Alltoall 1717 1513085.
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 0.060 0.133 71.171 71.173
qs_mol_dyn_low 1 2.0 0.004 0.008 70.682 70.700
qs_forces 11 3.9 0.002 0.002 70.610 70.611
qs_energies 11 4.9 0.003 0.009 67.197 67.204
scf_env_do_scf 11 5.9 0.001 0.001 58.139 58.142
scf_env_do_scf_inner_loop 99 6.5 0.002 0.008 49.883 49.885
dbcsr_multiply_generic 2055 12.4 0.115 0.127 38.679 38.921
velocity_verlet 10 3.0 0.001 0.002 37.167 37.169
qs_scf_new_mos 99 7.5 0.001 0.001 33.709 33.857
qs_scf_loop_do_ot 99 8.5 0.001 0.002 33.709 33.857
multiply_cannon 2055 13.4 0.223 0.244 31.362 32.725
ot_scf_mini 99 9.5 0.003 0.004 32.013 32.139
multiply_cannon_loop 2055 14.4 0.925 0.953 29.906 30.943
ot_mini 99 10.5 0.001 0.001 18.831 18.958
multiply_cannon_multrec 24660 15.4 7.568 9.078 13.826 15.522
rebuild_ks_matrix 110 8.3 0.000 0.001 13.902 14.020
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 13.901 14.019
qs_ot_get_derivative 99 11.5 0.001 0.001 12.933 13.065
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.200 12.304
mp_waitall_1 176588 16.5 8.138 11.373 8.138 11.373
multiply_cannon_metrocomm3 24660 15.4 0.069 0.071 5.367 8.230
init_scf_loop 11 6.9 0.000 0.002 8.216 8.224
multiply_cannon_sync_h2d 24660 15.4 7.005 7.954 7.005 7.954
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.575 7.307
apply_single 110 13.6 0.000 0.001 6.575 7.307
sum_up_and_integrate 110 10.3 0.052 0.059 6.728 6.760
integrate_v_rspace 110 11.3 0.002 0.002 6.675 6.710
qs_ot_get_p 110 10.4 0.001 0.002 6.442 6.644
init_scf_run 11 5.9 0.000 0.001 6.565 6.566
scf_env_initial_rho_setup 11 6.9 0.001 0.002 6.565 6.566
dbcsr_mm_accdrv_process 52282 16.1 4.661 5.354 6.099 6.436
prepare_preconditioner 11 7.9 0.000 0.000 6.140 6.166
make_preconditioner 11 8.9 0.000 0.002 6.140 6.166
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.942 5.951
calculate_rho_elec 110 8.6 0.039 0.047 5.941 5.950
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.705 5.867
ot_diis_step 99 11.5 0.010 0.010 5.836 5.836
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.799 5.475
make_m2s 4110 13.4 0.057 0.060 4.489 5.059
make_images 4110 14.4 0.403 0.448 4.378 4.947
qs_ot_p2m_diag 48 11.0 0.028 0.044 4.358 4.389
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.897 3.898
pw_transfer 1331 11.6 0.065 0.071 3.530 3.687
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.602 3.605
wfi_extrapolate 11 7.9 0.001 0.001 3.597 3.597
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.423 3.580
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.415 3.482
density_rs2pw 110 9.6 0.004 0.004 3.230 3.422
grid_integrate_task_list 110 12.3 3.144 3.320 3.144 3.320
rs_pw_transfer 902 11.9 0.012 0.014 3.123 3.303
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.283 3.286
cp_fm_redistribute_end 48 14.0 2.450 3.250 2.453 3.252
cp_fm_diag_elpa_base 48 14.0 0.763 3.099 0.794 3.189
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.102 3.163
make_images_data 4110 15.4 0.047 0.051 2.527 3.109
calculate_dm_sparse 110 9.5 0.001 0.001 3.046 3.075
hybrid_alltoall_any 4261 16.3 0.102 0.441 2.196 2.992
cp_fm_cholesky_invert 11 10.9 2.966 2.974 2.966 2.974
mp_sum_l 6514 12.8 2.215 2.951 2.215 2.951
fft_wrap_pw1pw2_140 451 13.1 0.199 0.220 2.793 2.947
fft3d_ps 1111 14.6 1.075 1.275 2.782 2.921
calculate_first_density_matrix 1 7.0 0.001 0.005 2.876 2.880
potential_pw2rs 110 12.3 0.008 0.008 2.580 2.616
grid_collocate_task_list 110 9.6 2.120 2.303 2.120 2.303
mp_alltoall_d11v 2046 13.8 1.893 2.124 1.893 2.124
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.025 2.053
jit_kernel_multiply 10 16.3 1.086 2.015 1.086 2.015
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.927 1.928
mp_allgather_i34 2055 14.4 0.784 1.927 0.784 1.927
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.867 1.888
dbcsr_complete_redistribute 325 12.2 0.287 0.477 1.364 1.708
multiply_cannon_metrocomm4 22605 15.4 0.072 0.075 0.780 1.671
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.595 1.603
acc_transpose_blocks 24660 15.4 0.110 0.113 1.540 1.565
mp_alltoall_z22v 1111 16.6 1.433 1.525 1.433 1.525
cp_fm_cholesky_decompose 22 10.9 1.512 1.519 1.512 1.519
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.369 1.477
mp_irecv_dv 57340 16.2 0.658 1.461 0.658 1.461
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=71.173000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=555.181818, yerr=7.158189
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/11/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 404.681598E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 3346752 0.0% 0.0% 100.0%
average stack size 0.0 0.0 294.1
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 660.221952E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65578.
MP_Allreduce 9838 559.
MP_Sync 100
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_comm_split 48
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.088 63.208 63.210
qs_mol_dyn_low 1 2.0 0.004 0.013 62.778 62.787
qs_forces 11 3.9 0.002 0.002 62.687 62.689
qs_energies 11 4.9 0.001 0.002 59.436 59.441
scf_env_do_scf 11 5.9 0.000 0.001 51.368 51.368
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 42.002 42.002
velocity_verlet 10 3.0 0.001 0.002 34.092 34.094
dbcsr_multiply_generic 2055 12.4 0.109 0.113 29.357 29.727
qs_scf_new_mos 99 7.5 0.001 0.001 26.874 26.986
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.874 26.985
ot_scf_mini 99 9.5 0.002 0.003 25.610 25.737
multiply_cannon 2055 13.4 0.212 0.228 22.341 23.664
multiply_cannon_loop 2055 14.4 0.614 0.634 21.015 22.182
ot_mini 99 10.5 0.001 0.001 14.290 14.422
rebuild_ks_matrix 110 8.3 0.000 0.000 12.595 12.763
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.594 12.762
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.115 11.266
multiply_cannon_multrec 16440 15.4 3.918 5.039 9.683 10.816
mp_waitall_1 139946 16.5 7.461 10.693 7.461 10.693
qs_ot_get_derivative 99 11.5 0.001 0.001 9.796 9.923
init_scf_loop 11 6.9 0.000 0.000 9.325 9.326
multiply_cannon_metrocomm3 16440 15.4 0.043 0.044 4.369 7.544
prepare_preconditioner 11 7.9 0.000 0.000 7.518 7.535
make_preconditioner 11 8.9 0.000 0.000 7.518 7.535
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.859 7.203
sum_up_and_integrate 110 10.3 0.060 0.061 6.721 6.737
integrate_v_rspace 110 11.3 0.002 0.003 6.660 6.677
qs_ot_get_p 110 10.4 0.001 0.001 6.363 6.522
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.836 5.850
calculate_rho_elec 110 8.6 0.058 0.058 5.836 5.849
dbcsr_mm_accdrv_process 34862 16.1 4.586 5.275 5.619 5.816
init_scf_run 11 5.9 0.000 0.001 5.604 5.604
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.604 5.604
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.988 5.414
apply_single 110 13.6 0.000 0.000 4.987 5.414
make_m2s 4110 13.4 0.050 0.052 4.527 4.883
make_images 4110 14.4 0.396 0.513 4.410 4.764
qs_ot_p2m_diag 48 11.0 0.042 0.045 4.485 4.495
ot_diis_step 99 11.5 0.011 0.011 4.436 4.437
multiply_cannon_sync_h2d 16440 15.4 3.678 4.349 3.678 4.349
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.087 4.088
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.305 3.949
pw_transfer 1331 11.6 0.064 0.071 3.410 3.441
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.430 3.432
grid_integrate_task_list 110 12.3 3.188 3.400 3.188 3.400
cp_fm_redistribute_end 48 14.0 2.134 3.386 2.137 3.387
cp_fm_diag_elpa_base 48 14.0 1.181 3.220 1.243 3.328
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.303 3.314
rs_pw_transfer 902 11.9 0.010 0.011 3.075 3.233
density_rs2pw 110 9.6 0.004 0.004 3.056 3.214
cp_fm_cholesky_invert 11 10.9 3.118 3.126 3.118 3.126
wfi_extrapolate 11 7.9 0.001 0.001 3.094 3.094
make_images_data 4110 15.4 0.043 0.048 2.568 3.052
hybrid_alltoall_any 4261 16.3 0.105 0.372 2.230 2.868
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 2.836 2.838
fft_wrap_pw1pw2_140 451 13.1 0.209 0.212 2.725 2.736
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.628 2.697
fft3d_ps 1111 14.6 1.065 1.087 2.620 2.628
calculate_dm_sparse 110 9.5 0.001 0.001 2.570 2.602
mp_sum_l 6514 12.8 1.908 2.545 1.908 2.545
potential_pw2rs 110 12.3 0.010 0.010 2.522 2.540
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.455 2.499
mp_alltoall_d11v 2046 13.8 1.939 2.461 1.939 2.461
calculate_first_density_matrix 1 7.0 0.001 0.001 2.418 2.420
multiply_cannon_metrocomm4 14385 15.4 0.045 0.048 0.872 2.349
grid_collocate_task_list 110 9.6 2.181 2.315 2.181 2.315
mp_irecv_dv 48980 15.7 0.803 2.229 0.803 2.229
dbcsr_complete_redistribute 325 12.2 0.325 0.365 1.606 2.065
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 2.009 2.011
cp_fm_upper_to_full 70 13.6 1.392 1.875 1.392 1.875
cp_fm_cholesky_decompose 22 10.9 1.830 1.848 1.830 1.848
mp_allgather_i34 2055 14.4 0.650 1.742 0.650 1.742
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.685 1.701
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.491 1.502
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.357 1.492
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.035 1.484
rs_gather_matrices 110 12.3 0.138 0.151 0.895 1.398
mp_waitany 17072 13.8 1.231 1.397 1.231 1.397
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.369 1.380
mp_alltoall_z22v 1111 16.6 1.341 1.357 1.341 1.357
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=63.210000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=625.272727, yerr=8.812088
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/12/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 601.317074E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 4916280 0.0% 0.0% 100.0%
average stack size 0.0 0.0 200.2
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 724.967424E+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 3521 65576.
MP_Allreduce 9838 600.
MP_Sync 100
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_comm_split 48
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 0.045 0.111 64.788 64.789
qs_mol_dyn_low 1 2.0 0.003 0.004 64.448 64.458
qs_forces 11 3.9 0.013 0.014 64.383 64.384
qs_energies 11 4.9 0.006 0.012 60.993 60.998
scf_env_do_scf 11 5.9 0.001 0.002 52.580 52.583
scf_env_do_scf_inner_loop 99 6.5 0.002 0.010 40.717 40.720
velocity_verlet 10 3.0 0.002 0.002 36.869 36.871
dbcsr_multiply_generic 2055 12.4 0.114 0.120 29.677 29.894
qs_scf_new_mos 99 7.5 0.001 0.001 26.242 26.372
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.241 26.372
ot_scf_mini 99 9.5 0.003 0.003 24.611 24.712
multiply_cannon 2055 13.4 0.245 0.265 22.625 23.489
multiply_cannon_loop 2055 14.4 0.887 0.908 21.063 21.646
ot_mini 99 10.5 0.001 0.001 14.171 14.298
multiply_cannon_multrec 24660 15.4 4.188 6.849 12.697 13.926
rebuild_ks_matrix 110 8.3 0.000 0.001 11.993 12.098
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 11.993 12.098
init_scf_loop 11 6.9 0.001 0.003 11.820 11.828
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.614 10.710
qs_ot_get_derivative 99 11.5 0.001 0.001 10.010 10.117
prepare_preconditioner 11 7.9 0.000 0.000 10.016 10.032
make_preconditioner 11 8.9 0.001 0.006 10.016 10.032
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.240 9.696
dbcsr_mm_accdrv_process 52304 16.0 6.817 8.262 8.361 9.334
mp_waitall_1 121746 16.5 4.347 6.518 4.347 6.518
sum_up_and_integrate 110 10.3 0.068 0.071 6.369 6.383
integrate_v_rspace 110 11.3 0.002 0.003 6.301 6.316
make_m2s 4110 13.4 0.060 0.062 5.431 5.946
make_images 4110 14.4 0.580 0.703 5.290 5.802
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.667 5.677
calculate_rho_elec 110 8.6 0.077 0.081 5.666 5.676
qs_ot_get_p 110 10.4 0.001 0.002 5.463 5.619
init_scf_run 11 5.9 0.000 0.001 5.594 5.595
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.594 5.595
cp_fm_upper_to_full 70 13.8 3.250 4.627 3.250 4.627
ot_diis_step 99 11.5 0.011 0.012 4.124 4.125
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.030 4.104
apply_single 110 13.6 0.000 0.000 4.030 4.104
dbcsr_complete_redistribute 325 12.2 0.416 0.462 2.688 3.832
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.621 3.636
grid_integrate_task_list 110 12.3 3.282 3.431 3.282 3.431
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.370 3.419
multiply_cannon_sync_h2d 24660 15.4 3.170 3.344 3.170 3.344
pw_transfer 1331 11.6 0.064 0.071 3.300 3.324
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.140 3.267
multiply_cannon_metrocomm3 24660 15.4 0.036 0.036 1.463 3.247
make_images_data 4110 15.4 0.047 0.052 2.724 3.245
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.194 3.222
hybrid_alltoall_any 4261 16.3 0.119 0.458 2.346 3.213
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.140 3.140
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.048 3.105
wfi_extrapolate 11 7.9 0.001 0.001 3.019 3.019
calculate_dm_sparse 110 9.5 0.001 0.001 2.939 2.973
density_rs2pw 110 9.6 0.004 0.004 2.812 2.940
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.767 2.880
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.865 2.867
cp_fm_cholesky_invert 11 10.9 2.819 2.827 2.819 2.827
mp_alltoall_i22 605 13.7 1.646 2.787 1.646 2.787
fft_wrap_pw1pw2_140 451 13.1 0.202 0.213 2.697 2.727
rs_pw_transfer 902 11.9 0.010 0.011 2.501 2.681
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.590 2.592
cp_fm_redistribute_end 48 14.0 1.296 2.565 1.297 2.566
cp_fm_diag_elpa_base 48 14.0 1.188 2.428 1.265 2.537
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.495 2.530
fft3d_ps 1111 14.6 1.067 1.101 2.498 2.512
calculate_first_density_matrix 1 7.0 0.001 0.005 2.486 2.489
grid_collocate_task_list 110 9.6 2.222 2.426 2.222 2.426
qs_energies_init_hamiltonians 11 5.9 0.012 0.014 2.311 2.311
jit_kernel_multiply 12 15.7 1.211 2.294 1.211 2.294
potential_pw2rs 110 12.3 0.012 0.013 2.140 2.148
mp_alltoall_d11v 2046 13.8 1.817 1.967 1.817 1.967
cp_fm_cholesky_decompose 22 10.9 1.712 1.761 1.712 1.761
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.700 1.733
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.601 1.705
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.625 1.637
mp_sum_l 6514 12.8 0.945 1.620 0.945 1.620
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.544 1.563
acc_transpose_blocks 24660 15.4 0.106 0.109 1.526 1.553
multiply_cannon_metrocomm4 20550 15.4 0.057 0.059 0.860 1.463
mp_irecv_dv 62702 16.1 0.762 1.391 0.762 1.391
mp_allgather_i34 2055 14.4 0.664 1.364 0.664 1.364
qs_env_update_s_mstruct 11 6.9 0.000 0.001 1.243 1.351
mp_waitany 13376 13.8 1.101 1.325 1.101 1.325
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=64.789000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=687.272727, yerr=7.782158
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/13/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 807.299199E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 1438408 0.0% 0.0% 100.0%
average stack size 0.0 0.0 684.2
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 823.365632E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65574.
MP_Allreduce 9838 640.
MP_Sync 100
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_comm_split 48
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 0.014 0.043 57.812 57.812
qs_mol_dyn_low 1 2.0 0.003 0.004 57.569 57.577
qs_forces 11 3.9 0.002 0.002 57.499 57.500
qs_energies 11 4.9 0.001 0.002 53.802 53.806
scf_env_do_scf 11 5.9 0.000 0.001 45.249 45.249
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 36.722 36.722
velocity_verlet 10 3.0 0.002 0.002 32.342 32.345
dbcsr_multiply_generic 2055 12.4 0.105 0.107 23.800 23.949
qs_scf_new_mos 99 7.5 0.001 0.001 22.122 22.171
qs_scf_loop_do_ot 99 8.5 0.001 0.001 22.121 22.171
ot_scf_mini 99 9.5 0.002 0.002 20.819 20.856
multiply_cannon 2055 13.4 0.251 0.262 17.726 19.078
multiply_cannon_loop 2055 14.4 0.321 0.336 16.224 16.502
rebuild_ks_matrix 110 8.3 0.000 0.000 11.807 11.828
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.807 11.828
ot_mini 99 10.5 0.001 0.001 11.033 11.057
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.517 10.537
mp_waitall_1 103326 16.6 6.568 8.694 6.568 8.694
multiply_cannon_multrec 8220 15.4 3.201 4.618 7.490 8.497
init_scf_loop 11 6.9 0.000 0.000 8.466 8.467
qs_ot_get_derivative 99 11.5 0.001 0.001 7.156 7.186
prepare_preconditioner 11 7.9 0.000 0.000 6.795 6.805
make_preconditioner 11 8.9 0.000 0.000 6.795 6.805
sum_up_and_integrate 110 10.3 0.079 0.080 6.441 6.459
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.367 6.455
integrate_v_rspace 110 11.3 0.003 0.004 6.361 6.381
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.866 5.879
calculate_rho_elec 110 8.6 0.115 0.115 5.866 5.878
qs_ot_get_p 110 10.4 0.001 0.001 5.740 5.781
init_scf_run 11 5.9 0.000 0.001 5.378 5.379
scf_env_initial_rho_setup 11 6.9 0.002 0.002 5.378 5.378
dbcsr_mm_accdrv_process 17442 15.9 2.920 3.687 4.159 5.040
make_m2s 4110 13.4 0.039 0.040 4.552 4.914
make_images 4110 14.4 0.639 0.696 4.422 4.781
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 3.149 4.607
qs_ot_p2m_diag 48 11.0 0.081 0.084 4.182 4.187
ot_diis_step 99 11.5 0.012 0.012 3.844 3.844
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.838 3.839
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.756 3.788
apply_single 110 13.6 0.000 0.000 3.756 3.788
grid_integrate_task_list 110 12.3 3.360 3.505 3.360 3.505
pw_transfer 1331 11.6 0.065 0.069 3.453 3.467
cp_fm_cholesky_invert 11 10.9 3.461 3.466 3.461 3.466
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.346 3.363
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.217 3.218
cp_fm_redistribute_end 48 14.0 0.823 3.175 0.829 3.177
cp_fm_diag_elpa_base 48 14.0 2.160 2.967 2.337 3.125
make_images_data 4110 15.4 0.038 0.043 2.528 3.080
density_rs2pw 110 9.6 0.004 0.004 2.909 3.076
multiply_cannon_sync_h2d 8220 15.4 2.902 2.989 2.902 2.989
hybrid_alltoall_any 4261 16.3 0.199 0.861 2.396 2.960
wfi_extrapolate 11 7.9 0.001 0.001 2.879 2.879
fft_wrap_pw1pw2_140 451 13.1 0.211 0.215 2.795 2.819
rs_pw_transfer 902 11.9 0.010 0.010 2.498 2.698
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.676 2.677
calculate_dm_sparse 110 9.5 0.001 0.001 2.639 2.673
fft3d_ps 1111 14.6 1.120 1.141 2.605 2.617
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.610 2.611
grid_collocate_task_list 110 9.6 2.327 2.538 2.327 2.538
calculate_first_density_matrix 1 7.0 0.001 0.001 2.404 2.406
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.216 2.228
potential_pw2rs 110 12.3 0.015 0.015 2.156 2.168
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.058 2.074
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.041 2.055
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.774 1.993
mp_alltoall_d11v 2046 13.8 1.786 1.977 1.786 1.977
cp_fm_cholesky_decompose 22 10.9 1.871 1.891 1.871 1.891
mp_allgather_i34 2055 14.4 0.685 1.823 0.685 1.823
dbcsr_complete_redistribute 325 12.2 0.552 0.566 1.590 1.695
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.522 1.642
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.629 1.633
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.545 1.558
jit_kernel_multiply 8 15.8 0.924 1.370 0.924 1.370
qs_create_task_list 11 7.9 0.001 0.001 1.236 1.334
generate_qs_task_list 11 8.9 0.379 0.448 1.235 1.333
mp_waitany 9240 13.8 1.103 1.313 1.103 1.313
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.257 1.283
mp_alltoall_z22v 1111 16.6 1.252 1.266 1.252 1.266
multiply_cannon_metrocomm4 6165 15.4 0.017 0.019 0.472 1.197
multiply_cannon_metrocomm1 8220 15.4 0.021 0.021 0.897 1.183
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=57.812000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=774.636364, yerr=10.020640
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/14/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 1.612391E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 1464624 0.0% 0.0% 100.0%
average stack size 0.0 0.0 672.0
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 1.340051E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67104.
MP_Allreduce 9672 819.
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 0.020 0.047 88.459 88.460
qs_mol_dyn_low 1 2.0 0.003 0.004 88.120 88.128
qs_forces 11 3.9 0.002 0.002 88.046 88.047
qs_energies 11 4.9 0.002 0.002 83.920 83.922
scf_env_do_scf 11 5.9 0.001 0.001 73.733 73.734
velocity_verlet 10 3.0 0.002 0.002 56.452 56.458
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 44.495 44.497
dbcsr_multiply_generic 2055 12.4 0.120 0.121 29.421 29.461
init_scf_loop 11 6.9 0.000 0.000 29.163 29.165
qs_scf_new_mos 99 7.5 0.001 0.001 27.595 27.623
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.594 27.622
prepare_preconditioner 11 7.9 0.000 0.000 27.187 27.194
make_preconditioner 11 8.9 0.000 0.000 27.187 27.194
make_full_inverse_cholesky 11 9.9 0.000 0.000 21.245 26.641
ot_scf_mini 99 9.5 0.002 0.002 25.807 25.831
multiply_cannon 2055 13.4 0.348 0.374 21.897 22.808
multiply_cannon_loop 2055 14.4 0.343 0.362 19.994 20.292
cp_fm_upper_to_full 70 14.2 12.739 18.387 12.739 18.387
ot_mini 99 10.5 0.001 0.001 14.085 14.108
rebuild_ks_matrix 110 8.3 0.001 0.001 13.479 13.520
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.014 13.478 13.520
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.253 12.290
dbcsr_complete_redistribute 325 12.2 1.050 1.080 7.624 10.914
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.554 9.845
multiply_cannon_multrec 8220 15.4 4.382 4.579 9.663 9.791
qs_ot_get_derivative 99 11.5 0.001 0.001 9.336 9.359
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.927 9.190
mp_waitall_1 84994 16.7 8.091 8.961 8.091 8.961
mp_alltoall_i22 605 13.7 5.549 8.855 5.549 8.855
sum_up_and_integrate 110 10.3 0.150 0.152 6.848 6.863
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.713 6.749
calculate_rho_elec 110 8.6 0.227 0.227 6.712 6.748
integrate_v_rspace 110 11.3 0.003 0.003 6.697 6.714
qs_ot_get_p 110 10.4 0.001 0.001 6.376 6.398
make_m2s 4110 13.4 0.043 0.044 5.670 6.234
make_images 4110 14.4 0.877 0.908 5.481 6.044
init_scf_run 11 5.9 0.000 0.001 5.892 5.892
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.891 5.891
cp_fm_cholesky_invert 11 10.9 5.830 5.835 5.830 5.835
dbcsr_mm_accdrv_process 11614 15.7 3.285 3.666 5.137 5.391
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 4.962 5.257
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.787 5.223
apply_single 110 13.6 0.000 0.000 4.787 5.223
ot_diis_step 99 11.5 0.018 0.038 4.723 4.723
qs_ot_p2m_diag 48 11.0 0.150 0.155 4.678 4.685
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.212 4.214
pw_transfer 1331 11.6 0.073 0.074 3.990 3.995
hybrid_alltoall_any 4261 16.3 0.255 0.552 3.135 3.967
multiply_cannon_sync_h2d 8220 15.4 3.953 3.960 3.953 3.960
make_images_data 4110 15.4 0.042 0.046 3.176 3.915
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 3.874 3.880
grid_integrate_task_list 110 12.3 3.677 3.748 3.677 3.748
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.679 3.680
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.555 3.555
cp_fm_diag_elpa_base 48 14.0 3.010 3.205 3.552 3.552
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.958 3.422
fft_wrap_pw1pw2_140 451 13.1 0.215 0.222 3.314 3.317
wfi_extrapolate 11 7.9 0.001 0.001 3.302 3.302
calculate_dm_sparse 110 9.5 0.001 0.001 3.257 3.281
density_rs2pw 110 9.6 0.004 0.004 3.149 3.172
fft3d_ps 1111 14.6 1.269 1.281 3.084 3.092
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.040 3.043
grid_collocate_task_list 110 9.6 2.643 2.667 2.643 2.667
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.507 2.518
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.478 2.488
calculate_first_density_matrix 1 7.0 0.001 0.001 2.478 2.480
rs_pw_transfer 902 11.9 0.010 0.011 2.284 2.338
mp_alltoall_d11v 2046 13.8 2.229 2.304 2.229 2.304
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.205 2.263
potential_pw2rs 110 12.3 0.021 0.022 2.209 2.215
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.098 2.211
cp_fm_cholesky_decompose 22 10.9 2.143 2.165 2.143 2.165
qs_create_task_list 11 7.9 0.008 0.009 1.918 1.966
generate_qs_task_list 11 8.9 0.741 0.797 1.910 1.958
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.837 1.862
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.835 1.839
jit_kernel_multiply 10 15.3 1.649 1.827 1.649 1.827
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.772 1.788
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=88.460000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1214.181818, yerr=44.984478
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420241154048 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528896499712 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514757E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755939872 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 627.945472E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175955098256
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4075 56898.
MP_Allreduce 11228 786.
MP_Sync 170
MP_Alltoall 2226 2955660.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_comm_split 83
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.031 207.222 207.222
qs_mol_dyn_low 1 2.0 0.003 0.004 206.862 206.876
qs_forces 11 3.9 0.003 0.003 206.768 206.769
qs_energies 11 4.9 0.001 0.002 201.213 201.228
scf_env_do_scf 11 5.9 0.001 0.001 184.589 184.593
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 162.925 162.927
dbcsr_multiply_generic 2507 12.6 0.180 0.185 124.718 125.654
velocity_verlet 10 3.0 0.001 0.002 124.664 124.666
qs_scf_new_mos 117 7.6 0.001 0.001 123.398 123.606
qs_scf_loop_do_ot 117 8.6 0.001 0.001 123.397 123.605
ot_scf_mini 117 9.6 0.003 0.003 116.768 116.982
multiply_cannon 2507 13.6 0.236 0.243 101.011 102.964
multiply_cannon_loop 2507 14.6 2.125 2.196 98.676 100.467
ot_mini 117 10.6 0.001 0.001 66.231 66.445
multiply_cannon_multrec 60168 15.6 33.047 34.985 41.517 43.469
qs_ot_get_derivative 117 11.6 0.001 0.002 41.436 41.648
rebuild_ks_matrix 128 8.3 0.001 0.001 33.664 33.974
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 33.664 33.973
mp_waitall_1 267128 16.5 28.931 31.544 28.931 31.544
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.306 30.590
multiply_cannon_sync_h2d 60168 15.6 27.157 29.201 27.157 29.201
qs_ot_get_p 128 10.4 0.001 0.001 28.625 28.956
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.219 25.027
apply_single 128 13.6 0.001 0.001 24.219 25.026
ot_diis_step 117 11.6 0.007 0.008 24.463 24.464
qs_ot_p2m_diag 83 11.4 0.077 0.091 21.656 21.711
init_scf_loop 11 6.9 0.000 0.000 21.578 21.579
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 19.448 19.645
cp_dbcsr_syevd 83 12.4 0.004 0.005 19.016 19.017
multiply_cannon_metrocomm3 60168 15.6 0.116 0.120 15.666 18.048
prepare_preconditioner 11 7.9 0.000 0.000 16.944 16.996
make_preconditioner 11 8.9 0.000 0.000 16.944 16.996
make_full_inverse_cholesky 11 9.9 0.000 0.000 16.155 16.337
cp_fm_diag_elpa 83 13.4 0.000 0.001 15.891 15.899
cp_fm_redistribute_end 83 14.4 12.547 15.796 12.562 15.800
cp_fm_diag_elpa_base 83 14.4 3.186 15.405 3.223 15.508
make_m2s 5014 13.6 0.106 0.115 14.309 14.771
make_images 5014 14.6 0.404 0.424 14.125 14.593
sum_up_and_integrate 128 10.3 0.089 0.108 14.374 14.390
integrate_v_rspace 128 11.3 0.003 0.004 14.285 14.305
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.164 13.263
calculate_rho_elec 128 8.7 0.045 0.063 13.163 13.263
init_scf_run 11 5.9 0.000 0.001 12.495 12.496
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.495 12.496
cp_fm_cholesky_invert 11 10.9 9.879 9.888 9.879 9.888
mp_sum_l 7870 13.0 8.136 9.384 8.136 9.384
wfi_extrapolate 11 7.9 0.001 0.001 9.211 9.211
calculate_dm_sparse 128 9.5 0.001 0.001 8.505 8.606
dbcsr_mm_accdrv_process 124484 16.2 3.331 3.485 8.028 8.571
multiply_cannon_metrocomm1 60168 15.6 0.089 0.094 6.242 8.276
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 7.962 8.093
make_images_data 5014 15.6 0.066 0.071 7.032 8.022
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.857 7.915
density_rs2pw 128 9.7 0.006 0.007 6.972 7.586
pw_transfer 1547 11.6 0.075 0.106 7.203 7.504
grid_integrate_task_list 128 12.3 7.126 7.490 7.126 7.490
hybrid_alltoall_any 5200 16.5 0.291 2.263 6.121 7.461
fft_wrap_pw1pw2 1291 12.7 0.010 0.013 6.999 7.271
rs_pw_transfer 1046 11.9 0.016 0.018 6.163 6.723
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.682 6.690
fft_wrap_pw1pw2_140 523 13.2 0.442 0.508 5.843 6.051
fft3d_ps 1291 14.7 2.094 2.572 5.796 6.015
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.787 5.879
mp_alltoall_d11v 2415 14.1 4.367 5.620 4.367 5.620
grid_collocate_task_list 128 9.7 4.726 5.056 4.726 5.056
cp_fm_cholesky_decompose 22 10.9 4.843 4.857 4.843 4.857
potential_pw2rs 128 12.3 0.009 0.010 4.789 4.829
mp_sum_d 4455 12.2 3.578 4.278 3.578 4.278
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=207.222000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=593.727273, yerr=6.165755
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 2.166472E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 5925696 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.6
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 832.749568E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2385600
MPI messages size (bytes):
total size 4.069300E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.705776E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70188 2295595008
32768 < size <= 131072 716032 54973693952
131072 < size <= 4194304 1363760 1386318135296
4194304 < size <= 16777216 153648 1453843137296
16777216 < size 67056 1171888537600
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4083 57228.
MP_Allreduce 11247 947.
MP_Sync 168
MP_Alltoall 1955 5526701.
MP_SendRecv 11938 47072.
MP_ISendRecv 11938 47072.
MP_Wait 25718
MP_comm_split 82
MP_ISend 11660 212488.
MP_IRecv 11660 212488.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.013 0.028 191.804 191.806
qs_mol_dyn_low 1 2.0 0.003 0.003 191.482 191.496
qs_forces 11 3.9 0.003 0.003 191.399 191.400
qs_energies 11 4.9 0.002 0.002 184.612 184.623
scf_env_do_scf 11 5.9 0.001 0.001 168.165 168.175
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 133.959 133.962
velocity_verlet 10 3.0 0.001 0.002 120.441 120.442
dbcsr_multiply_generic 2485 12.5 0.185 0.193 97.777 99.129
qs_scf_new_mos 116 7.6 0.001 0.001 94.967 95.498
qs_scf_loop_do_ot 116 8.6 0.001 0.001 94.966 95.497
ot_scf_mini 116 9.6 0.004 0.004 90.176 90.756
multiply_cannon 2485 13.5 0.476 0.531 77.183 81.804
multiply_cannon_loop 2485 14.5 1.242 1.287 73.868 76.583
ot_mini 116 10.6 0.001 0.001 50.178 50.708
mp_waitall_1 212858 16.6 25.442 38.570 25.442 38.570
multiply_cannon_multrec 29820 15.5 21.702 26.928 31.325 36.776
init_scf_loop 11 6.9 0.000 0.000 34.111 34.113
rebuild_ks_matrix 127 8.3 0.001 0.001 32.919 33.448
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.019 32.918 33.447
qs_ks_update_qs_env 127 7.6 0.001 0.001 29.633 30.112
prepare_preconditioner 11 7.9 0.000 0.000 29.693 29.757
make_preconditioner 11 8.9 0.000 0.000 29.693 29.757
multiply_cannon_metrocomm3 29820 15.5 0.093 0.099 16.089 29.248
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.330 28.865
qs_ot_get_derivative 116 11.6 0.001 0.002 28.151 28.714
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 22.091 23.279
apply_single 127 13.6 0.001 0.001 22.090 23.279
qs_ot_get_p 127 10.4 0.001 0.001 21.790 22.466
ot_diis_step 116 11.6 0.014 0.015 21.861 21.863
multiply_cannon_sync_h2d 29820 15.5 19.046 21.181 19.046 21.181
cp_fm_cholesky_invert 11 10.9 17.467 17.482 17.467 17.482
qs_ot_p2m_diag 82 11.4 0.185 0.213 16.912 16.959
make_m2s 4970 13.5 0.091 0.095 14.725 16.155
make_images 4970 14.5 1.162 1.345 14.514 15.938
cp_dbcsr_syevd 82 12.4 0.005 0.005 15.703 15.704
sum_up_and_integrate 127 10.3 0.116 0.130 14.542 14.571
integrate_v_rspace 127 11.3 0.003 0.003 14.426 14.461
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.259 13.301
calculate_rho_elec 127 8.7 0.087 0.104 13.258 13.300
cp_fm_diag_elpa 82 13.4 0.000 0.001 12.496 12.511
cp_fm_redistribute_end 82 14.4 7.296 12.428 7.306 12.429
cp_fm_diag_elpa_base 82 14.4 4.880 11.910 5.095 12.257
init_scf_run 11 5.9 0.000 0.001 11.661 11.663
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.661 11.663
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 11.073 11.513
make_images_data 4970 15.5 0.064 0.071 9.024 10.828
multiply_cannon_metrocomm4 27335 15.5 0.093 0.107 3.714 10.749
mp_irecv_dv 68888 16.3 3.525 10.364 3.525 10.364
hybrid_alltoall_any 5155 16.4 0.338 1.479 7.765 10.029
dbcsr_mm_accdrv_process 61726 16.2 4.533 5.275 9.092 9.715
wfi_extrapolate 11 7.9 0.001 0.001 8.486 8.486
pw_transfer 1535 11.6 0.084 0.102 7.960 8.015
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 7.738 7.790
grid_integrate_task_list 127 12.3 7.123 7.553 7.123 7.553
density_rs2pw 127 9.7 0.006 0.006 7.043 7.513
cp_fm_cholesky_decompose 22 10.9 7.282 7.375 7.282 7.375
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.288 7.031
fft_wrap_pw1pw2_140 519 13.2 0.465 0.513 6.691 6.746
calculate_dm_sparse 127 9.5 0.001 0.001 6.542 6.680
rs_pw_transfer 1038 11.9 0.014 0.016 5.971 6.469
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.211 6.222
fft3d_ps 1281 14.7 2.753 2.905 6.121 6.147
mp_sum_l 7804 13.0 4.100 6.083 4.100 6.083
mp_alltoall_d11v 2401 14.1 4.345 5.468 4.345 5.468
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.313 5.459
qs_ot_get_orbitals 116 10.6 0.001 0.001 5.374 5.445
mp_allgather_i34 2485 14.5 1.893 5.308 1.893 5.308
grid_collocate_task_list 127 9.7 4.871 5.159 4.871 5.159
potential_pw2rs 127 12.3 0.014 0.016 5.046 5.070
dbcsr_complete_redistribute 393 12.7 0.768 0.844 3.297 4.158
mp_sum_d 4441 12.1 2.641 3.976 2.641 3.976
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=191.806000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=791.818182, yerr=4.628335
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430458527744 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613083000832 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228655E+12 0.0% 0.0% 100.0%
flops max/rank 2.950886E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806314816 0.0% 0.0% 100.0%
number of processed stacks 4015680 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1694.9
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 949.833728E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1051232
MPI messages size (bytes):
total size 2.737039E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.603648E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 284160 37245419520
131072 < size <= 4194304 665408 1004401590272
4194304 < size <= 16777216 66080 937889582784
16777216 < size 28896 757491302400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4113 56816.
MP_Allreduce 11321 983.
MP_Sync 170
MP_Alltoall 1724 9394160.
MP_SendRecv 7998 75008.
MP_ISendRecv 7998 75008.
MP_Wait 21986
MP_comm_split 83
MP_ISend 11836 275177.
MP_IRecv 11836 275177.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.035 179.621 179.622
qs_mol_dyn_low 1 2.0 0.003 0.004 179.174 179.186
qs_forces 11 3.9 0.003 0.003 179.063 179.067
qs_energies 11 4.9 0.014 0.034 172.543 172.551
scf_env_do_scf 11 5.9 0.001 0.012 156.925 156.926
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 120.623 120.625
velocity_verlet 10 3.0 0.019 0.020 115.497 115.499
dbcsr_multiply_generic 2527 12.6 0.182 0.187 83.289 84.469
qs_scf_new_mos 118 7.6 0.001 0.001 83.219 83.602
qs_scf_loop_do_ot 118 8.6 0.001 0.002 83.218 83.601
ot_scf_mini 118 9.6 0.004 0.004 78.981 79.446
multiply_cannon 2527 13.6 0.505 0.535 62.626 66.961
multiply_cannon_loop 2527 14.6 0.862 0.899 59.262 61.835
ot_mini 118 10.6 0.001 0.001 43.755 44.226
mp_waitall_1 171878 16.6 26.181 36.265 26.181 36.265
init_scf_loop 11 6.9 0.001 0.003 36.186 36.188
prepare_preconditioner 11 7.9 0.000 0.000 32.155 32.203
make_preconditioner 11 8.9 0.000 0.001 32.155 32.203
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.772 31.213
rebuild_ks_matrix 129 8.3 0.001 0.001 30.608 31.130
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.018 30.608 31.129
qs_ks_update_qs_env 129 7.6 0.001 0.001 27.665 28.140
multiply_cannon_multrec 20216 15.6 13.454 16.930 22.202 25.612
multiply_cannon_metrocomm3 20216 15.6 0.061 0.066 15.683 25.342
qs_ot_get_derivative 118 11.6 0.001 0.002 23.677 24.143
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 20.180 21.214
apply_single 129 13.6 0.001 0.001 20.179 21.214
qs_ot_get_p 129 10.4 0.001 0.001 20.365 20.894
ot_diis_step 118 11.6 0.018 0.019 19.971 19.971
make_m2s 5054 13.6 0.083 0.087 15.307 16.451
multiply_cannon_sync_h2d 20216 15.6 14.271 16.236 14.271 16.236
make_images 5054 14.6 1.208 1.304 15.071 16.213
qs_ot_p2m_diag 83 11.4 0.265 0.272 15.872 15.898
cp_fm_cholesky_invert 11 10.9 15.259 15.271 15.259 15.271
cp_dbcsr_syevd 83 12.4 0.005 0.005 14.789 14.791
sum_up_and_integrate 129 10.3 0.135 0.147 14.493 14.513
integrate_v_rspace 129 11.3 0.003 0.004 14.358 14.382
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.428 13.483
calculate_rho_elec 129 8.7 0.132 0.147 13.427 13.482
cp_fm_diag_elpa 83 13.4 0.000 0.001 11.586 11.590
cp_fm_redistribute_end 83 14.4 4.390 11.518 4.414 11.521
cp_fm_diag_elpa_base 83 14.4 6.672 10.889 7.083 11.354
make_images_data 5054 15.6 0.062 0.070 9.524 11.076
init_scf_run 11 5.9 0.000 0.001 10.538 10.540
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.538 10.540
hybrid_alltoall_any 5240 16.5 0.434 1.978 8.210 9.930
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.345 9.684
multiply_cannon_metrocomm4 17689 15.6 0.062 0.073 3.467 9.386
mp_irecv_dv 50625 16.2 3.345 9.141 3.345 9.141
dbcsr_mm_accdrv_process 41830 16.2 4.472 5.312 8.208 8.393
pw_transfer 1559 11.6 0.086 0.107 7.979 8.093
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 7.754 7.871
grid_integrate_task_list 129 12.3 7.381 7.856 7.381 7.856
cp_fm_cholesky_decompose 22 10.9 7.813 7.843 7.813 7.843
wfi_extrapolate 11 7.9 0.001 0.001 7.479 7.479
density_rs2pw 129 9.7 0.006 0.006 6.772 7.387
cp_fm_upper_to_full 105 14.5 5.658 7.368 5.658 7.368
fft_wrap_pw1pw2_140 527 13.2 0.480 0.528 6.699 6.818
dbcsr_complete_redistribute 395 12.7 1.172 1.197 4.573 6.319
fft3d_ps 1301 14.7 2.712 2.923 6.012 6.080
rs_pw_transfer 1054 12.0 0.014 0.014 5.431 5.988
calculate_dm_sparse 129 9.5 0.001 0.001 5.855 5.973
grid_collocate_task_list 129 9.7 5.116 5.705 5.116 5.705
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.449 5.457
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 4.763 5.435
mp_sum_l 7930 13.1 3.527 5.228 3.527 5.228
mp_allgather_i34 2527 14.6 1.864 5.105 1.864 5.105
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.348 5.095
mp_alltoall_d11v 2423 14.1 4.315 5.076 4.315 5.076
potential_pw2rs 129 12.3 0.020 0.023 4.753 4.772
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.536 4.663
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.149 4.187
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.363 4.073
mp_alltoall_i22 718 14.1 2.014 3.866 2.014 3.866
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 3.771 3.772
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=179.622000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=890.363636, yerr=14.561922
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 4.387242E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 6026880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1129.3
marketing flops 145.651870E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 1.136402E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1153224
MPI messages size (bytes):
total size 2.039489E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.768511E+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 322096 36390305792
131072 < size <= 4194304 721976 792118951936
4194304 < size <= 16777216 70800 669922227920
16777216 < size 30960 541065216000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4123 56680.
MP_Allreduce 11346 1064.
MP_Sync 172
MP_Alltoall 1724 12509605.
MP_SendRecv 5934 75008.
MP_ISendRecv 5934 75008.
MP_Wait 22612
MP_comm_split 84
MP_ISend 15064 244788.
MP_IRecv 15064 244788.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.057 0.163 193.098 193.099
qs_mol_dyn_low 1 2.0 0.003 0.004 192.558 192.571
qs_forces 11 3.9 0.003 0.003 192.437 192.444
qs_energies 11 4.9 0.002 0.005 185.381 185.394
scf_env_do_scf 11 5.9 0.001 0.002 168.450 168.465
velocity_verlet 10 3.0 0.002 0.002 128.106 128.108
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 120.111 120.112
qs_scf_new_mos 118 7.6 0.001 0.001 83.463 83.777
qs_scf_loop_do_ot 118 8.6 0.001 0.001 83.462 83.777
dbcsr_multiply_generic 2529 12.6 0.191 0.195 80.926 81.660
ot_scf_mini 118 9.6 0.003 0.005 78.902 79.245
multiply_cannon 2529 13.6 0.559 0.594 55.255 58.405
multiply_cannon_loop 2529 14.6 1.193 1.226 51.456 53.348
init_scf_loop 11 6.9 0.000 0.002 48.198 48.199
prepare_preconditioner 11 7.9 0.000 0.000 44.048 44.076
make_preconditioner 11 8.9 0.000 0.001 44.048 44.076
ot_mini 118 10.6 0.001 0.001 43.344 43.677
make_full_inverse_cholesky 11 9.9 0.000 0.000 37.689 42.638
multiply_cannon_multrec 30348 15.6 14.028 19.793 26.059 31.262
rebuild_ks_matrix 129 8.3 0.001 0.001 29.809 30.128
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.018 29.808 30.127
mp_waitall_1 149172 16.7 18.414 28.461 18.414 28.461
qs_ks_update_qs_env 129 7.6 0.001 0.001 26.990 27.278
qs_ot_get_derivative 118 11.6 0.001 0.002 23.354 23.698
make_m2s 5058 13.6 0.098 0.104 21.187 22.253
make_images 5058 14.6 1.993 2.303 20.877 21.942
qs_ot_get_p 129 10.4 0.001 0.001 21.420 21.785
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.413 19.982
apply_single 129 13.6 0.001 0.001 19.412 19.982
ot_diis_step 118 11.6 0.018 0.019 19.873 19.876
cp_fm_cholesky_invert 11 10.9 17.403 17.413 17.403 17.413
qs_ot_p2m_diag 84 11.4 0.347 0.394 17.180 17.236
cp_fm_upper_to_full 106 14.7 11.035 16.334 11.035 16.334
cp_dbcsr_syevd 84 12.4 0.005 0.006 15.844 15.845
multiply_cannon_metrocomm3 30348 15.6 0.047 0.050 6.809 15.649
sum_up_and_integrate 129 10.3 0.142 0.153 14.561 14.587
integrate_v_rspace 129 11.3 0.003 0.004 14.419 14.451
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.583 13.619
calculate_rho_elec 129 8.7 0.176 0.192 13.582 13.619
make_images_data 5058 15.6 0.065 0.069 11.526 13.507
dbcsr_complete_redistribute 397 12.7 1.513 1.627 9.131 12.881
hybrid_alltoall_any 5245 16.5 0.529 2.237 10.269 12.595
cp_fm_diag_elpa 84 13.4 0.000 0.001 12.514 12.518
multiply_cannon_sync_h2d 30348 15.6 11.755 12.465 11.755 12.465
cp_fm_redistribute_end 84 14.4 2.169 12.417 2.189 12.423
cp_fm_diag_elpa_base 84 14.4 9.579 11.738 10.188 12.244
dbcsr_mm_accdrv_process 62780 16.2 7.382 8.311 11.597 12.143
copy_fm_to_dbcsr 210 11.7 0.002 0.003 7.662 11.400
init_scf_run 11 5.9 0.000 0.001 10.963 10.964
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.963 10.964
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.337 9.994
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 9.720 9.968
mp_alltoall_i22 720 14.1 5.595 9.355 5.595 9.355
pw_transfer 1559 11.6 0.086 0.100 8.245 8.366
cp_fm_cholesky_decompose 22 10.9 8.147 8.283 8.147 8.283
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 8.018 8.142
grid_integrate_task_list 129 12.3 7.606 8.068 7.606 8.068
wfi_extrapolate 11 7.9 0.001 0.001 7.796 7.796
multiply_cannon_metrocomm4 25290 15.6 0.076 0.087 2.862 7.275
fft_wrap_pw1pw2_140 527 13.2 0.485 0.497 7.023 7.179
mp_irecv_dv 76751 16.2 2.718 7.009 2.718 7.009
density_rs2pw 129 9.7 0.006 0.006 6.566 6.873
calculate_dm_sparse 129 9.5 0.001 0.001 6.250 6.321
fft3d_ps 1301 14.7 2.825 2.928 6.237 6.315
grid_collocate_task_list 129 9.7 5.253 5.718 5.253 5.718
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.535 5.588
mp_alltoall_d11v 2429 14.1 5.129 5.432 5.129 5.432
rs_pw_transfer 1054 12.0 0.013 0.015 4.977 5.300
potential_pw2rs 129 12.3 0.023 0.023 4.621 4.646
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.471 4.549
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.390 4.486
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.450 4.451
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.340 4.396
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=193.099000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1074.636364, yerr=17.884847
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 5.820059E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 1944496 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3448.5
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.515262E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 238560
MPI messages size (bytes):
total size 1.321104E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.537828E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 112800 59139686400
4194304 < size <= 16777216 104112 545846722560
16777216 < size 20064 716108638608
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8852 52.
MP_Alltoall 9584 804353.
MP_ISend 39716 2104723.
MP_IRecv 39716 2103824.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3997 58279.
MP_Allreduce 10990 1175.
MP_Sync 86
MP_Alltoall 1700 18828148.
MP_SendRecv 3810 122880.
MP_ISendRecv 3810 122880.
MP_Wait 16000
MP_ISend 10600 423612.
MP_IRecv 10600 423612.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.047 0.079 174.751 174.751
qs_mol_dyn_low 1 2.0 0.003 0.004 174.243 174.256
qs_forces 11 3.9 0.003 0.003 174.145 174.152
qs_energies 11 4.9 0.001 0.002 166.777 166.785
scf_env_do_scf 11 5.9 0.001 0.008 149.484 149.496
velocity_verlet 10 3.0 0.013 0.014 113.586 113.590
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 112.700 112.702
qs_scf_new_mos 116 7.6 0.001 0.001 76.794 76.934
qs_scf_loop_do_ot 116 8.6 0.001 0.001 76.793 76.934
dbcsr_multiply_generic 2485 12.5 0.180 0.186 75.077 75.610
ot_scf_mini 116 9.6 0.003 0.004 72.383 72.552
multiply_cannon 2485 13.5 0.580 0.607 54.798 58.996
multiply_cannon_loop 2485 14.5 0.442 0.458 50.142 51.184
ot_mini 116 10.6 0.001 0.001 40.150 40.329
init_scf_loop 11 6.9 0.000 0.000 36.620 36.623
mp_waitall_1 124680 16.7 26.993 35.225 26.993 35.225
prepare_preconditioner 11 7.9 0.000 0.000 32.692 32.717
make_preconditioner 11 8.9 0.000 0.000 32.692 32.717
make_full_inverse_cholesky 11 9.9 0.000 0.000 30.500 30.768
rebuild_ks_matrix 127 8.3 0.001 0.001 28.804 28.936
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.017 28.803 28.935
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.286 26.409
multiply_cannon_multrec 9940 15.5 10.344 14.422 17.670 22.140
multiply_cannon_metrocomm3 9940 15.5 0.023 0.024 12.903 20.459
qs_ot_get_derivative 116 11.6 0.001 0.002 20.235 20.404
ot_diis_step 116 11.6 0.019 0.020 19.838 19.838
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.403 19.731
apply_single 127 13.6 0.001 0.001 19.403 19.731
qs_ot_get_p 127 10.4 0.001 0.001 19.152 19.345
cp_fm_cholesky_invert 11 10.9 18.857 18.865 18.857 18.865
make_m2s 4970 13.5 0.066 0.071 16.509 18.577
make_images 4970 14.5 2.284 2.650 16.206 18.273
qs_ot_p2m_diag 82 11.4 0.489 0.497 15.272 15.294
sum_up_and_integrate 127 10.3 0.179 0.189 14.201 14.245
cp_dbcsr_syevd 82 12.4 0.005 0.005 14.108 14.110
integrate_v_rspace 127 11.3 0.004 0.004 14.022 14.076
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.301 13.339
calculate_rho_elec 127 8.7 0.256 0.267 13.301 13.339
make_images_data 4970 15.5 0.052 0.060 10.152 12.436
hybrid_alltoall_any 5155 16.4 0.839 3.799 9.872 12.155
multiply_cannon_sync_h2d 9940 15.5 11.404 11.979 11.404 11.979
cp_fm_diag_elpa 82 13.4 0.000 0.000 10.985 10.998
cp_fm_diag_elpa_base 82 14.4 10.736 10.824 10.977 10.988
init_scf_run 11 5.9 0.000 0.001 10.435 10.435
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.435 10.435
qs_ot_get_derivative_diag 76 12.4 0.002 0.003 8.177 8.290
cp_fm_cholesky_decompose 22 10.9 8.117 8.273 8.117 8.273
grid_integrate_task_list 127 12.3 7.702 8.064 7.702 8.064
multiply_cannon_metrocomm1 9940 15.5 0.028 0.029 4.577 8.012
pw_transfer 1535 11.6 0.083 0.089 7.961 7.978
fft_wrap_pw1pw2 1281 12.7 0.010 0.010 7.741 7.762
dbcsr_mm_accdrv_process 20590 16.1 2.620 3.418 6.963 7.579
mp_allgather_i34 2485 14.5 2.870 7.487 2.870 7.487
wfi_extrapolate 11 7.9 0.001 0.001 7.443 7.443
fft_wrap_pw1pw2_140 519 13.2 0.500 0.529 6.754 6.783
density_rs2pw 127 9.7 0.005 0.006 6.108 6.324
calculate_dm_sparse 127 9.5 0.001 0.001 6.143 6.224
fft3d_ps 1281 14.7 2.711 2.802 5.910 5.936
dbcsr_complete_redistribute 393 12.7 2.124 2.213 5.441 5.847
grid_collocate_task_list 127 9.7 5.442 5.798 5.442 5.798
mp_alltoall_d11v 2401 14.1 5.079 5.634 5.079 5.634
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.298 5.299
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.250 5.258
multiply_cannon_metrocomm4 7455 15.5 0.024 0.027 1.846 5.254
mp_irecv_dv 28618 15.9 1.812 5.172 1.812 5.172
rs_pw_transfer 1038 11.9 0.012 0.013 4.304 4.524
potential_pw2rs 127 12.3 0.026 0.027 4.241 4.259
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.102 4.158
copy_fm_to_dbcsr 208 11.6 0.002 0.002 3.555 3.899
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.561 3.876
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.785 3.826
copy_dbcsr_to_fm 185 11.7 0.004 0.004 3.641 3.736
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.591 3.640
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=174.751000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1400.181818, yerr=48.262197
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430451062784 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986258751488 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1990717341696 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1991997444096 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613050155008 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239170400256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239170400256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.260896E+12 0.0% 0.0% 100.0%
flops max/rank 11.789544E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806803520 0.0% 0.0% 100.0%
number of processed stacks 1985072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3429.0
marketing flops 145.678848E+12
-------------------------------------------------------------------------------
# multiplications 2542
max memory usage/rank 3.111645E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101680
MPI messages size (bytes):
total size 1.145405E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.264802E+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 46168 35869687808
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592076688
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 58888.
MP_Allreduce 11047 1512.
MP_Sync 87
MP_Alltoall 1724 36993678.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.059 294.217 294.217
qs_mol_dyn_low 1 2.0 0.003 0.004 293.585 293.597
qs_forces 11 3.9 0.003 0.003 293.492 293.496
qs_energies 11 4.9 0.001 0.001 284.902 284.918
scf_env_do_scf 11 5.9 0.001 0.002 262.465 262.476
velocity_verlet 10 3.0 0.002 0.002 212.517 212.524
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 136.671 136.673
init_scf_loop 11 6.9 0.000 0.000 125.530 125.532
prepare_preconditioner 11 7.9 0.000 0.000 120.763 120.786
make_preconditioner 11 8.9 0.000 0.000 120.763 120.786
make_full_inverse_cholesky 11 9.9 0.000 0.000 96.938 117.939
qs_scf_new_mos 118 7.6 0.001 0.001 93.704 93.824
qs_scf_loop_do_ot 118 8.6 0.001 0.001 93.703 93.823
ot_scf_mini 118 9.6 0.004 0.004 88.810 88.857
dbcsr_multiply_generic 2542 12.6 0.214 0.222 85.714 86.347
cp_fm_upper_to_full 105 14.8 52.831 75.682 52.831 75.682
multiply_cannon 2542 13.6 0.716 0.762 60.598 61.097
multiply_cannon_loop 2542 14.6 0.477 0.488 56.832 58.307
ot_mini 118 10.6 0.001 0.001 45.642 45.690
dbcsr_complete_redistribute 395 12.7 4.015 4.057 29.595 42.344
copy_fm_to_dbcsr 209 11.7 0.002 0.002 26.155 38.879
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 23.778 36.325
cp_fm_cholesky_invert 11 10.9 34.529 34.536 34.529 34.536
mp_alltoall_i22 718 14.1 21.587 34.443 21.587 34.443
rebuild_ks_matrix 129 8.3 0.001 0.001 34.214 34.265
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 34.213 34.264
mp_waitall_1 105100 16.8 29.445 32.999 29.445 32.999
qs_ks_update_qs_env 129 7.6 0.001 0.001 31.957 32.006
qs_ot_get_p 129 10.4 0.001 0.001 27.257 27.306
qs_ot_get_derivative 118 11.6 0.002 0.002 25.033 25.078
qs_ot_p2m_diag 83 11.4 0.879 0.883 23.038 23.069
cp_dbcsr_syevd 83 12.4 0.006 0.006 21.274 21.275
make_m2s 5084 13.6 0.078 0.082 20.293 21.263
multiply_cannon_metrocomm3 10168 15.6 0.024 0.025 19.847 21.199
make_images 5084 14.6 3.830 3.955 19.810 20.784
ot_diis_step 118 11.6 0.022 0.023 20.571 20.572
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.757 20.002
apply_single 129 13.6 0.001 0.001 19.756 20.002
multiply_cannon_multrec 10168 15.6 10.636 12.402 18.506 18.592
cp_fm_diag_elpa 83 13.4 0.000 0.000 17.969 17.970
cp_fm_diag_elpa_base 83 14.4 13.561 15.204 17.964 17.964
sum_up_and_integrate 129 10.3 0.325 0.328 16.044 16.136
integrate_v_rspace 129 11.3 0.004 0.004 15.719 15.810
multiply_cannon_sync_h2d 10168 15.6 15.787 15.796 15.787 15.796
qs_rho_update_rho_low 129 7.7 0.001 0.001 15.352 15.374
calculate_rho_elec 129 8.7 0.487 0.488 15.351 15.374
make_images_data 5084 15.6 0.062 0.066 10.815 12.561
hybrid_alltoall_any 5270 16.5 1.312 3.065 10.839 12.493
init_scf_run 11 5.9 0.000 0.001 12.443 12.444
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.443 12.444
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.726 9.759
dbcsr_mm_accdrv_process 20986 16.1 3.885 5.671 7.629 9.512
cp_fm_cholesky_decompose 22 10.9 9.368 9.446 9.368 9.446
wfi_extrapolate 11 7.9 0.001 0.001 8.827 8.827
grid_integrate_task_list 129 12.3 8.600 8.772 8.600 8.772
pw_transfer 1559 11.6 0.091 0.092 8.678 8.685
fft_wrap_pw1pw2 1301 12.7 0.011 0.011 8.443 8.450
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 7.974 7.976
fft_wrap_pw1pw2_140 527 13.2 0.543 0.545 7.454 7.467
mp_alltoall_d11v 2423 14.1 7.226 7.384 7.226 7.384
calculate_dm_sparse 129 9.5 0.001 0.001 7.074 7.174
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.596 6.655
fft3d_ps 1301 14.7 2.779 2.787 6.471 6.480
grid_collocate_task_list 129 9.7 6.385 6.418 6.385 6.418
density_rs2pw 129 9.7 0.005 0.005 6.255 6.285
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.077 6.125
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=294.217000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2744.000000, yerr=194.491879
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/21/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 2.766000E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 419739 0.0% 0.0% 100.0%
average stack size 0.0 0.0 22952.9
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 1.261564E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255669.
MP_Allreduce 3059 6274.
MP_Sync 4
MP_Alltoall 54 44080663.
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.128 85.116 85.117
qs_energies 1 2.0 0.000 0.001 84.558 84.566
ls_scf 1 3.0 0.000 0.000 83.651 83.659
dbcsr_multiply_generic 111 6.7 0.015 0.016 72.690 72.863
multiply_cannon 111 7.7 0.017 0.021 55.887 57.223
multiply_cannon_loop 111 8.7 0.210 0.223 52.419 53.829
ls_scf_main 1 4.0 0.000 0.000 52.201 52.201
density_matrix_trs4 2 5.0 0.002 0.003 46.732 46.823
ls_scf_init_scf 1 4.0 0.000 0.000 28.387 28.388
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.215 27.263
mp_waitall_1 11031 10.9 22.546 25.322 22.546 25.322
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.110 25.135
multiply_cannon_multrec 2664 9.7 8.134 8.892 15.544 17.352
multiply_cannon_sync_h2d 2664 9.7 13.478 15.396 13.478 15.396
make_m2s 222 7.7 0.008 0.011 13.071 13.675
make_images 222 8.7 0.099 0.109 13.049 13.653
multiply_cannon_metrocomm1 2664 9.7 0.009 0.010 9.695 12.621
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.524 9.541
make_images_data 222 9.7 0.004 0.005 7.645 8.200
dbcsr_mm_accdrv_process 4760 10.4 0.514 0.648 7.028 8.042
hybrid_alltoall_any 227 10.6 0.214 1.833 6.514 7.689
dbcsr_mm_accdrv_process_sort 4760 11.4 6.312 7.218 6.312 7.218
calculate_norms 4752 9.8 5.498 6.080 5.498 6.080
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.068 5.157
mp_sum_l 807 5.4 3.262 4.763 3.262 4.763
multiply_cannon_metrocomm4 2442 9.7 0.012 0.016 2.080 4.206
mp_irecv_dv 6231 10.9 2.062 4.180 2.062 4.180
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.428 3.769
make_images_sizes 222 9.7 0.000 0.000 0.726 3.347
mp_alltoall_i44 222 10.7 0.726 3.346 0.726 3.346
arnoldi_extremal 4 6.8 0.000 0.000 3.236 3.276
arnoldi_normal_ev 4 7.8 0.001 0.004 3.236 3.276
build_subspace 16 8.4 0.009 0.012 3.129 3.132
ls_scf_post 1 4.0 0.000 0.000 3.063 3.071
ls_scf_store_result 1 5.0 0.000 0.000 2.874 2.916
dbcsr_special_finalize 555 9.7 0.005 0.006 2.346 2.745
dbcsr_merge_single_wm 555 10.7 0.462 0.631 2.339 2.737
make_images_pack 222 9.7 2.206 2.626 2.208 2.628
dbcsr_matrix_vector_mult 304 9.0 0.003 0.010 2.328 2.570
dbcsr_sort_data 658 11.4 2.135 2.465 2.135 2.465
dbcsr_matrix_vector_mult_local 304 10.0 2.066 2.460 2.068 2.461
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.200 2.270
buffer_matrices_ensure_size 222 8.7 1.751 2.099 1.751 2.099
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.779 1.780
rebuild_ks_matrix 3 7.3 0.000 0.000 1.769 1.771
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.769 1.771
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=85.117000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1141.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/22/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 5.588524E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 368848 0.0% 0.0% 100.0%
average stack size 0.0 0.0 26119.8
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 2.098532E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 10339.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.060 0.100 91.924 91.926
qs_energies 1 2.0 0.001 0.008 91.252 91.306
ls_scf 1 3.0 0.001 0.005 89.924 89.974
dbcsr_multiply_generic 111 6.7 0.015 0.016 75.763 76.183
multiply_cannon 111 7.7 0.029 0.039 53.358 57.297
ls_scf_main 1 4.0 0.000 0.003 55.438 55.443
multiply_cannon_loop 111 8.7 0.116 0.122 50.055 53.584
density_matrix_trs4 2 5.0 0.002 0.004 49.790 49.987
ls_scf_init_scf 1 4.0 0.001 0.002 30.919 30.927
mp_waitall_1 9105 10.9 21.391 30.475 21.391 30.475
ls_scf_init_matrix_S 1 5.0 0.000 0.002 29.626 29.763
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 27.209 27.225
multiply_cannon_multrec 1332 9.7 13.127 16.695 22.458 27.140
multiply_cannon_metrocomm3 1332 9.7 0.006 0.008 11.731 20.694
make_m2s 222 7.7 0.006 0.007 15.682 16.442
make_images 222 8.7 1.573 1.937 15.652 16.413
dbcsr_mm_accdrv_process 4041 10.4 0.272 0.453 8.928 10.539
dbcsr_mm_accdrv_process_sort 4041 11.4 8.520 10.086 8.520 10.086
make_images_data 222 9.7 0.004 0.004 9.054 10.037
hybrid_alltoall_any 227 10.6 0.521 2.445 8.432 9.608
mp_sum_l 807 5.4 5.428 8.544 5.428 8.544
multiply_cannon_metrocomm4 1221 9.7 0.006 0.008 3.229 7.534
mp_irecv_dv 3311 11.0 3.209 7.480 3.209 7.480
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.226 6.901
calculate_norms 2376 9.8 6.003 6.695 6.003 6.695
multiply_cannon_sync_h2d 1332 9.7 4.827 6.168 4.827 6.168
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.021 5.220
arnoldi_extremal 4 6.8 0.000 0.002 4.796 4.822
arnoldi_normal_ev 4 7.8 0.001 0.008 4.795 4.822
build_subspace 16 8.4 0.014 0.021 4.522 4.529
ls_scf_post 1 4.0 0.002 0.016 3.566 3.615
dbcsr_matrix_vector_mult 304 9.0 0.005 0.017 3.197 3.421
ls_scf_store_result 1 5.0 0.000 0.000 3.243 3.365
dbcsr_matrix_vector_mult_local 304 10.0 2.747 3.232 2.749 3.234
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.527 2.612
mp_allgather_i34 111 8.7 0.996 2.531 0.996 2.531
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 1.231 2.505
make_images_pack 222 9.7 2.031 2.444 2.033 2.446
dbcsr_sort_data 436 11.2 1.850 2.061 1.850 2.061
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.838 1.848
dbcsr_data_new 4174 10.1 1.609 1.842 1.609 1.842
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=91.926000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1685.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/23/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 8.404608E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 353133 0.0% 0.0% 100.0%
average stack size 0.0 0.0 27282.1
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 2.698703E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265470.
MP_Allreduce 3058 11181.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.031 0.052 95.207 95.208
qs_energies 1 2.0 0.000 0.001 94.641 94.643
ls_scf 1 3.0 0.000 0.002 93.246 93.249
dbcsr_multiply_generic 111 6.7 0.015 0.016 77.938 78.302
ls_scf_main 1 4.0 0.000 0.002 58.331 58.347
multiply_cannon 111 7.7 0.042 0.117 53.515 57.427
multiply_cannon_loop 111 8.7 0.100 0.106 49.923 54.180
density_matrix_trs4 2 5.0 0.002 0.003 52.409 52.575
mp_waitall_1 7281 11.0 24.521 34.611 24.521 34.611
ls_scf_init_scf 1 4.0 0.000 0.001 31.282 31.285
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.127 30.226
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 27.364 27.377
multiply_cannon_multrec 888 9.7 12.572 14.985 21.279 24.141
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.560 23.446
make_m2s 222 7.7 0.006 0.007 17.292 18.544
make_images 222 8.7 1.974 2.292 17.254 18.502
hybrid_alltoall_any 227 10.6 0.620 2.869 9.458 11.005
make_images_data 222 9.7 0.003 0.004 9.897 10.917
mp_sum_l 807 5.4 5.573 9.724 5.573 9.724
dbcsr_mm_accdrv_process 3754 10.4 0.259 0.723 8.236 9.394
dbcsr_mm_accdrv_process_sort 3754 11.4 7.782 8.979 7.782 8.979
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.266 7.787
multiply_cannon_sync_h2d 888 9.7 6.020 7.398 6.020 7.398
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 3.683 7.064
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.445 6.988
mp_irecv_dv 2335 11.1 2.430 6.942 2.430 6.942
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.320 5.480
arnoldi_extremal 4 6.8 0.000 0.000 5.218 5.235
arnoldi_normal_ev 4 7.8 0.001 0.005 5.218 5.235
build_subspace 16 8.4 0.014 0.020 4.912 4.918
calculate_norms 1584 9.8 4.315 4.756 4.315 4.756
mp_allgather_i34 111 8.7 1.424 3.974 1.424 3.974
dbcsr_matrix_vector_mult 304 9.0 0.005 0.017 3.501 3.825
ls_scf_post 1 4.0 0.000 0.001 3.632 3.636
dbcsr_matrix_vector_mult_local 304 10.0 3.033 3.610 3.035 3.612
ls_scf_store_result 1 5.0 0.000 0.000 3.349 3.462
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.750 2.827
make_images_sizes 222 9.7 0.000 0.000 1.099 2.751
mp_alltoall_i44 222 10.7 1.099 2.751 1.099 2.751
dbcsr_sort_data 325 11.1 1.855 2.137 1.855 2.137
make_images_pack 222 9.7 1.817 2.125 1.820 2.128
dbcsr_data_release 9322 10.9 1.310 2.005 1.310 2.005
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=95.208000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2262.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/24/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 10.747127E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 369794 0.0% 0.0% 100.0%
average stack size 0.0 0.0 26053.0
marketing flops 1.742116E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 3.360100E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 13371.
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 0.059 0.097 96.460 96.462
qs_energies 1 2.0 0.000 0.001 95.730 95.742
ls_scf 1 3.0 0.001 0.003 94.060 94.073
dbcsr_multiply_generic 111 6.7 0.020 0.031 77.728 78.019
ls_scf_main 1 4.0 0.001 0.003 58.190 58.192
multiply_cannon 111 7.7 0.059 0.139 51.245 56.186
density_matrix_trs4 2 5.0 0.003 0.007 52.177 52.314
multiply_cannon_loop 111 8.7 0.114 0.126 46.108 48.988
ls_scf_init_scf 1 4.0 0.001 0.002 32.630 32.631
ls_scf_init_matrix_S 1 5.0 0.001 0.003 31.454 31.525
mp_waitall_1 6369 11.0 21.898 29.079 21.898 29.079
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.970 28.986
multiply_cannon_multrec 1332 9.7 14.310 17.387 22.273 25.429
make_m2s 222 7.7 0.007 0.008 20.871 22.323
make_images 222 8.7 3.148 3.607 20.820 22.275
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 8.633 16.739
make_images_data 222 9.7 0.004 0.005 11.568 13.206
hybrid_alltoall_any 227 10.6 0.799 3.812 10.885 12.695
dbcsr_mm_accdrv_process 3641 10.4 0.213 0.405 7.591 9.117
dbcsr_mm_accdrv_process_sort 3641 11.4 7.216 8.702 7.216 8.702
mp_sum_l 807 5.4 4.049 7.285 4.049 7.285
multiply_cannon_sync_h2d 1332 9.7 5.468 6.210 5.468 6.210
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.070 6.138
mp_irecv_dv 3229 10.9 2.047 6.057 2.047 6.057
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.064 5.596
arnoldi_extremal 4 6.8 0.000 0.000 5.427 5.442
arnoldi_normal_ev 4 7.8 0.001 0.004 5.427 5.442
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.528 5.207
build_subspace 16 8.4 0.015 0.021 5.077 5.087
mp_allgather_i34 111 8.7 2.264 4.757 2.264 4.757
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.537 4.737
calculate_norms 2376 9.8 4.196 4.520 4.196 4.520
dbcsr_matrix_vector_mult 304 9.0 0.006 0.017 3.681 3.983
dbcsr_matrix_vector_mult_local 304 10.0 3.214 3.709 3.216 3.711
dbcsr_sort_data 658 11.4 3.065 3.313 3.065 3.313
ls_scf_post 1 4.0 0.000 0.001 3.239 3.252
dbcsr_special_finalize 555 9.7 0.006 0.007 2.818 3.125
dbcsr_merge_single_wm 555 10.7 0.537 0.658 2.810 3.116
ls_scf_store_result 1 5.0 0.000 0.000 2.964 3.035
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.899 2.949
dbcsr_data_release 10477 10.7 1.587 2.416 1.587 2.416
dbcsr_finalize 304 7.8 0.050 0.061 1.797 1.970
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=96.462000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2706.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/25/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 15.383312E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 336818 0.0% 0.0% 100.0%
average stack size 0.0 0.0 28603.7
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 4.629918E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265558.
MP_Allreduce 3049 15663.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.042 0.054 99.042 99.042
qs_energies 1 2.0 0.000 0.000 98.241 98.246
ls_scf 1 3.0 0.000 0.000 96.319 96.323
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.020 78.216
ls_scf_main 1 4.0 0.000 0.000 62.030 62.031
multiply_cannon 111 7.7 0.095 0.169 55.404 60.973
density_matrix_trs4 2 5.0 0.002 0.002 55.082 55.196
multiply_cannon_loop 111 8.7 0.069 0.076 51.006 52.642
mp_waitall_1 5436 11.0 26.802 32.475 26.802 32.475
ls_scf_init_scf 1 4.0 0.000 0.000 30.710 30.715
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.507 29.565
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.353 27.362
multiply_cannon_multrec 444 9.7 13.885 16.715 21.009 23.181
make_m2s 222 7.7 0.004 0.005 17.862 20.237
make_images 222 8.7 3.724 4.415 17.800 20.177
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.063 16.608
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.488 14.599
make_images_data 222 9.7 0.003 0.004 10.049 12.398
hybrid_alltoall_any 227 10.6 0.786 3.749 9.893 12.314
dbcsr_mm_accdrv_process 3003 10.4 0.165 0.342 6.837 7.995
multiply_cannon_sync_h2d 444 9.7 6.598 7.696 6.598 7.696
dbcsr_mm_accdrv_process_sort 3003 11.4 6.521 7.644 6.521 7.644
mp_allgather_i34 111 8.7 2.637 7.026 2.637 7.026
arnoldi_extremal 4 6.8 0.000 0.000 5.885 5.896
arnoldi_normal_ev 4 7.8 0.002 0.005 5.885 5.896
build_subspace 16 8.4 0.015 0.020 5.488 5.494
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.573 4.716
mp_sum_l 807 5.4 2.798 4.462 2.798 4.462
dbcsr_matrix_vector_mult 304 9.0 0.007 0.017 4.229 4.441
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.658 4.427
mp_irecv_dv 1241 11.2 1.639 4.403 1.639 4.403
dbcsr_matrix_vector_mult_local 304 10.0 3.754 4.233 3.756 4.235
calculate_norms 792 9.8 3.549 3.656 3.549 3.656
ls_scf_post 1 4.0 0.000 0.000 3.579 3.583
make_images_sizes 222 9.7 0.000 0.000 1.166 3.484
mp_alltoall_i44 222 10.7 1.166 3.484 1.166 3.484
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.355 3.433
ls_scf_store_result 1 5.0 0.000 0.000 3.362 3.393
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.884 3.197
dbcsr_finalize 304 7.8 0.062 0.078 2.200 2.309
dbcsr_merge_all 275 8.9 0.474 0.531 2.050 2.144
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=99.042000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3659.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/2216e2348d729b940ade5c6e00d35197f22708d2_performance_tests/26/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 23 x 23 x 23 234439235724792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 234.439236E+12 0.0% 0.0% 100.0%
flops max/rank 30.358840E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 9634225188 0.0% 0.0% 100.0%
number of processed stacks 339931 0.0% 0.0% 100.0%
average stack size 0.0 0.0 28341.7
marketing flops 1.742118E+15
-------------------------------------------------------------------------------
# multiplications 111
max memory usage/rank 8.728912E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284111.
MP_Allreduce 3043 21950.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.080 0.105 108.852 108.852
qs_energies 1 2.0 0.000 0.000 107.428 107.445
ls_scf 1 3.0 0.000 0.000 104.540 104.557
dbcsr_multiply_generic 111 6.7 0.023 0.027 78.370 78.513
ls_scf_main 1 4.0 0.000 0.000 66.675 66.675
density_matrix_trs4 2 5.0 0.002 0.003 57.771 57.833
multiply_cannon 111 7.7 0.212 0.303 50.424 51.986
multiply_cannon_loop 111 8.7 0.067 0.070 46.959 48.145
ls_scf_init_scf 1 4.0 0.000 0.000 34.161 34.161
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.807 32.832
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.980 29.993
mp_waitall_1 4527 11.1 22.928 27.047 22.928 27.047
make_m2s 222 7.7 0.005 0.005 24.095 25.066
make_images 222 8.7 4.593 4.986 23.989 24.957
multiply_cannon_multrec 444 9.7 17.794 18.506 22.448 23.107
hybrid_alltoall_any 227 10.6 1.659 3.610 13.059 15.766
make_images_data 222 9.7 0.003 0.003 13.295 15.736
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.986 11.557
multiply_cannon_sync_h2d 444 9.7 8.818 8.883 8.818 8.883
arnoldi_extremal 4 6.8 0.000 0.000 7.613 7.628
arnoldi_normal_ev 4 7.8 0.003 0.009 7.613 7.628
build_subspace 16 8.4 0.026 0.037 7.032 7.046
dbcsr_matrix_vector_mult 304 9.0 0.009 0.027 5.665 5.784
dbcsr_matrix_vector_mult_local 304 10.0 5.170 5.493 5.172 5.496
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.069 5.333
ls_scf_dm_to_ks 2 5.0 0.000 0.000 4.831 4.922
dbcsr_mm_accdrv_process 1814 10.4 0.184 0.324 4.487 4.619
dbcsr_mm_accdrv_process_sort 1814 11.4 4.189 4.330 4.189 4.330
ls_scf_post 1 4.0 0.000 0.000 3.705 3.722
make_images_sizes 222 9.7 0.000 0.000 1.442 3.548
mp_alltoall_i44 222 10.7 1.442 3.548 1.442 3.548
ls_scf_store_result 1 5.0 0.000 0.000 3.456 3.469
mp_allgather_i34 111 8.7 1.112 3.446 1.112 3.446
calculate_norms 792 9.8 3.267 3.331 3.267 3.331
dbcsr_finalize 304 7.8 0.082 0.090 3.108 3.154
dbcsr_data_release 12724 10.6 2.374 3.036 2.374 3.036
dbcsr_merge_all 275 8.9 0.891 0.918 2.869 2.905
dbcsr_complete_redistribute 5 7.6 1.433 1.464 2.785 2.902
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.858 2.858
matrix_ls_to_qs 2 6.0 0.000 0.000 2.443 2.573
dbcsr_sort_data 325 11.1 2.442 2.498 2.442 2.498
mp_sum_l 807 5.4 1.520 2.331 1.520 2.331
dbcsr_new_transposed 4 7.5 0.243 0.249 2.312 2.321
dbcsr_frobenius_norm 74 6.6 2.056 2.132 2.197 2.245
dbcsr_add_d 103 6.2 0.000 0.000 2.164 2.232
dbcsr_add_anytype 103 7.2 0.860 0.891 2.164 2.231
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.207 2.208
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=108.852000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6911.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 2216e2348d729b940ade5c6e00d35197f22708d2
Summary: empty
Status: OK