=== This is the CP2K Performance-Test ===
Updating b9b716585..b22e23451
Fast-forward
data/xc_section/CAM-B3LYP.sec | 44 +++++++++++++++++++++++++++++++++++++++++++
1 file changed, 44 insertions(+)
create mode 100644 data/xc_section/CAM-B3LYP.sec
Current branch master is up to date.
Already up to date.
Current branch master is up to date.
GIT Revision: b22e23451d9407afd60f6019384990b988ad9a21
################# 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, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.1.0, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.1,
# SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Author: Matthias Krack (03.02.2023)
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${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=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2022.11.001
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.1.0
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.1
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/01
job id: 45396981
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/02
job id: 45396983
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/03
job id: 45396985
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/04
job id: 45396986
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/05
job id: 45396987
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/06
job id: 45396988
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/07
job id: 45396989
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/08
job id: 45396994
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/09
job id: 45396999
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/10
job id: 45397003
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/11
job id: 45397007
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/12
job id: 45397010
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/13
job id: 45397012
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/14
job id: 45397014
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/15
job id: 45397016
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/16
job id: 45397018
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/17
job id: 45397022
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/18
job id: 45397023
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/19
job id: 45397025
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/20
job id: 45397027
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/21
job id: 45397029
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/22
job id: 45397031
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/23
job id: 45397032
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/24
job id: 45397033
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/25
job id: 45397034
--- 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/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/26
job id: 45397035
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: 512 H2O (4 NVE MD steps on 64 nodes)
input file: benchmarks/QS/00512_H2O/H2O-512_md.inp
required files: []
output file: result.log
# nodes = 64
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/27
job id: 45397036
--- Point ---
name: 601
plot: h2o_512_md
regex: CP2K
label: (64n/12r/1t)
--- Point ---
name: 602
plot: h2o_512_md_mem
regex: Estimated peak process memory
label: (64n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 0.023 0.037 142.202 142.203
farming_run 1 2.0 141.575 141.580 142.165 142.169
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.460863E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1103589.
MP_Allreduce 491 2254389.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.026 118.045 118.045
qs_energies 1 2.0 0.000 0.000 117.735 117.738
mp2_main 1 3.0 0.000 0.000 114.611 114.614
mp2_gpw_main 1 4.0 0.037 0.043 113.373 113.376
mp2_ri_gpw_compute_in 1 5.0 0.180 0.182 93.595 94.113
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 55.383 55.901
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.168 41.706 47.162
get_2c_integrals 1 6.0 0.000 0.000 37.164 38.031
integrate_v_rspace 273 8.0 0.434 0.449 25.112 30.309
pw_transfer 6555 10.6 0.373 0.384 27.341 27.860
grid_integrate_task_list 273 9.0 20.929 26.563 20.929 26.563
fft_wrap_pw1pw2 5465 11.4 0.046 0.049 26.007 26.539
fft_wrap_pw1pw2_100 2178 12.4 1.162 1.266 23.481 24.020
rpa_ri_compute_en 1 5.0 0.001 0.001 19.650 19.778
compute_2c_integrals 1 7.0 0.002 0.003 19.482 19.482
compute_2c_integrals_loop_lm 1 8.0 0.008 0.009 18.838 19.016
mp2_eri_2c_integrate_gpw 1 9.0 2.400 2.444 18.829 19.007
cp_fm_cholesky_decompose 12 8.2 17.730 18.679 17.730 18.679
cholesky_decomp 1 7.0 0.000 0.000 16.535 17.411
fft3d_s 5443 13.4 16.118 16.635 16.140 16.654
ao_to_mo_and_store_B_mult_1 272 7.0 10.850 15.592 10.850 15.592
calculate_wavefunction 272 8.0 5.417 5.574 12.516 13.162
rpa_num_int 1 6.0 0.000 0.001 10.640 10.641
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.522 10.611
calc_mat_Q 8 8.0 0.000 0.000 9.326 9.423
contract_S_to_Q 8 9.0 0.000 0.000 8.747 8.845
calc_potential_gpw 544 9.5 0.005 0.006 8.245 8.632
parallel_gemm_fm 14 9.1 0.000 0.000 8.330 8.441
parallel_gemm_fm_cosma 14 10.1 8.330 8.441 8.330 8.441
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.174 8.409
create_integ_mat 1 6.0 0.022 0.028 8.395 8.396
potential_pw2rs 545 10.0 0.106 0.108 7.662 8.354
collocate_single_gaussian 272 10.0 0.040 0.042 7.423 7.738
array2fm 1 7.0 0.000 0.000 6.682 7.009
pw_scatter_s 2720 13.7 4.443 4.662 4.443 4.662
pw_gather_s 2722 13.2 3.829 4.157 3.829 4.157
array2fm_buffer_send 1 8.0 2.946 3.163 2.946 3.163
scf_env_do_scf 1 3.0 0.000 0.000 2.692 2.692
init_scf_loop 1 4.0 0.000 0.000 2.505 2.505
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=113.376143, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2732.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 0.033 0.042 413.572 413.574
farming_run 1 2.0 412.399 412.426 413.486 413.490
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.225863E+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 743 386399.
MP_Allreduce 1941 22272.
MP_Sync 37
MP_Alltoall 77 2368424.
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 0.018 0.055 213.311 213.312
qs_energies 1 2.0 0.003 0.023 212.991 212.997
scf_env_do_scf 1 3.0 0.000 0.000 107.325 107.326
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.754 105.763
rebuild_ks_matrix 4 6.0 0.000 0.000 105.752 105.762
qs_ks_build_kohn_sham_matrix 4 7.0 0.069 0.079 105.752 105.762
mp2_main 1 3.0 0.002 0.012 105.250 105.259
hfx_ks_matrix 4 8.0 0.001 0.001 105.224 105.227
integrate_four_center 4 9.0 0.144 0.458 105.223 105.227
mp2_gpw_main 1 4.0 0.039 0.074 103.138 103.154
integrate_four_center_main 4 10.0 0.121 0.595 96.999 98.962
integrate_four_center_bin 267 11.0 96.878 98.707 96.878 98.707
init_scf_loop 1 4.0 0.000 0.000 92.373 92.374
mp2_ri_gpw_compute_in 1 5.0 0.065 0.072 75.121 76.143
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.003 54.459 55.472
mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.165 42.079 47.254
integrate_v_rspace 95 8.0 0.398 0.563 28.578 33.579
pw_transfer 2240 10.6 0.144 0.162 30.042 30.437
fft_wrap_pw1pw2 1868 11.4 0.018 0.021 29.057 29.464
mp2_ri_gpw_compute_en 1 5.0 0.056 0.065 27.750 29.415
grid_integrate_task_list 95 9.0 23.752 28.975 23.752 28.975
fft_wrap_pw1pw2_100 730 12.4 1.275 1.447 26.640 27.012
ao_to_mo_and_store_B_mult_1 91 7.0 10.700 26.969 10.700 26.969
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.853 1.908 26.092 26.103
get_2c_integrals 1 6.0 0.000 0.000 20.327 20.604
compute_2c_integrals 1 7.0 0.003 0.004 19.279 19.295
compute_2c_integrals_loop_lm 1 8.0 0.002 0.008 18.850 19.050
mp2_eri_2c_integrate_gpw 1 9.0 1.736 1.856 18.848 19.049
fft3d_s 1823 13.4 18.426 18.860 18.439 18.873
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.947 14.947
calculate_wavefunction 91 8.0 2.012 2.038 9.737 9.955
mp2_ri_gpw_compute_en_comm 22 7.0 0.504 0.517 9.104 9.442
potential_pw2rs 186 10.0 0.033 0.035 8.733 9.287
mp2_ri_gpw_compute_en_expansio 172 7.0 0.551 0.578 8.748 9.132
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.235 8.573
local_gemm 172 8.0 8.196 8.558 8.196 8.558
calc_potential_gpw 182 9.5 0.002 0.002 7.923 8.158
collocate_single_gaussian 91 10.0 0.019 0.050 7.875 8.073
mp_sendrecv_dm3 2068 8.0 7.121 7.488 7.121 7.488
mp2_ri_gpw_compute_en_ener 172 7.0 6.342 6.440 6.342 6.440
pw_gather_s 912 13.2 4.903 5.389 4.903 5.389
mp_sync 37 10.5 2.936 4.999 2.936 4.999
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=103.144269, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1513.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/03/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 29.277748E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 5055360 0.0% 0.0% 100.0%
average stack size 0.0 0.0 29.1
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 452.706304E+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 272.
MP_Sync 530
MP_Alltoall 2083 578578.
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 0.016 0.035 54.255 54.257
qs_mol_dyn_low 1 2.0 0.003 0.004 53.974 53.987
qs_forces 11 3.9 0.002 0.003 53.481 53.482
qs_energies 11 4.9 0.001 0.002 51.871 51.884
scf_env_do_scf 11 5.9 0.001 0.002 45.654 45.654
scf_env_do_scf_inner_loop 108 6.5 0.002 0.008 43.560 43.560
dbcsr_multiply_generic 2286 12.5 0.092 0.095 32.942 33.415
qs_scf_new_mos 108 7.5 0.000 0.001 33.003 33.291
qs_scf_loop_do_ot 108 8.5 0.000 0.001 33.003 33.291
ot_scf_mini 108 9.5 0.002 0.002 31.382 31.567
multiply_cannon 2286 13.5 0.187 0.191 25.813 27.741
multiply_cannon_loop 2286 14.5 1.447 1.528 24.942 26.885
velocity_verlet 10 3.0 0.015 0.024 25.687 25.692
ot_mini 108 10.5 0.001 0.001 19.026 19.272
qs_ot_get_derivative 108 11.5 0.001 0.002 16.133 16.306
mp_waitall_1 245248 16.5 7.981 14.337 7.981 14.337
multiply_cannon_metrocomm3 54864 15.5 0.069 0.074 5.522 12.882
multiply_cannon_multrec 54864 15.5 4.347 6.763 7.826 11.387
rebuild_ks_matrix 119 8.3 0.000 0.000 8.206 8.340
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.206 8.339
qs_ot_get_p 119 10.4 0.001 0.001 7.824 8.133
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.122 7.242
multiply_cannon_sync_h2d 54864 15.5 6.058 6.731 6.058 6.731
mp_sum_l 7207 12.9 4.717 6.418 4.717 6.418
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.679 5.778
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.123 5.595
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.219 5.272
init_scf_run 11 5.9 0.000 0.001 5.004 5.004
scf_env_initial_rho_setup 11 6.9 0.000 0.000 5.003 5.004
dbcsr_mm_accdrv_process 76910 16.1 1.094 1.772 3.400 4.793
sum_up_and_integrate 119 10.3 0.012 0.014 4.742 4.749
integrate_v_rspace 119 11.3 0.002 0.003 4.731 4.738
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.468 4.604
calculate_rho_elec 119 8.7 0.012 0.017 4.467 4.603
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.491 4.492
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.278 4.278
cp_fm_redistribute_end 50 14.0 2.182 4.251 2.188 4.255
cp_fm_diag_elpa_base 50 14.0 2.060 4.148 2.065 4.157
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.928 3.125
apply_single 119 13.6 0.000 0.000 2.928 3.125
calculate_dm_sparse 119 9.5 0.000 0.001 2.960 3.109
jit_kernel_multiply 13 15.8 2.244 2.933 2.244 2.933
rs_pw_transfer 974 11.9 0.012 0.013 2.802 2.930
multiply_cannon_metrocomm1 54864 15.5 0.053 0.057 1.671 2.886
calculate_first_density_matrix 1 7.0 0.001 0.001 2.785 2.788
ot_diis_step 108 11.5 0.006 0.006 2.644 2.645
density_rs2pw 119 9.7 0.004 0.004 2.469 2.607
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.499 2.500
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.311 2.377
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.200 2.276
pw_transfer 1439 11.6 0.053 0.058 2.165 2.243
make_m2s 4572 13.5 0.054 0.056 2.101 2.181
acc_transpose_blocks 54864 15.5 0.217 0.248 1.693 2.172
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 2.088 2.170
wfi_extrapolate 11 7.9 0.001 0.001 2.156 2.156
grid_integrate_task_list 119 12.3 2.044 2.134 2.044 2.134
make_images 4572 14.5 0.133 0.138 2.019 2.098
init_scf_loop 11 6.9 0.000 0.000 2.077 2.078
potential_pw2rs 119 12.3 0.004 0.004 2.045 2.051
fft3d_ps 1201 14.6 0.369 0.472 1.859 1.933
mp_sum_d 4135 12.0 1.193 1.842 1.193 1.842
fft_wrap_pw1pw2_140 487 13.2 0.082 0.095 1.645 1.729
mp_alltoall_d11v 2130 13.8 1.422 1.697 1.422 1.697
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.500 1.525
grid_collocate_task_list 119 9.7 1.291 1.378 1.291 1.378
mp_waitany 12084 13.8 1.239 1.362 1.239 1.362
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.133 1.147
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=54.257000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=431.272727, yerr=0.962091
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 489.082880E+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 54
MP_Alltoall 2060 1031878.
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 0.036 0.067 65.484 65.485
qs_mol_dyn_low 1 2.0 0.003 0.004 65.085 65.092
qs_forces 11 3.9 0.005 0.023 65.001 65.007
qs_energies 11 4.9 0.003 0.009 62.701 62.714
scf_env_do_scf 11 5.9 0.001 0.002 55.333 55.333
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 49.102 49.105
qs_scf_new_mos 108 7.5 0.001 0.001 32.502 32.812
qs_scf_loop_do_ot 108 8.5 0.001 0.001 32.502 32.812
velocity_verlet 10 3.0 0.001 0.002 32.585 32.586
dbcsr_multiply_generic 2286 12.5 0.101 0.105 30.115 31.394
ot_scf_mini 108 9.5 0.003 0.003 31.126 31.329
multiply_cannon 2286 13.5 0.212 0.223 20.195 24.314
multiply_cannon_loop 2286 14.5 0.902 0.978 18.013 22.053
ot_mini 108 10.5 0.002 0.011 18.214 18.555
mp_waitall_1 200699 16.5 12.210 16.806 12.210 16.806
qs_ot_get_derivative 108 11.5 0.001 0.001 14.883 15.096
multiply_cannon_metrocomm3 27432 15.5 0.067 0.071 6.799 12.947
rebuild_ks_matrix 119 8.3 0.000 0.000 11.850 12.269
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.015 11.849 12.269
qs_ks_update_qs_env 119 7.6 0.001 0.001 10.539 10.927
qs_ot_get_p 119 10.4 0.001 0.002 8.976 9.313
rs_pw_transfer 974 11.9 0.010 0.011 8.086 8.576
multiply_cannon_multrec 27432 15.5 1.823 3.946 5.766 8.456
sum_up_and_integrate 119 10.3 0.024 0.027 8.080 8.097
integrate_v_rspace 119 11.3 0.002 0.003 8.055 8.071
qs_rho_update_rho_low 119 7.7 0.001 0.001 7.926 7.979
calculate_rho_elec 119 8.7 0.021 0.024 7.925 7.978
mp_sum_l 7207 12.9 4.213 6.860 4.213 6.860
density_rs2pw 119 9.7 0.004 0.004 6.082 6.593
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.649 6.513
init_scf_loop 11 6.9 0.001 0.006 6.130 6.134
make_m2s 4572 13.5 0.052 0.054 5.266 5.958
make_images 4572 14.5 0.201 0.235 5.179 5.868
dbcsr_mm_accdrv_process 47894 16.0 3.061 5.235 3.873 5.730
init_scf_run 11 5.9 0.000 0.001 5.724 5.725
scf_env_initial_rho_setup 11 6.9 0.000 0.002 5.724 5.725
qs_ot_p2m_diag 50 11.0 0.009 0.013 5.666 5.701
potential_pw2rs 119 12.3 0.006 0.006 5.457 5.518
cp_dbcsr_syevd 50 12.0 0.003 0.003 5.029 5.029
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.625 5.015
apply_single 119 13.6 0.000 0.000 3.624 5.015
prepare_preconditioner 11 7.9 0.000 0.001 4.637 4.670
make_preconditioner 11 8.9 0.000 0.001 4.637 4.669
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.449 4.541
pw_transfer 1439 11.6 0.065 0.071 4.192 4.265
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 4.099 4.176
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.073 4.076
cp_fm_redistribute_end 50 14.0 2.039 3.947 2.066 3.959
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 3.798 3.916
fft3d_ps 1201 14.6 0.525 0.579 3.802 3.875
cp_fm_diag_elpa_base 50 14.0 1.818 3.683 1.868 3.771
make_images_data 4572 15.5 0.044 0.050 2.219 3.603
ot_diis_step 108 11.5 0.011 0.015 3.133 3.135
mp_sendrecv_dv 16779 12.7 3.048 3.113 3.048 3.113
wfi_extrapolate 11 7.9 0.001 0.001 3.051 3.051
make_images_sizes 4572 15.5 0.005 0.005 2.228 3.030
mp_alltoall_i44 4572 16.5 2.223 3.026 2.223 3.026
mp_alltoall_d11v 2130 13.8 2.779 2.977 2.779 2.977
mp_allgather_i34 2286 14.5 1.616 2.827 1.616 2.827
calculate_dm_sparse 119 9.5 0.000 0.001 2.675 2.806
hybrid_alltoall_any 4725 16.4 0.052 0.114 1.904 2.784
multiply_cannon_sync_h2d 27432 15.5 1.928 2.619 1.928 2.619
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.583 2.587
calculate_first_density_matrix 1 7.0 0.001 0.003 2.562 2.565
cp_fm_cholesky_invert 11 10.9 2.519 2.533 2.519 2.533
mp_alltoall_z22v 1201 16.6 2.311 2.469 2.311 2.469
fft_wrap_pw1pw2_140 487 13.2 0.077 0.083 2.169 2.247
jit_kernel_multiply 11 16.0 0.758 2.170 0.758 2.170
rs_pw_transfer_RS2PW_50 119 11.7 0.264 0.283 2.029 2.122
rs_pw_transfer_PW2RS_50 119 14.3 0.584 0.603 1.925 2.038
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.889 2.024
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.894 2.023
dbcsr_complete_redistribute 329 12.2 0.077 0.094 1.882 2.018
mp_sum_d 4135 12.0 1.300 1.988 1.300 1.988
grid_integrate_task_list 119 12.3 1.824 1.931 1.824 1.931
multiply_cannon_metrocomm1 27432 15.5 0.030 0.034 0.812 1.815
copy_dbcsr_to_fm 153 11.3 0.002 0.002 1.489 1.608
acc_transpose_blocks 27432 15.5 0.108 0.113 1.226 1.546
mp_waitany 5720 13.7 0.997 1.540 0.997 1.540
cp_fm_cholesky_decompose 22 10.9 1.502 1.508 1.502 1.508
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.475 1.501
dbcsr_dot_sd 1205 11.9 0.072 0.089 0.875 1.443
arnoldi_extremal 119 11.4 0.002 0.004 1.317 1.432
arnoldi_normal_ev 119 12.4 0.008 0.010 1.314 1.430
rs_pw_transfer_RS2PW_140 130 11.5 0.140 0.148 0.914 1.422
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.380 1.414
yz_to_x 368 15.5 0.110 0.130 1.282 1.385
grid_collocate_task_list 119 9.7 1.241 1.370 1.241 1.370
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=65.485000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.090909, yerr=1.504813
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 523.362304E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63497.
MP_Allreduce 10075 307.
MP_Sync 54
MP_Alltoall 1821 708770.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.040 0.100 39.835 39.838
qs_mol_dyn_low 1 2.0 0.004 0.008 39.428 39.433
qs_forces 11 3.9 0.004 0.009 37.321 37.324
qs_energies 11 4.9 0.004 0.007 35.730 35.738
scf_env_do_scf 11 5.9 0.001 0.002 28.165 28.165
scf_env_do_scf_inner_loop 108 6.5 0.005 0.008 25.557 25.558
dbcsr_multiply_generic 2286 12.5 0.094 0.096 20.325 20.439
velocity_verlet 10 3.0 0.001 0.002 18.054 18.119
multiply_cannon 2286 13.5 0.197 0.206 16.351 17.076
qs_scf_new_mos 108 7.5 0.001 0.001 16.867 16.892
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.867 16.891
multiply_cannon_loop 2286 14.5 0.633 0.657 15.437 16.168
ot_scf_mini 108 9.5 0.003 0.003 16.119 16.134
ot_mini 108 10.5 0.001 0.001 9.703 9.717
multiply_cannon_multrec 18288 15.5 1.952 2.833 7.071 8.811
qs_ot_get_derivative 108 11.5 0.001 0.001 8.071 8.084
rebuild_ks_matrix 119 8.3 0.000 0.000 6.818 6.837
qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.020 6.818 6.837
mp_waitall_1 158411 16.6 4.732 6.237 4.732 6.237
init_scf_run 11 5.9 0.000 0.001 6.191 6.191
scf_env_initial_rho_setup 11 6.9 0.001 0.002 6.190 6.191
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.054 6.070
dbcsr_mm_accdrv_process 38222 16.0 4.332 5.715 5.035 5.976
calculate_first_density_matrix 1 7.0 0.002 0.004 4.927 4.929
sum_up_and_integrate 119 10.3 0.030 0.033 4.348 4.354
multiply_cannon_metrocomm3 18288 15.5 0.044 0.045 2.492 4.335
integrate_v_rspace 119 11.3 0.003 0.004 4.317 4.327
calculate_dm_sparse 119 9.5 0.000 0.001 4.286 4.297
qs_ot_get_p 119 10.4 0.001 0.005 4.074 4.126
acc_transpose_blocks 18288 15.5 0.075 0.076 2.494 3.754
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 3.664 3.671
qs_rho_update_rho_low 119 7.7 0.001 0.002 3.593 3.601
calculate_rho_elec 119 8.7 0.031 0.032 3.593 3.600
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.803 3.443
acc_transpose_blocks_kernels 18288 16.5 0.211 0.219 2.049 3.298
jit_kernel_transpose 5 15.6 1.838 3.083 1.838 3.083
rs_pw_transfer 974 11.9 0.009 0.010 2.680 2.937
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.736 2.761
multiply_cannon_metrocomm1 18288 15.5 0.028 0.029 0.979 2.731
init_scf_loop 11 6.9 0.003 0.006 2.576 2.578
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.148 2.456
apply_single 119 13.6 0.000 0.000 2.148 2.456
make_m2s 4572 13.5 0.044 0.045 2.214 2.407
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.380 2.380
density_rs2pw 119 9.7 0.004 0.006 2.106 2.341
make_images 4572 14.5 0.190 0.202 2.129 2.321
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.142 2.150
potential_pw2rs 119 12.3 0.007 0.008 2.108 2.117
pw_transfer 1439 11.6 0.066 0.071 2.099 2.110
mp_sum_dm 438 4.9 2.028 2.098 2.028 2.098
md_write_output 11 3.9 0.018 0.553 0.066 2.073
md_output 10 3.0 0.000 0.000 0.080 2.072
update_particle_set 20 4.0 0.000 0.000 2.005 2.070
jit_kernel_multiply 10 16.0 0.649 2.063 0.649 2.063
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.035 2.043
cp_fm_diag_elpa_base 50 14.0 2.008 2.021 2.033 2.041
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.005 2.018
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.920 1.922
grid_integrate_task_list 119 12.3 1.805 1.900 1.805 1.900
mp_sum_l 7207 12.9 1.423 1.790 1.423 1.790
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.756 1.763
prepare_preconditioner 11 7.9 0.000 0.000 1.731 1.734
make_preconditioner 11 8.9 0.002 0.004 1.731 1.734
fft3d_ps 1201 14.6 0.530 0.549 1.683 1.697
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.587 1.672
ot_diis_step 108 11.5 0.011 0.011 1.598 1.598
multiply_cannon_sync_h2d 18288 15.5 1.369 1.536 1.369 1.536
write_trajectory 44 4.9 0.048 1.513 0.048 1.520
fft_wrap_pw1pw2_140 487 13.2 0.092 0.095 1.504 1.518
grid_collocate_task_list 119 9.7 1.216 1.374 1.216 1.374
make_images_data 4572 15.5 0.044 0.047 1.020 1.243
wfi_extrapolate 11 7.9 0.001 0.001 1.196 1.197
hybrid_alltoall_any 4725 16.4 0.055 0.112 0.887 1.126
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 1.050 1.053
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.002 1.030
mp_waitany 9880 13.7 0.621 0.899 0.621 0.899
rs_pw_transfer_RS2PW_140 130 11.5 0.125 0.142 0.589 0.849
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.838 0.841
mp_alltoall_d11v 2130 13.8 0.722 0.817 0.722 0.817
mp_alltoall_z22v 1201 16.6 0.720 0.813 0.720 0.813
mp_allgather_i34 2286 14.5 0.372 0.802 0.372 0.802
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=39.838000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=498.272727, yerr=1.958242
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 545.513472E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63496.
MP_Allreduce 10074 349.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.032 0.043 56.552 56.553
qs_mol_dyn_low 1 2.0 0.003 0.007 56.136 56.145
qs_forces 11 3.9 0.002 0.002 56.050 56.053
qs_energies 11 4.9 0.001 0.001 53.917 53.925
scf_env_do_scf 11 5.9 0.001 0.002 47.294 47.295
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 40.848 40.850
velocity_verlet 10 3.0 0.002 0.002 29.940 29.942
qs_scf_new_mos 108 7.5 0.001 0.001 26.727 26.812
qs_scf_loop_do_ot 108 8.5 0.001 0.001 26.726 26.811
ot_scf_mini 108 9.5 0.002 0.003 25.465 25.513
dbcsr_multiply_generic 2286 12.5 0.099 0.101 24.850 25.189
multiply_cannon 2286 13.5 0.234 0.273 17.162 18.328
multiply_cannon_loop 2286 14.5 0.943 0.978 15.339 16.355
ot_mini 108 10.5 0.001 0.001 14.766 14.858
qs_ot_get_derivative 108 11.5 0.003 0.011 12.486 12.543
rebuild_ks_matrix 119 8.3 0.000 0.000 9.858 10.019
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 9.858 10.019
multiply_cannon_multrec 27432 15.5 2.300 2.933 8.647 8.964
qs_ks_update_qs_env 119 7.6 0.001 0.001 8.815 8.962
mp_waitall_1 137007 16.6 6.868 8.401 6.868 8.401
dbcsr_mm_accdrv_process 47916 15.9 5.210 7.029 6.252 7.457
qs_ot_get_p 119 10.4 0.001 0.001 7.227 7.333
qs_rho_update_rho_low 119 7.7 0.001 0.001 6.706 6.738
calculate_rho_elec 119 8.7 0.040 0.046 6.706 6.737
init_scf_loop 11 6.9 0.000 0.000 6.382 6.386
sum_up_and_integrate 119 10.3 0.035 0.037 6.342 6.350
rs_pw_transfer 974 11.9 0.009 0.010 6.167 6.340
integrate_v_rspace 119 11.3 0.003 0.003 6.307 6.316
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 4.772 5.251
make_m2s 4572 13.5 0.055 0.056 4.948 5.229
prepare_preconditioner 11 7.9 0.000 0.000 5.171 5.185
make_preconditioner 11 8.9 0.000 0.000 5.171 5.185
make_images 4572 14.5 0.272 0.334 4.840 5.118
density_rs2pw 119 9.7 0.004 0.004 4.939 5.101
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.641 5.044
init_scf_run 11 5.9 0.000 0.001 4.941 4.942
scf_env_initial_rho_setup 11 6.9 0.000 0.000 4.941 4.941
qs_ot_p2m_diag 50 11.0 0.015 0.023 4.808 4.818
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.312 4.313
potential_pw2rs 119 12.3 0.009 0.009 3.867 3.880
cp_fm_diag_elpa 50 13.0 0.000 0.000 3.679 3.688
cp_fm_diag_elpa_base 50 14.0 3.618 3.646 3.676 3.686
pw_transfer 1439 11.6 0.067 0.072 3.508 3.551
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 1.900 3.522
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 3.414 3.461
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 3.352 3.386
mp_sum_l 7207 12.9 2.332 3.242 2.332 3.242
fft3d_ps 1201 14.6 0.586 0.643 3.090 3.135
make_images_data 4572 15.5 0.045 0.048 2.393 2.874
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.464 2.863
apply_single 119 13.6 0.000 0.000 2.464 2.863
calculate_dm_sparse 119 9.5 0.000 0.000 2.430 2.512
wfi_extrapolate 11 7.9 0.001 0.001 2.476 2.477
fft_wrap_pw1pw2_140 487 13.2 0.086 0.092 2.423 2.471
hybrid_alltoall_any 4725 16.4 0.062 0.150 2.006 2.426
calculate_first_density_matrix 1 7.0 0.001 0.001 2.384 2.387
mp_sendrecv_dv 8211 12.7 2.307 2.327 2.307 2.327
cp_fm_cholesky_invert 11 10.9 2.232 2.244 2.232 2.244
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.233 2.237
make_images_sizes 4572 15.5 0.005 0.005 1.528 2.232
mp_alltoall_i44 4572 16.5 1.524 2.227 1.524 2.227
ot_diis_step 108 11.5 0.012 0.013 2.179 2.180
mp_alltoall_z22v 1201 16.6 2.088 2.131 2.088 2.131
mp_alltoall_d11v 2130 13.8 1.902 2.057 1.902 2.057
grid_integrate_task_list 119 12.3 1.798 1.906 1.798 1.906
dbcsr_complete_redistribute 329 12.2 0.124 0.172 1.567 1.887
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.787 1.835
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.696 1.749
jit_kernel_multiply 9 15.8 0.979 1.704 0.979 1.704
multiply_cannon_metrocomm1 27432 15.5 0.033 0.034 0.940 1.699
rs_pw_transfer_RS2PW_50 119 11.7 0.227 0.231 1.665 1.688
acc_transpose_blocks 27432 15.5 0.112 0.115 1.477 1.500
cp_fm_upper_to_full 72 14.2 0.960 1.421 0.960 1.421
mp_allgather_i34 2286 14.5 1.111 1.365 1.111 1.365
grid_collocate_task_list 119 9.7 1.221 1.364 1.221 1.364
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.342 1.355
cp_fm_cholesky_decompose 22 10.9 1.302 1.308 1.302 1.308
yz_to_x 606 15.1 0.188 0.221 1.277 1.298
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.958 1.280
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.238 1.241
rs_pw_transfer_PW2RS_50 119 14.3 0.312 0.323 1.179 1.237
x_to_yz 595 16.2 0.216 0.248 1.216 1.224
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.142 1.156
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=56.553000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=517.909091, yerr=3.058655
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 614.617088E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63495.
MP_Allreduce 10074 348.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.047 0.058 42.675 42.677
qs_mol_dyn_low 1 2.0 0.003 0.004 42.323 42.343
qs_forces 11 3.9 0.002 0.003 42.246 42.248
qs_energies 11 4.9 0.001 0.002 40.304 40.309
scf_env_do_scf 11 5.9 0.001 0.002 34.540 34.540
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 30.021 30.022
velocity_verlet 10 3.0 0.002 0.002 22.181 22.183
qs_scf_new_mos 108 7.5 0.001 0.001 17.530 17.591
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.529 17.590
dbcsr_multiply_generic 2286 12.5 0.092 0.095 17.265 17.457
ot_scf_mini 108 9.5 0.002 0.002 16.624 16.676
multiply_cannon 2286 13.5 0.235 0.247 11.943 12.811
multiply_cannon_loop 2286 14.5 0.331 0.342 10.721 11.142
ot_mini 108 10.5 0.001 0.001 9.276 9.338
rebuild_ks_matrix 119 8.3 0.000 0.000 9.185 9.254
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 9.185 9.254
qs_ks_update_qs_env 119 7.6 0.001 0.001 8.345 8.408
qs_ot_get_derivative 108 11.5 0.001 0.002 7.635 7.685
mp_waitall_1 115863 16.7 5.837 6.527 5.837 6.527
sum_up_and_integrate 119 10.3 0.038 0.042 6.212 6.217
multiply_cannon_multrec 9144 15.5 1.657 1.860 5.997 6.208
integrate_v_rspace 119 11.3 0.003 0.003 6.174 6.179
qs_rho_update_rho_low 119 7.7 0.001 0.001 5.537 5.579
calculate_rho_elec 119 8.7 0.060 0.061 5.536 5.578
qs_ot_get_p 119 10.4 0.001 0.001 5.028 5.084
rs_pw_transfer 974 11.9 0.008 0.008 4.554 4.739
init_scf_loop 11 6.9 0.000 0.000 4.490 4.491
dbcsr_mm_accdrv_process 12550 15.8 3.282 4.058 4.239 4.322
init_scf_run 11 5.9 0.000 0.001 4.176 4.176
scf_env_initial_rho_setup 11 6.9 0.000 0.000 4.175 4.176
make_m2s 4572 13.5 0.034 0.036 3.788 4.037
make_images 4572 14.5 0.270 0.304 3.697 3.946
density_rs2pw 119 9.7 0.004 0.004 3.614 3.772
potential_pw2rs 119 12.3 0.010 0.011 3.549 3.577
pw_transfer 1439 11.6 0.067 0.070 3.404 3.423
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 3.309 3.329
prepare_preconditioner 11 7.9 0.000 0.000 3.317 3.326
make_preconditioner 11 8.9 0.000 0.000 3.317 3.326
qs_ot_p2m_diag 50 11.0 0.022 0.023 3.262 3.267
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.113 3.177
fft3d_ps 1201 14.6 0.579 0.591 2.968 2.984
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.911 2.912
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.822 2.852
make_images_data 4572 15.5 0.038 0.042 1.937 2.554
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.463 2.472
cp_fm_diag_elpa_base 50 14.0 2.418 2.450 2.461 2.470
fft_wrap_pw1pw2_140 487 13.2 0.087 0.090 2.420 2.442
hybrid_alltoall_any 4725 16.4 0.063 0.177 1.723 2.417
calculate_first_density_matrix 1 7.0 0.009 0.022 2.396 2.397
multiply_cannon_metrocomm1 9144 15.5 0.022 0.023 2.063 2.356
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.260 2.284
calculate_dm_sparse 119 9.5 0.000 0.000 2.057 2.096
grid_integrate_task_list 119 12.3 1.948 2.041 1.948 2.041
make_images_sizes 4572 15.5 0.005 0.005 1.183 1.991
mp_alltoall_i44 4572 16.5 1.178 1.986 1.178 1.986
jit_kernel_multiply 10 15.7 0.918 1.918 0.918 1.918
mp_alltoall_z22v 1201 16.6 1.840 1.895 1.840 1.895
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.766 1.769
mp_alltoall_d11v 2130 13.8 1.588 1.709 1.588 1.709
wfi_extrapolate 11 7.9 0.001 0.001 1.704 1.704
ot_diis_step 108 11.5 0.013 0.013 1.590 1.590
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.493 1.509
cp_fm_cholesky_invert 11 10.9 1.494 1.499 1.494 1.499
grid_collocate_task_list 119 9.7 1.347 1.474 1.347 1.474
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.399 1.457
apply_single 119 13.6 0.000 0.000 1.399 1.457
mp_allgather_i34 2286 14.5 0.544 1.374 0.544 1.374
mp_sum_l 7207 12.9 1.106 1.373 1.106 1.373
cp_fm_cholesky_decompose 22 10.9 1.295 1.302 1.295 1.302
mp_sendrecv_dv 5355 12.7 1.273 1.300 1.273 1.300
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.276 1.286
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.271 1.273
x_to_yz 595 16.2 0.280 0.291 1.261 1.272
rs_pw_transfer_PW2RS_50 119 14.3 0.216 0.225 1.066 1.201
yz_to_x 606 15.1 0.258 0.269 1.117 1.142
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.985 0.995
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.964 0.970
acc_transpose_blocks 9144 15.5 0.037 0.038 0.954 0.961
rs_pw_transfer_RS2PW_50 119 11.7 0.199 0.208 0.866 0.941
mp_waitany 5200 13.7 0.788 0.937 0.788 0.937
dbcsr_complete_redistribute 329 12.2 0.184 0.193 0.879 0.930
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.868 0.920
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=42.677000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=580.272727, yerr=6.282601
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 731.312128E+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 0.034 0.059 43.758 43.759
qs_mol_dyn_low 1 2.0 0.003 0.004 43.461 43.469
qs_forces 11 3.9 0.002 0.002 43.394 43.394
qs_energies 11 4.9 0.001 0.001 41.399 41.402
scf_env_do_scf 11 5.9 0.003 0.018 35.326 35.326
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 27.491 27.492
velocity_verlet 10 3.0 0.014 0.015 24.415 24.420
dbcsr_multiply_generic 2286 12.5 0.098 0.101 18.351 18.560
qs_scf_new_mos 108 7.5 0.001 0.001 16.841 16.944
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.840 16.944
ot_scf_mini 108 9.5 0.002 0.002 15.718 15.826
multiply_cannon 2286 13.5 0.303 0.318 14.070 15.036
multiply_cannon_loop 2286 14.5 0.344 0.353 12.712 13.667
ot_mini 108 10.5 0.001 0.001 9.451 9.579
multiply_cannon_multrec 9144 15.5 3.379 4.678 8.663 8.768
rebuild_ks_matrix 119 8.3 0.000 0.000 7.674 7.811
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.674 7.810
init_scf_loop 11 6.9 0.000 0.000 7.788 7.789
qs_ot_get_derivative 108 11.5 0.016 0.065 7.366 7.469
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.978 7.102
prepare_preconditioner 11 7.9 0.000 0.000 6.806 6.820
make_preconditioner 11 8.9 0.000 0.000 6.806 6.820
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.366 6.697
dbcsr_mm_accdrv_process 12550 15.8 4.006 5.381 5.160 6.489
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.757 4.764
calculate_rho_elec 119 8.7 0.118 0.121 4.756 4.764
cp_fm_upper_to_full 72 14.2 3.146 4.533 3.146 4.533
sum_up_and_integrate 119 10.3 0.066 0.067 4.275 4.281
integrate_v_rspace 119 11.3 0.003 0.004 4.209 4.217
init_scf_run 11 5.9 0.000 0.001 3.962 3.962
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.962 3.962
mp_waitall_1 94719 16.7 2.974 3.928 2.974 3.928
qs_ot_get_p 119 10.4 0.001 0.001 3.404 3.544
pw_transfer 1439 11.6 0.069 0.070 3.105 3.113
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 3.006 3.014
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.585 2.995
dbcsr_complete_redistribute 329 12.2 0.286 0.292 2.031 2.853
make_m2s 4572 13.5 0.038 0.038 2.598 2.829
make_images 4572 14.5 0.352 0.385 2.477 2.709
density_rs2pw 119 9.7 0.004 0.004 2.650 2.672
fft3d_ps 1201 14.6 0.610 0.620 2.630 2.638
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.685 2.498
fft_wrap_pw1pw2_140 487 13.2 0.095 0.097 2.471 2.479
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.170 2.454
apply_single 119 13.6 0.000 0.000 2.170 2.454
multiply_cannon_metrocomm3 9144 15.5 0.019 0.020 1.564 2.451
calculate_first_density_matrix 1 7.0 0.001 0.001 2.332 2.333
mp_alltoall_i22 627 13.8 1.491 2.324 1.491 2.324
calculate_dm_sparse 119 9.5 0.000 0.000 2.244 2.263
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.435 2.241
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.044 2.092
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.072 2.074
grid_integrate_task_list 119 12.3 2.059 2.074 2.059 2.074
ot_diis_step 108 11.5 0.014 0.014 2.053 2.053
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.018 2.019
mp_sum_l 7207 12.9 1.142 1.906 1.142 1.906
rs_pw_transfer 974 11.9 0.009 0.010 1.853 1.872
jit_kernel_multiply 10 15.6 1.126 1.856 1.126 1.856
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.822 1.822
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.742 1.743
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.699 1.714
potential_pw2rs 119 12.3 0.014 0.014 1.633 1.643
hybrid_alltoall_any 4725 16.4 0.087 0.146 1.321 1.615
make_images_data 4572 15.5 0.042 0.045 1.305 1.597
wfi_extrapolate 11 7.9 0.001 0.001 1.563 1.563
grid_collocate_task_list 119 9.7 1.488 1.499 1.488 1.499
cp_fm_cholesky_invert 11 10.9 1.460 1.463 1.460 1.463
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.441 1.441
cp_fm_diag_elpa_base 50 14.0 1.293 1.349 1.440 1.440
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.270 1.278
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.225 1.247
mp_alltoall_d11v 2130 13.8 1.193 1.220 1.193 1.220
qs_env_update_s_mstruct 11 6.9 0.008 0.011 1.162 1.179
yz_to_x 606 15.1 0.463 0.476 1.102 1.115
mp_alltoall_z22v 1201 16.6 1.062 1.106 1.062 1.106
multiply_cannon_sync_h2d 9144 15.5 1.038 1.044 1.038 1.044
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.985 1.027
qs_create_task_list 11 7.9 0.001 0.001 0.959 0.974
generate_qs_task_list 11 8.9 0.366 0.385 0.958 0.974
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.914 0.927
acc_transpose_blocks 9144 15.5 0.038 0.038 0.917 0.921
x_to_yz 595 16.2 0.483 0.493 0.906 0.910
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=43.759000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=686.545455, yerr=15.686495
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 503.316480E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66218.
MP_Allreduce 9696 492.
MP_Sync 52
MP_Alltoall 1938 1379060.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.025 0.051 117.109 117.111
qs_mol_dyn_low 1 2.0 0.003 0.003 116.650 116.664
qs_forces 11 3.9 0.003 0.004 116.562 116.564
qs_energies 11 4.9 0.002 0.006 113.098 113.110
scf_env_do_scf 11 5.9 0.001 0.001 101.448 101.450
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 90.921 90.923
qs_scf_new_mos 99 7.5 0.000 0.001 65.097 65.203
qs_scf_loop_do_ot 99 8.5 0.000 0.001 65.096 65.203
ot_scf_mini 99 9.5 0.002 0.002 62.221 62.301
dbcsr_multiply_generic 2055 12.4 0.107 0.110 60.917 61.176
velocity_verlet 10 3.0 0.001 0.002 59.385 59.386
multiply_cannon 2055 13.4 0.184 0.190 46.170 47.233
multiply_cannon_loop 2055 14.4 1.507 1.543 43.847 44.969
ot_mini 99 10.5 0.001 0.001 33.400 33.488
qs_ot_get_derivative 99 11.5 0.001 0.002 26.402 26.520
mp_waitall_1 220248 16.4 18.533 20.013 18.533 20.013
rebuild_ks_matrix 110 8.3 0.000 0.000 19.602 19.722
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 19.602 19.721
qs_ot_get_p 110 10.4 0.001 0.001 18.805 18.919
qs_ks_update_qs_env 110 7.6 0.001 0.001 17.327 17.438
multiply_cannon_multrec 49320 15.4 11.676 12.323 16.603 17.154
qs_ot_p2m_diag 48 11.0 0.012 0.019 13.175 13.221
cp_dbcsr_syevd 48 12.0 0.002 0.003 11.829 11.831
qs_rho_update_rho_low 110 7.6 0.001 0.001 11.448 11.569
calculate_rho_elec 110 8.6 0.021 0.026 11.448 11.569
sum_up_and_integrate 110 10.3 0.036 0.042 10.534 10.548
integrate_v_rspace 110 11.3 0.003 0.004 10.498 10.521
init_scf_loop 11 6.9 0.000 0.000 10.445 10.448
cp_fm_diag_elpa 48 13.0 0.000 0.000 10.423 10.444
cp_fm_diag_elpa_base 48 14.0 10.369 10.400 10.420 10.441
multiply_cannon_sync_h2d 49320 15.4 9.312 9.860 9.312 9.860
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 9.305 9.826
init_scf_run 11 5.9 0.000 0.001 8.974 8.974
scf_env_initial_rho_setup 11 6.9 0.002 0.012 8.973 8.974
make_m2s 4110 13.4 0.061 0.064 8.671 8.890
density_rs2pw 110 9.6 0.004 0.005 7.966 8.886
make_images 4110 14.4 0.179 0.194 8.574 8.794
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 8.736 8.779
multiply_cannon_metrocomm3 49320 15.4 0.076 0.080 7.111 8.669
rs_pw_transfer 902 11.9 0.012 0.013 7.484 8.455
pw_transfer 1331 11.6 0.056 0.068 8.304 8.412
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 8.214 8.327
multiply_cannon_metrocomm1 49320 15.4 0.060 0.063 6.094 8.135
fft3d_ps 1111 14.6 0.779 0.879 7.751 7.849
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.339 7.795
apply_single 110 13.6 0.000 0.001 7.339 7.794
prepare_preconditioner 11 7.9 0.000 0.000 7.617 7.643
make_preconditioner 11 8.9 0.000 0.000 7.616 7.643
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.283 7.364
ot_diis_step 99 11.5 0.006 0.006 6.757 6.760
fft_wrap_pw1pw2_140 451 13.1 0.164 0.186 6.530 6.643
mp_sum_l 6514 12.8 5.632 6.621 5.632 6.621
mp_alltoall_z22v 1111 16.6 5.817 6.126 5.817 6.126
wfi_extrapolate 11 7.9 0.001 0.001 5.925 5.925
potential_pw2rs 110 12.3 0.006 0.007 5.707 5.766
dbcsr_mm_accdrv_process 87628 16.1 1.972 2.065 4.807 5.092
make_images_sizes 4110 15.4 0.004 0.005 3.450 4.705
mp_alltoall_i44 4110 16.4 3.446 4.701 3.446 4.701
calculate_dm_sparse 110 9.5 0.000 0.001 4.319 4.426
make_images_data 4110 15.4 0.043 0.046 3.741 4.283
qs_ot_get_orbitals 99 10.5 0.000 0.001 4.145 4.198
cp_fm_cholesky_invert 11 10.9 4.172 4.184 4.172 4.184
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 4.077 4.110
mp_alltoall_d11v 2046 13.8 3.704 4.109 3.704 4.109
hybrid_alltoall_any 4261 16.3 0.080 0.457 3.140 3.673
grid_integrate_task_list 110 12.3 3.207 3.445 3.207 3.445
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.367 3.393
mp_waitany 14300 13.8 2.505 3.372 2.505 3.372
x_to_yz 330 15.4 0.258 0.314 3.290 3.303
yz_to_x 341 15.4 0.213 0.243 2.989 3.172
calculate_first_density_matrix 1 7.0 0.004 0.027 2.872 2.882
mp_allgather_i34 2055 14.4 1.745 2.606 1.745 2.606
jit_kernel_multiply 13 15.9 2.553 2.586 2.553 2.586
parallel_gemm_fm 81 9.0 0.000 0.000 2.477 2.504
parallel_gemm_fm_cosma 81 10.0 2.477 2.504 2.477 2.504
cp_fm_cholesky_decompose 22 10.9 2.467 2.474 2.467 2.474
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.425 2.445
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=117.111000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=478.272727, yerr=1.813631
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 588.255232E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66437.
MP_Allreduce 9695 570.
MP_Sync 52
MP_Alltoall 1717 2463018.
MP_SendRecv 10340 26400.
MP_ISendRecv 10340 26400.
MP_Wait 22352
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.037 0.054 88.986 88.987
qs_mol_dyn_low 1 2.0 0.003 0.004 88.643 88.657
qs_forces 11 3.9 0.003 0.004 88.496 88.499
qs_energies 11 4.9 0.005 0.028 84.597 84.604
scf_env_do_scf 11 5.9 0.001 0.002 73.912 73.916
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 61.773 61.789
velocity_verlet 10 3.0 0.001 0.002 47.064 47.067
dbcsr_multiply_generic 2055 12.4 0.113 0.117 42.806 42.993
qs_scf_new_mos 99 7.5 0.001 0.001 40.791 40.937
qs_scf_loop_do_ot 99 8.5 0.001 0.001 40.791 40.936
ot_scf_mini 99 9.5 0.003 0.003 38.910 39.045
multiply_cannon 2055 13.4 0.224 0.245 32.780 34.431
multiply_cannon_loop 2055 14.4 0.927 0.952 30.906 31.860
ot_mini 99 10.5 0.001 0.001 21.508 21.647
rebuild_ks_matrix 110 8.3 0.000 0.000 16.843 16.943
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.016 16.842 16.942
qs_ot_get_derivative 99 11.5 0.001 0.001 15.570 15.719
multiply_cannon_multrec 24660 15.4 7.599 9.371 13.844 15.580
qs_ks_update_qs_env 110 7.6 0.001 0.001 14.922 15.010
mp_waitall_1 176588 16.5 11.565 14.021 11.565 14.021
init_scf_loop 11 6.9 0.001 0.006 12.045 12.070
qs_ot_get_p 110 10.4 0.001 0.001 10.069 10.198
prepare_preconditioner 11 7.9 0.001 0.004 9.660 9.684
make_preconditioner 11 8.9 0.000 0.001 9.659 9.684
make_full_inverse_cholesky 11 9.9 0.000 0.000 9.053 9.329
sum_up_and_integrate 110 10.3 0.053 0.058 9.058 9.073
integrate_v_rspace 110 11.3 0.003 0.003 9.005 9.020
multiply_cannon_metrocomm3 24660 15.4 0.067 0.069 5.752 8.992
qs_rho_update_rho_low 110 7.6 0.001 0.001 8.877 8.891
calculate_rho_elec 110 8.6 0.040 0.048 8.876 8.890
multiply_cannon_sync_h2d 24660 15.4 7.062 8.375 7.062 8.375
make_m2s 4110 13.4 0.057 0.059 7.091 7.669
make_images 4110 14.4 0.397 0.435 6.981 7.557
init_scf_run 11 5.9 0.000 0.001 7.556 7.557
scf_env_initial_rho_setup 11 6.9 0.000 0.002 7.556 7.556
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.641 7.366
apply_single 110 13.6 0.000 0.001 6.641 7.366
qs_ot_p2m_diag 48 11.0 0.029 0.044 7.323 7.350
rs_pw_transfer 902 11.9 0.012 0.013 6.092 6.671
cp_dbcsr_syevd 48 12.0 0.003 0.003 6.605 6.605
dbcsr_mm_accdrv_process 52282 16.1 4.737 5.625 6.086 6.407
density_rs2pw 110 9.6 0.004 0.004 5.776 6.392
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 5.470 6.176
ot_diis_step 99 11.5 0.010 0.010 5.870 5.870
pw_transfer 1331 11.6 0.067 0.074 5.396 5.622
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.552 5.564
cp_fm_diag_elpa_base 48 14.0 5.458 5.496 5.549 5.561
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 5.288 5.518
fft3d_ps 1111 14.6 1.111 1.330 4.645 4.870
make_images_data 4110 15.4 0.046 0.049 3.886 4.732
wfi_extrapolate 11 7.9 0.001 0.001 4.727 4.727
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 4.660 4.722
potential_pw2rs 110 12.3 0.008 0.008 4.618 4.682
cp_fm_cholesky_invert 11 10.9 4.622 4.634 4.622 4.634
hybrid_alltoall_any 4261 16.3 0.103 0.443 3.235 4.187
fft_wrap_pw1pw2_140 451 13.1 0.196 0.215 3.801 4.037
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.573 3.579
grid_integrate_task_list 110 12.3 3.100 3.349 3.100 3.349
calculate_dm_sparse 110 9.5 0.001 0.001 3.214 3.254
mp_alltoall_d11v 2046 13.8 2.828 3.243 2.828 3.243
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.125 3.156
cp_fm_cholesky_decompose 22 10.9 2.995 3.002 2.995 3.002
mp_alltoall_z22v 1111 16.6 2.736 2.923 2.736 2.923
mp_sum_l 6514 12.8 2.317 2.893 2.317 2.893
make_images_sizes 4110 15.4 0.005 0.005 1.844 2.808
mp_alltoall_i44 4110 16.4 1.840 2.803 1.840 2.803
calculate_first_density_matrix 1 7.0 0.002 0.010 2.707 2.710
mp_allgather_i34 2055 14.4 1.200 2.546 1.200 2.546
jit_kernel_multiply 13 16.3 0.997 2.497 0.997 2.497
mp_waitany 10164 13.8 1.789 2.493 1.789 2.493
grid_collocate_task_list 110 9.6 2.054 2.486 2.054 2.486
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.398 2.437
dbcsr_complete_redistribute 325 12.2 0.234 0.284 1.939 2.292
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.092 2.101
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 1.915 1.918
yz_to_x 451 15.2 0.269 0.308 1.767 1.914
rs_pw_transfer_RS2PW_140 121 11.5 0.207 0.217 1.147 1.827
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=88.987000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=556.000000, yerr=6.849021
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 664.715264E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66428.
MP_Allreduce 9694 567.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.056 0.094 63.698 63.700
qs_mol_dyn_low 1 2.0 0.003 0.005 63.245 63.256
qs_forces 11 3.9 0.003 0.005 63.166 63.168
qs_energies 11 4.9 0.008 0.021 59.959 59.963
scf_env_do_scf 11 5.9 0.001 0.002 51.629 51.631
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 42.367 42.367
velocity_verlet 10 3.0 0.011 0.018 33.981 33.983
dbcsr_multiply_generic 2055 12.4 0.105 0.109 29.688 29.983
qs_scf_new_mos 99 7.5 0.001 0.001 26.784 26.889
qs_scf_loop_do_ot 99 8.5 0.001 0.004 26.784 26.888
ot_scf_mini 99 9.5 0.003 0.003 25.547 25.665
multiply_cannon 2055 13.4 0.214 0.221 23.060 24.354
multiply_cannon_loop 2055 14.4 0.618 0.631 21.731 22.714
ot_mini 99 10.5 0.001 0.001 14.578 14.698
rebuild_ks_matrix 110 8.3 0.000 0.001 12.845 12.987
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.016 12.845 12.986
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.384 11.511
multiply_cannon_multrec 16440 15.4 4.002 5.098 10.111 11.260
mp_waitall_1 139946 16.5 7.520 10.744 7.520 10.744
qs_ot_get_derivative 99 11.5 0.001 0.001 10.109 10.228
init_scf_loop 11 6.9 0.002 0.006 9.228 9.231
multiply_cannon_metrocomm3 16440 15.4 0.042 0.044 4.528 7.607
prepare_preconditioner 11 7.9 0.000 0.001 7.436 7.453
make_preconditioner 11 8.9 0.000 0.001 7.436 7.453
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.783 7.123
sum_up_and_integrate 110 10.3 0.061 0.062 6.958 6.971
integrate_v_rspace 110 11.3 0.003 0.003 6.897 6.910
qs_ot_get_p 110 10.4 0.001 0.001 6.131 6.283
dbcsr_mm_accdrv_process 34862 16.1 4.746 5.644 5.963 6.097
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.085 6.095
calculate_rho_elec 110 8.6 0.059 0.059 6.084 6.094
init_scf_run 11 5.9 0.000 0.001 5.816 5.816
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.815 5.816
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.002 5.456
apply_single 110 13.6 0.000 0.000 5.002 5.456
make_m2s 4110 13.4 0.050 0.051 4.304 4.751
make_images 4110 14.4 0.391 0.506 4.190 4.639
ot_diis_step 99 11.5 0.011 0.011 4.433 4.434
density_rs2pw 110 9.6 0.005 0.006 3.196 4.416
rs_pw_transfer 902 11.9 0.011 0.012 3.118 4.341
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.647 4.306
multiply_cannon_sync_h2d 16440 15.4 3.694 4.234 3.694 4.234
qs_ot_p2m_diag 48 11.0 0.042 0.044 4.224 4.229
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.881 3.882
pw_transfer 1331 11.6 0.066 0.074 3.729 3.741
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.621 3.636
grid_integrate_task_list 110 12.3 3.161 3.401 3.161 3.401
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.333 3.342
cp_fm_diag_elpa_base 48 14.0 3.268 3.299 3.331 3.340
fft_wrap_pw1pw2_140 451 13.1 0.214 0.219 3.061 3.080
make_images_data 4110 15.4 0.043 0.047 2.447 3.010
wfi_extrapolate 11 7.9 0.001 0.001 2.959 2.959
fft3d_ps 1111 14.6 1.096 1.131 2.919 2.929
cp_fm_cholesky_invert 11 10.9 2.921 2.926 2.921 2.926
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.922 2.923
hybrid_alltoall_any 4261 16.3 0.106 0.373 2.169 2.891
calculate_first_density_matrix 1 7.0 0.003 0.012 2.763 2.764
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.648 2.717
calculate_dm_sparse 110 9.5 0.001 0.001 2.676 2.703
potential_pw2rs 110 12.3 0.011 0.011 2.682 2.691
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.578 2.629
mp_waitany 17072 13.8 1.297 2.573 1.297 2.573
grid_collocate_task_list 110 9.6 2.082 2.444 2.082 2.444
multiply_cannon_metrocomm4 14385 15.4 0.046 0.049 0.851 2.395
mp_sum_l 6514 12.8 1.738 2.380 1.738 2.380
mp_alltoall_d11v 2046 13.8 1.810 2.306 1.810 2.306
mp_irecv_dv 48980 15.7 0.781 2.267 0.781 2.267
rs_pw_transfer_RS2PW_140 121 11.5 0.178 0.189 0.923 2.138
cp_fm_cholesky_decompose 22 10.9 2.044 2.065 2.044 2.065
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 2.022 2.024
dbcsr_complete_redistribute 325 12.2 0.340 0.384 1.406 1.855
cp_fm_upper_to_full 70 14.2 1.381 1.744 1.381 1.744
jit_kernel_multiply 8 16.3 0.830 1.711 0.830 1.711
mp_allgather_i34 2055 14.4 0.650 1.678 0.650 1.678
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.665 1.678
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.620 1.630
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.363 1.483
copy_fm_to_dbcsr 174 11.2 0.001 0.002 0.969 1.411
rs_gather_matrices 110 12.3 0.233 0.262 0.980 1.399
acc_transpose_blocks 16440 15.4 0.072 0.075 1.360 1.370
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.299 1.308
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=63.700000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=628.090909, yerr=8.425659
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 724.553728E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66426.
MP_Allreduce 9694 608.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.044 0.081 68.425 68.439
qs_mol_dyn_low 1 2.0 0.003 0.005 67.951 67.964
qs_forces 11 3.9 0.003 0.007 67.868 67.869
qs_energies 11 4.9 0.008 0.028 64.443 64.447
scf_env_do_scf 11 5.9 0.001 0.001 55.565 55.568
scf_env_do_scf_inner_loop 99 6.5 0.003 0.008 43.518 43.521
velocity_verlet 10 3.0 0.002 0.002 38.461 38.463
dbcsr_multiply_generic 2055 12.4 0.111 0.118 31.314 31.530
qs_scf_new_mos 99 7.5 0.001 0.001 28.113 28.224
qs_scf_loop_do_ot 99 8.5 0.001 0.001 28.113 28.223
ot_scf_mini 99 9.5 0.003 0.004 26.467 26.573
multiply_cannon 2055 13.4 0.249 0.267 23.706 24.778
multiply_cannon_loop 2055 14.4 0.883 0.910 21.956 22.610
ot_mini 99 10.5 0.001 0.001 15.386 15.494
multiply_cannon_multrec 24660 15.4 4.245 6.905 12.972 14.156
rebuild_ks_matrix 110 8.3 0.000 0.000 12.462 12.568
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.016 12.462 12.567
init_scf_loop 11 6.9 0.001 0.004 12.006 12.013
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.034 11.128
qs_ot_get_derivative 99 11.5 0.001 0.001 11.025 11.125
prepare_preconditioner 11 7.9 0.000 0.001 10.223 10.240
make_preconditioner 11 8.9 0.000 0.001 10.222 10.240
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.434 9.910
dbcsr_mm_accdrv_process 52304 16.0 7.028 8.525 8.579 9.537
mp_waitall_1 121746 16.5 5.079 7.315 5.079 7.315
sum_up_and_integrate 110 10.3 0.067 0.071 6.661 6.674
integrate_v_rspace 110 11.3 0.003 0.003 6.593 6.609
make_m2s 4110 13.4 0.060 0.062 5.725 6.272
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.230 6.237
calculate_rho_elec 110 8.6 0.078 0.081 6.229 6.237
make_images 4110 14.4 0.577 0.693 5.584 6.128
init_scf_run 11 5.9 0.000 0.001 6.087 6.088
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.087 6.087
qs_ot_get_p 110 10.4 0.001 0.001 5.869 6.038
cp_fm_upper_to_full 70 14.2 3.438 4.963 3.438 4.963
ot_diis_step 99 11.5 0.011 0.011 4.308 4.308
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.147 4.232
apply_single 110 13.6 0.000 0.000 4.146 4.232
dbcsr_complete_redistribute 325 12.2 0.417 0.457 2.799 3.975
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.945 3.960
pw_transfer 1331 11.6 0.065 0.075 3.908 3.937
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.802 3.835
density_rs2pw 110 9.6 0.004 0.004 3.180 3.691
multiply_cannon_metrocomm3 24660 15.4 0.035 0.037 1.906 3.673
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.599 3.649
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.488 3.545
make_images_data 4110 15.4 0.046 0.050 2.894 3.476
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.469 3.469
grid_integrate_task_list 110 12.3 3.272 3.426 3.272 3.426
hybrid_alltoall_any 4261 16.3 0.120 0.457 2.519 3.412
copy_fm_to_dbcsr 174 11.2 0.001 0.002 2.159 3.316
multiply_cannon_sync_h2d 24660 15.4 3.162 3.284 3.162 3.284
wfi_extrapolate 11 7.9 0.001 0.001 3.244 3.244
fft_wrap_pw1pw2_140 451 13.1 0.203 0.215 3.199 3.234
calculate_dm_sparse 110 9.5 0.001 0.001 3.098 3.134
fft3d_ps 1111 14.6 1.095 1.134 3.093 3.121
rs_pw_transfer 902 11.9 0.010 0.011 2.547 3.069
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.016 3.017
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.921 2.930
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.779 2.929
cp_fm_diag_elpa_base 48 14.0 2.775 2.837 2.919 2.929
mp_alltoall_i22 605 13.7 1.715 2.922 1.715 2.922
cp_fm_cholesky_invert 11 10.9 2.770 2.779 2.770 2.779
calculate_first_density_matrix 1 7.0 0.013 0.026 2.752 2.755
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.582 2.618
jit_kernel_multiply 12 15.9 1.220 2.511 1.220 2.511
grid_collocate_task_list 110 9.6 2.180 2.412 2.180 2.412
potential_pw2rs 110 12.3 0.012 0.013 2.311 2.319
mp_alltoall_d11v 2046 13.8 1.879 2.302 1.879 2.302
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 2.278 2.279
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.822 1.854
mp_sum_l 6514 12.8 1.218 1.852 1.218 1.852
mp_allgather_i34 2055 14.4 0.846 1.845 0.846 1.845
cp_fm_cholesky_decompose 22 10.9 1.664 1.720 1.664 1.720
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.598 1.704
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.686 1.702
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.677 1.687
mp_waitany 13376 13.8 1.137 1.643 1.137 1.643
acc_transpose_blocks 24660 15.4 0.107 0.109 1.564 1.594
multiply_cannon_metrocomm4 20550 15.4 0.062 0.066 0.843 1.525
mp_irecv_dv 62702 16.1 0.740 1.444 0.740 1.444
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=68.439000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=688.090909, yerr=6.529020
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 834.166784E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66424.
MP_Allreduce 9694 649.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.034 83.597 83.599
qs_mol_dyn_low 1 2.0 0.003 0.004 83.303 83.312
qs_forces 11 3.9 0.003 0.003 83.234 83.239
qs_energies 11 4.9 0.001 0.001 79.075 79.081
scf_env_do_scf 11 5.9 0.001 0.002 68.692 68.692
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 54.157 54.158
velocity_verlet 10 3.0 0.002 0.002 46.404 46.407
qs_scf_new_mos 99 7.5 0.001 0.001 33.149 33.275
qs_scf_loop_do_ot 99 8.5 0.001 0.001 33.148 33.274
ot_scf_mini 99 9.5 0.002 0.002 31.405 31.495
dbcsr_multiply_generic 2055 12.4 0.105 0.111 30.648 30.981
multiply_cannon 2055 13.4 0.256 0.267 21.245 22.972
multiply_cannon_loop 2055 14.4 0.323 0.334 18.987 19.727
ot_mini 99 10.5 0.001 0.001 15.478 15.608
rebuild_ks_matrix 110 8.3 0.000 0.000 15.251 15.406
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 15.250 15.406
mp_waitall_1 103326 16.6 11.862 14.723 11.862 14.723
init_scf_loop 11 6.9 0.000 0.000 14.464 14.467
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.599 13.743
prepare_preconditioner 11 7.9 0.000 0.000 12.418 12.434
make_preconditioner 11 8.9 0.000 0.000 12.418 12.434
make_full_inverse_cholesky 11 9.9 0.000 0.000 11.803 11.958
qs_ot_get_derivative 99 11.5 0.001 0.001 10.796 10.894
qs_ot_get_p 110 10.4 0.001 0.001 10.700 10.856
qs_rho_update_rho_low 110 7.6 0.001 0.001 9.913 9.957
calculate_rho_elec 110 8.6 0.116 0.116 9.912 9.956
sum_up_and_integrate 110 10.3 0.080 0.082 8.947 8.960
integrate_v_rspace 110 11.3 0.003 0.003 8.866 8.880
multiply_cannon_multrec 8220 15.4 3.201 4.391 7.449 8.296
qs_ot_p2m_diag 48 11.0 0.081 0.085 7.938 7.947
make_m2s 4110 13.4 0.039 0.040 7.102 7.830
make_images 4110 14.4 0.639 0.696 6.969 7.699
cp_fm_cholesky_invert 11 10.9 7.371 7.380 7.371 7.380
cp_dbcsr_syevd 48 12.0 0.003 0.003 7.375 7.377
init_scf_run 11 5.9 0.000 0.001 6.790 6.790
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.789 6.790
density_rs2pw 110 9.6 0.004 0.004 6.124 6.520
cp_fm_diag_elpa 48 13.0 0.000 0.000 6.360 6.368
cp_fm_diag_elpa_base 48 14.0 6.266 6.328 6.357 6.366
pw_transfer 1331 11.6 0.067 0.073 6.272 6.299
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 6.164 6.192
multiply_cannon_metrocomm3 8220 15.4 0.017 0.018 3.576 5.881
rs_pw_transfer 902 11.9 0.010 0.010 5.347 5.794
make_images_data 4110 15.4 0.038 0.043 4.302 5.499
fft3d_ps 1111 14.6 1.164 1.202 5.410 5.448
dbcsr_mm_accdrv_process 17442 15.9 2.820 3.941 4.115 5.046
fft_wrap_pw1pw2_140 451 13.1 0.212 0.216 4.966 4.997
hybrid_alltoall_any 4261 16.3 0.200 0.859 3.914 4.921
ot_diis_step 99 11.5 0.012 0.013 4.563 4.564
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.366 4.498
apply_single 110 13.6 0.000 0.000 4.365 4.498
wfi_extrapolate 11 7.9 0.001 0.001 4.328 4.329
multiply_cannon_metrocomm1 8220 15.4 0.021 0.022 3.328 4.074
potential_pw2rs 110 12.3 0.015 0.015 3.876 3.889
mp_alltoall_d11v 2046 13.8 3.372 3.744 3.372 3.744
grid_integrate_task_list 110 12.3 3.370 3.650 3.370 3.650
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.483 3.537
mp_allgather_i34 2055 14.4 1.427 3.438 1.427 3.438
cp_fm_cholesky_decompose 22 10.9 3.255 3.266 3.255 3.266
calculate_dm_sparse 110 9.5 0.001 0.001 3.152 3.207
mp_alltoall_z22v 1111 16.6 3.137 3.167 3.137 3.167
multiply_cannon_sync_h2d 8220 15.4 2.854 2.963 2.854 2.963
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.922 2.958
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.901 2.905
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.868 2.871
make_images_sizes 4110 15.4 0.005 0.005 1.380 2.700
mp_alltoall_i44 4110 16.4 1.375 2.695 1.375 2.695
grid_collocate_task_list 110 9.6 2.319 2.619 2.319 2.619
mp_waitany 9240 13.8 2.111 2.558 2.111 2.558
dbcsr_complete_redistribute 325 12.2 0.536 0.554 2.270 2.491
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.280 2.342
calculate_first_density_matrix 1 7.0 0.001 0.001 2.307 2.308
x_to_yz 550 16.1 0.574 0.583 2.169 2.183
mp_sum_l 6514 12.8 1.611 2.125 1.611 2.125
yz_to_x 561 15.1 0.523 0.532 2.064 2.085
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.992 2.005
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.773 1.984
rs_gather_matrices 110 12.3 0.322 0.365 1.515 1.871
qs_env_update_s_mstruct 11 6.9 0.001 0.002 1.741 1.854
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.728 1.773
jit_kernel_multiply 10 15.8 0.988 1.736 0.988 1.736
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.682 1.712
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.640 1.708
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=83.599000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=790.181818, yerr=11.271994
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 1.361539E+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 0.074 0.086 106.594 106.596
qs_mol_dyn_low 1 2.0 0.003 0.004 106.041 106.055
qs_forces 11 3.9 0.003 0.003 105.972 105.974
qs_energies 11 4.9 0.001 0.002 101.302 101.304
scf_env_do_scf 11 5.9 0.003 0.004 89.926 89.926
velocity_verlet 10 3.0 0.002 0.002 67.326 67.332
scf_env_do_scf_inner_loop 99 6.5 0.003 0.008 56.284 56.285
init_scf_loop 11 6.9 0.000 0.000 33.539 33.540
dbcsr_multiply_generic 2055 12.4 0.118 0.123 32.810 32.910
qs_scf_new_mos 99 7.5 0.001 0.001 32.575 32.616
qs_scf_loop_do_ot 99 8.5 0.001 0.001 32.574 32.616
prepare_preconditioner 11 7.9 0.000 0.000 31.121 31.130
make_preconditioner 11 8.9 0.000 0.000 31.121 31.130
ot_scf_mini 99 9.5 0.002 0.002 30.520 30.570
make_full_inverse_cholesky 11 9.9 0.000 0.000 24.947 30.499
multiply_cannon 2055 13.4 0.349 0.373 22.624 23.706
multiply_cannon_loop 2055 14.4 0.342 0.346 20.656 20.917
cp_fm_upper_to_full 70 14.2 13.135 19.007 13.135 19.007
rebuild_ks_matrix 110 8.3 0.001 0.001 17.079 17.116
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 17.078 17.115
ot_mini 99 10.5 0.001 0.001 16.087 16.125
qs_ks_update_qs_env 110 7.6 0.001 0.001 15.328 15.362
mp_waitall_1 84994 16.7 11.019 12.232 11.019 12.232
dbcsr_complete_redistribute 325 12.2 1.019 1.054 8.205 11.704
qs_ot_get_derivative 99 11.5 0.002 0.002 11.438 11.482
qs_rho_update_rho_low 110 7.6 0.001 0.001 11.103 11.138
calculate_rho_elec 110 8.6 0.228 0.228 11.103 11.137
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.927 10.417
multiply_cannon_multrec 8220 15.4 4.428 4.597 9.867 9.992
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.158 9.615
sum_up_and_integrate 110 10.3 0.150 0.154 9.543 9.563
integrate_v_rspace 110 11.3 0.003 0.004 9.393 9.412
mp_alltoall_i22 605 13.7 5.737 9.247 5.737 9.247
make_m2s 4110 13.4 0.044 0.045 8.204 8.721
qs_ot_get_p 110 10.4 0.001 0.001 8.589 8.634
make_images 4110 14.4 0.882 0.926 8.013 8.531
cp_fm_cholesky_invert 11 10.9 8.024 8.029 8.024 8.029
pw_transfer 1331 11.6 0.076 0.076 7.713 7.723
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 7.596 7.605
init_scf_run 11 5.9 0.000 0.001 6.837 6.837
scf_env_initial_rho_setup 11 6.9 0.000 0.000 6.836 6.837
fft3d_ps 1111 14.6 1.304 1.325 6.790 6.797
density_rs2pw 110 9.6 0.004 0.004 6.509 6.548
fft_wrap_pw1pw2_140 451 13.1 0.221 0.223 6.386 6.395
qs_ot_p2m_diag 48 11.0 0.151 0.156 6.357 6.365
make_images_data 4110 15.4 0.041 0.044 4.907 6.101
hybrid_alltoall_any 4261 16.3 0.258 0.562 4.591 5.816
cp_dbcsr_syevd 48 12.0 0.003 0.003 5.778 5.778
dbcsr_mm_accdrv_process 11614 15.7 3.341 3.677 5.294 5.551
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.747 5.222
apply_single 110 13.6 0.000 0.000 4.747 5.221
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 4.786 5.061
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.960 4.960
cp_fm_diag_elpa_base 48 14.0 4.343 4.611 4.958 4.958
rs_pw_transfer 902 11.9 0.011 0.011 4.674 4.711
ot_diis_step 99 11.5 0.015 0.016 4.624 4.624
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.677 4.136
potential_pw2rs 110 12.3 0.021 0.021 4.124 4.133
wfi_extrapolate 11 7.9 0.001 0.001 4.060 4.060
multiply_cannon_sync_h2d 8220 15.4 3.947 3.956 3.947 3.956
mp_alltoall_d11v 2046 13.8 3.668 3.790 3.668 3.790
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.775 3.777
grid_integrate_task_list 110 12.3 3.680 3.745 3.680 3.745
mp_alltoall_z22v 1111 16.6 3.604 3.620 3.604 3.620
calculate_dm_sparse 110 9.5 0.001 0.001 3.548 3.590
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.277 3.300
cp_fm_cholesky_decompose 22 10.9 3.263 3.288 3.263 3.288
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.217 3.222
yz_to_x 561 15.1 0.935 0.939 2.843 2.857
make_images_sizes 4110 15.4 0.004 0.005 1.307 2.739
mp_alltoall_i44 4110 16.4 1.302 2.735 1.302 2.735
grid_collocate_task_list 110 9.6 2.643 2.664 2.643 2.664
x_to_yz 550 16.1 0.933 0.939 2.630 2.635
calculate_first_density_matrix 1 7.0 0.001 0.002 2.600 2.603
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.571 2.582
qs_env_update_s_mstruct 11 6.9 0.010 0.019 2.310 2.371
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.357 2.360
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.107 2.217
copy_dbcsr_to_fm 151 11.3 0.003 0.003 2.099 2.138
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=106.596000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1207.727273, yerr=63.111864
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430457200640 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 1986252263424 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992001093632 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 11613077360640 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239162695680 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239162695680 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 94.232994E+12 0.0% 0.0% 100.0%
flops max/rank 1.103378E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806382528 0.0% 0.0% 100.0%
number of processed stacks 12049536 0.0% 0.0% 100.0%
average stack size 0.0 0.0 564.9
marketing flops 145.651870E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 630.706176E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10439712
MPI messages size (bytes):
total size 4.526311E+12
min size 0.000000E+00
max size 4.537280E+06
average size 433.566688E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3611304 96569262080
32768 < size <= 131072 1307392 74803314688
131072 < size <= 4194304 5190114 3200149557592
4194304 < size <= 16777216 263934 1154745839544
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 57188.
MP_Allreduce 11106 798.
MP_Sync 88
MP_Alltoall 2242 2521910.
MP_SendRecv 24510 18752.
MP_ISendRecv 24510 18752.
MP_Wait 42802
MP_ISend 16140 108019.
MP_IRecv 16140 108019.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.088 248.896 248.897
qs_mol_dyn_low 1 2.0 0.007 0.032 248.120 248.134
qs_forces 11 3.9 0.019 0.063 247.970 247.972
qs_energies 11 4.9 0.004 0.020 241.811 241.833
scf_env_do_scf 11 5.9 0.001 0.001 222.913 222.917
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 193.829 193.832
velocity_verlet 10 3.0 0.004 0.016 149.919 149.923
qs_scf_new_mos 118 7.6 0.001 0.001 144.761 145.004
qs_scf_loop_do_ot 118 8.6 0.001 0.001 144.760 145.003
ot_scf_mini 118 9.6 0.003 0.003 137.671 137.861
dbcsr_multiply_generic 2529 12.6 0.178 0.182 133.329 134.166
multiply_cannon 2529 13.6 0.246 0.256 105.209 107.226
multiply_cannon_loop 2529 14.6 2.052 2.131 101.865 104.129
ot_mini 118 10.6 0.001 0.001 71.969 72.175
qs_ot_get_derivative 118 11.6 0.001 0.004 46.436 46.643
multiply_cannon_multrec 60696 15.6 33.389 36.992 41.771 45.216
qs_ot_get_p 129 10.4 0.001 0.001 42.711 42.947
mp_waitall_1 269446 16.5 35.426 41.021 35.426 41.021
rebuild_ks_matrix 129 8.3 0.001 0.001 38.762 39.046
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.017 38.761 39.045
qs_ks_update_qs_env 129 7.6 0.001 0.001 34.994 35.227
qs_ot_p2m_diag 84 11.4 0.080 0.092 33.886 33.962
cp_dbcsr_syevd 84 12.4 0.005 0.005 31.122 31.127
multiply_cannon_sync_h2d 60696 15.6 27.403 30.571 27.403 30.571
init_scf_loop 11 6.9 0.000 0.000 28.998 29.001
cp_fm_diag_elpa 84 13.4 0.000 0.000 26.350 26.388
cp_fm_diag_elpa_base 84 14.4 26.224 26.299 26.344 26.381
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 24.929 25.526
apply_single 129 13.6 0.001 0.001 24.928 25.526
ot_diis_step 118 11.6 0.008 0.008 25.247 25.251
prepare_preconditioner 11 7.9 0.000 0.000 23.551 23.597
make_preconditioner 11 8.9 0.000 0.000 23.550 23.597
make_full_inverse_cholesky 11 9.9 0.000 0.000 22.582 22.904
qs_ot_get_derivative_diag 78 12.4 0.002 0.002 22.041 22.183
qs_rho_update_rho_low 129 7.7 0.001 0.001 18.990 19.080
calculate_rho_elec 129 8.7 0.046 0.065 18.989 19.079
make_m2s 5058 13.6 0.103 0.109 18.199 18.772
make_images 5058 14.6 0.407 0.427 18.019 18.599
multiply_cannon_metrocomm3 60696 15.6 0.115 0.121 15.950 18.582
sum_up_and_integrate 129 10.3 0.090 0.107 18.414 18.444
integrate_v_rspace 129 11.3 0.004 0.005 18.324 18.358
rs_pw_transfer 1054 12.0 0.017 0.020 12.037 15.696
density_rs2pw 129 9.7 0.006 0.007 12.029 15.206
init_scf_run 11 5.9 0.000 0.001 14.317 14.317
scf_env_initial_rho_setup 11 6.9 0.000 0.001 14.316 14.317
cp_fm_cholesky_invert 11 10.9 13.303 13.316 13.303 13.316
multiply_cannon_metrocomm1 60696 15.6 0.088 0.094 8.849 12.771
pw_transfer 1559 11.6 0.078 0.102 10.256 10.789
wfi_extrapolate 11 7.9 0.001 0.001 10.601 10.602
fft_wrap_pw1pw2 1301 12.7 0.011 0.013 10.047 10.589
make_images_data 5058 15.6 0.066 0.072 8.453 10.194
mp_sum_l 7936 13.1 8.667 10.149 8.667 10.149
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 9.657 9.808
hybrid_alltoall_any 5245 16.5 0.291 2.258 6.912 9.377
fft3d_ps 1301 14.7 2.113 2.860 8.839 9.316
calculate_dm_sparse 129 9.5 0.001 0.001 9.025 9.144
dbcsr_mm_accdrv_process 125516 16.2 3.061 3.327 7.942 8.655
potential_pw2rs 129 12.3 0.009 0.011 8.117 8.461
qs_ot_get_orbitals 118 10.6 0.001 0.001 8.182 8.290
fft_wrap_pw1pw2_140 527 13.2 0.431 0.510 7.593 8.164
cp_fm_cholesky_decompose 22 10.9 7.916 7.931 7.916 7.931
grid_integrate_task_list 129 12.3 6.957 7.741 6.957 7.741
mp_alltoall_d11v 2429 14.1 5.817 7.253 5.817 7.253
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.004 7.224 7.233
make_images_sizes 5058 15.6 0.007 0.007 4.346 6.745
mp_alltoall_i44 5058 16.6 4.340 6.738 4.340 6.738
mp_waitany 16140 13.9 3.367 6.678 3.367 6.678
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 6.096 6.203
grid_collocate_task_list 129 9.7 4.587 5.861 4.587 5.861
rs_pw_transfer_RS2PW_140 140 11.5 0.281 0.294 2.356 5.624
mp_alltoall_z22v 1301 16.7 4.541 4.997 4.541 4.997
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=248.897000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=597.363636, yerr=8.336914
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022950912 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963542011904 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444706349056 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019182452736 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019182452736 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796564E+12 0.0% 0.0% 100.0%
flops max/rank 2.166473E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499488 0.0% 0.0% 100.0%
number of processed stacks 5925696 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.6
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 829.054976E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2385600
MPI messages size (bytes):
total size 4.069299E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.705776E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70188 2295595008
32768 < size <= 131072 716032 54973693952
131072 < size <= 4194304 1363760 1386318135296
4194304 < size <= 16777216 153648 1453842592400
16777216 < size 67056 1171888537600
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3995 58319.
MP_Allreduce 10986 968.
MP_Sync 86
MP_Alltoall 1955 4591051.
MP_SendRecv 11938 47072.
MP_ISendRecv 11938 47072.
MP_Wait 25718
MP_ISend 11660 212488.
MP_IRecv 11660 212488.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.020 0.034 256.596 256.597
qs_mol_dyn_low 1 2.0 0.003 0.004 256.042 256.219
qs_forces 11 3.9 0.005 0.005 255.941 255.943
qs_energies 11 4.9 0.001 0.002 248.392 248.447
scf_env_do_scf 11 5.9 0.001 0.001 229.378 229.388
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 178.637 178.640
velocity_verlet 10 3.0 0.001 0.002 160.135 160.138
qs_scf_new_mos 116 7.6 0.001 0.001 129.280 129.779
qs_scf_loop_do_ot 116 8.6 0.001 0.001 129.279 129.778
ot_scf_mini 116 9.6 0.004 0.004 123.568 124.095
dbcsr_multiply_generic 2485 12.5 0.186 0.193 112.736 117.233
multiply_cannon 2485 13.5 0.477 0.530 83.196 97.179
multiply_cannon_loop 2485 14.5 1.240 1.287 78.436 88.918
ot_mini 116 10.6 0.001 0.001 62.502 63.103
mp_waitall_1 212858 16.6 35.655 58.008 35.655 58.008
init_scf_loop 11 6.9 0.000 0.000 50.571 50.574
multiply_cannon_metrocomm3 29820 15.5 0.090 0.101 20.992 46.980
prepare_preconditioner 11 7.9 0.000 0.000 45.419 45.474
make_preconditioner 11 8.9 0.000 0.000 45.419 45.474
make_full_inverse_cholesky 11 9.9 0.000 0.000 43.561 44.271
qs_ot_get_p 127 10.4 0.001 0.001 40.830 41.494
rebuild_ks_matrix 127 8.3 0.001 0.001 38.453 39.435
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.018 38.452 39.435
qs_ot_get_derivative 116 11.6 0.001 0.002 38.026 38.577
qs_ks_update_qs_env 127 7.6 0.001 0.001 34.501 35.424
multiply_cannon_multrec 29820 15.5 20.769 25.428 30.416 35.423
qs_ot_p2m_diag 82 11.4 0.186 0.214 33.406 33.564
cp_dbcsr_syevd 82 12.4 0.005 0.006 31.590 31.596
cp_fm_diag_elpa 82 13.4 0.000 0.000 26.931 26.958
cp_fm_diag_elpa_base 82 14.4 26.543 26.705 26.919 26.942
cp_fm_cholesky_invert 11 10.9 26.438 26.464 26.438 26.464
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 22.758 25.654
apply_single 127 13.6 0.001 0.001 22.758 25.654
ot_diis_step 116 11.6 0.014 0.015 23.939 23.941
multiply_cannon_sync_h2d 29820 15.5 18.242 21.184 18.242 21.184
make_m2s 4970 13.5 0.088 0.091 18.595 20.354
make_images 4970 14.5 1.146 1.336 18.391 20.151
qs_rho_update_rho_low 127 7.7 0.001 0.001 19.169 19.203
calculate_rho_elec 127 8.7 0.088 0.105 19.168 19.202
sum_up_and_integrate 127 10.3 0.115 0.132 18.753 18.793
integrate_v_rspace 127 11.3 0.004 0.004 18.637 18.685
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 16.076 16.455
density_rs2pw 127 9.7 0.006 0.006 12.326 14.893
rs_pw_transfer 1038 11.9 0.014 0.015 11.644 14.149
init_scf_run 11 5.9 0.000 0.001 13.742 13.743
scf_env_initial_rho_setup 11 6.9 0.000 0.001 13.742 13.743
make_images_data 4970 15.5 0.062 0.070 11.089 13.396
cp_fm_cholesky_decompose 22 10.9 12.909 13.003 12.909 13.003
mp_sum_l 7804 13.0 9.201 12.538 9.201 12.538
hybrid_alltoall_any 5155 16.4 0.341 1.497 9.308 12.169
pw_transfer 1535 11.6 0.087 0.101 11.827 11.927
fft_wrap_pw1pw2 1281 12.7 0.011 0.011 11.601 11.703
multiply_cannon_metrocomm4 27335 15.5 0.098 0.113 3.868 11.128
mp_irecv_dv 68888 16.3 3.679 10.743 3.679 10.743
wfi_extrapolate 11 7.9 0.001 0.001 10.127 10.128
fft3d_ps 1281 14.7 2.735 2.918 9.988 10.047
dbcsr_mm_accdrv_process 61726 16.2 4.527 5.325 9.159 9.745
fft_wrap_pw1pw2_140 519 13.2 0.460 0.520 9.103 9.232
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 7.958 8.745
potential_pw2rs 127 12.3 0.016 0.018 8.598 8.649
calculate_dm_sparse 127 9.5 0.001 0.001 7.463 7.655
grid_integrate_task_list 127 12.3 7.042 7.553 7.042 7.553
mp_alltoall_d11v 2401 14.1 6.529 7.459 6.529 7.459
qs_ot_get_orbitals 116 10.6 0.001 0.001 6.825 7.085
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 7.018 7.025
mp_sum_d 4448 12.1 4.886 6.996 4.886 6.996
mp_allgather_i34 2485 14.5 3.337 6.949 3.337 6.949
dbcsr_complete_redistribute 393 12.7 0.773 0.897 5.267 6.476
mp_waitany 11660 13.9 3.473 6.154 3.473 6.154
dbcsr_dot_sd 1305 12.0 0.765 0.840 4.351 6.100
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.624 5.872
grid_collocate_task_list 127 9.7 4.666 5.832 4.666 5.832
make_images_sizes 4970 15.5 0.007 0.008 3.130 5.401
mp_alltoall_i44 4970 16.5 3.123 5.395 3.123 5.395
mp_alltoall_z22v 1281 16.7 5.123 5.365 5.123 5.365
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=256.597000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=789.090909, yerr=4.679832
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022121472 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444702699520 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796573E+12 0.0% 0.0% 100.0%
flops max/rank 2.906045E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499744 0.0% 0.0% 100.0%
number of processed stacks 3951168 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1697.1
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 932.597760E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1033760
MPI messages size (bytes):
total size 2.695213E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.607194E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 279168 36591108096
131072 < size <= 4194304 654272 987691483136
4194304 < size <= 16777216 65184 925172769472
16777216 < size 28448 745747251200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4001 58227.
MP_Allreduce 11000 1007.
MP_Sync 86
MP_Alltoall 1700 9383497.
MP_SendRecv 7874 75008.
MP_ISendRecv 7874 75008.
MP_Wait 21654
MP_ISend 11660 275234.
MP_IRecv 11660 275234.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.021 0.034 245.012 245.013
qs_mol_dyn_low 1 2.0 0.003 0.005 244.571 244.584
qs_forces 11 3.9 0.005 0.005 244.452 244.458
qs_energies 11 4.9 0.001 0.002 236.709 236.730
scf_env_do_scf 11 5.9 0.001 0.002 218.412 218.413
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 163.821 163.822
velocity_verlet 10 3.0 0.002 0.003 156.583 156.585
qs_scf_new_mos 116 7.6 0.001 0.001 116.669 117.293
qs_scf_loop_do_ot 116 8.6 0.001 0.001 116.668 117.292
ot_scf_mini 116 9.6 0.003 0.004 111.854 112.538
dbcsr_multiply_generic 2485 12.5 0.178 0.183 97.758 99.929
multiply_cannon 2485 13.5 0.496 0.522 70.331 78.408
multiply_cannon_loop 2485 14.5 0.850 0.872 65.635 70.595
ot_mini 116 10.6 0.001 0.001 54.768 55.666
init_scf_loop 11 6.9 0.000 0.000 54.461 54.464
mp_waitall_1 169034 16.6 38.293 51.164 38.293 51.164
prepare_preconditioner 11 7.9 0.000 0.000 49.456 49.568
make_preconditioner 11 8.9 0.000 0.000 49.456 49.568
make_full_inverse_cholesky 11 9.9 0.000 0.000 46.580 48.290
qs_ot_get_p 127 10.4 0.001 0.001 40.338 41.270
rebuild_ks_matrix 127 8.3 0.001 0.001 36.723 37.452
qs_ks_build_kohn_sham_matrix 127 9.3 0.015 0.017 36.722 37.451
multiply_cannon_metrocomm3 19880 15.5 0.058 0.065 22.645 35.247
qs_ot_get_derivative 116 11.6 0.002 0.002 33.191 33.924
qs_ot_p2m_diag 82 11.4 0.264 0.270 33.721 33.783
qs_ks_update_qs_env 127 7.6 0.001 0.001 32.994 33.659
cp_dbcsr_syevd 82 12.4 0.005 0.005 31.696 31.702
cp_fm_diag_elpa 82 13.4 0.000 0.000 26.693 26.708
cp_fm_diag_elpa_base 82 14.4 26.092 26.325 26.687 26.701
cp_fm_cholesky_invert 11 10.9 26.217 26.237 26.217 26.237
multiply_cannon_multrec 19880 15.5 12.694 15.987 21.294 24.691
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 21.207 22.825
apply_single 127 13.6 0.001 0.001 21.207 22.824
ot_diis_step 116 11.6 0.018 0.018 21.336 21.338
make_m2s 4970 13.5 0.079 0.082 19.280 20.794
make_images 4970 14.5 1.178 1.263 19.053 20.571
sum_up_and_integrate 127 10.3 0.131 0.145 19.017 19.045
integrate_v_rspace 127 11.3 0.004 0.004 18.886 18.914
qs_rho_update_rho_low 127 7.7 0.001 0.001 18.782 18.864
calculate_rho_elec 127 8.7 0.131 0.146 18.781 18.863
multiply_cannon_sync_h2d 19880 15.5 13.658 15.330 13.658 15.330
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 14.109 14.594
make_images_data 4970 15.5 0.059 0.067 12.098 14.208
cp_fm_cholesky_decompose 22 10.9 13.188 13.204 13.188 13.204
density_rs2pw 127 9.7 0.006 0.006 10.985 13.150
hybrid_alltoall_any 5155 16.4 0.433 1.981 10.295 12.716
init_scf_run 11 5.9 0.000 0.001 12.582 12.583
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.582 12.583
pw_transfer 1535 11.6 0.088 0.109 12.022 12.203
fft_wrap_pw1pw2 1281 12.7 0.011 0.012 11.796 11.984
rs_pw_transfer 1038 11.9 0.014 0.014 9.692 11.728
mp_sum_l 7804 13.0 6.395 11.123 6.395 11.123
fft3d_ps 1281 14.7 2.653 2.877 10.083 10.233
multiply_cannon_metrocomm4 17395 15.5 0.063 0.075 3.495 9.727
fft_wrap_pw1pw2_140 519 13.2 0.464 0.511 9.515 9.707
mp_irecv_dv 49801 16.2 3.378 9.484 3.378 9.484
wfi_extrapolate 11 7.9 0.001 0.001 9.457 9.458
mp_alltoall_d11v 2401 14.1 7.640 8.373 7.640 8.373
dbcsr_mm_accdrv_process 41158 16.2 4.663 5.361 8.154 8.283
dbcsr_complete_redistribute 393 12.7 1.172 1.206 6.078 8.234
cp_fm_upper_to_full 104 14.8 6.344 8.167 6.344 8.167
potential_pw2rs 127 12.3 0.020 0.022 7.874 7.956
grid_integrate_task_list 127 12.3 7.148 7.739 7.148 7.739
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.820 7.490
mp_allgather_i34 2485 14.5 3.221 7.305 3.221 7.305
calculate_dm_sparse 127 9.5 0.001 0.001 6.537 6.656
copy_fm_to_dbcsr 208 11.6 0.002 0.002 4.325 6.476
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.999 6.004
grid_collocate_task_list 127 9.7 4.861 5.749 4.861 5.749
mp_alltoall_z22v 1281 16.7 5.463 5.734 5.463 5.734
mp_sum_d 4449 12.1 3.134 5.488 3.134 5.488
mp_waitany 11660 13.9 3.234 5.486 3.234 5.486
qs_ot_get_orbitals 116 10.6 0.001 0.001 5.235 5.329
make_images_sizes 4970 15.5 0.007 0.007 2.622 5.219
mp_alltoall_i44 4970 16.5 2.616 5.212 2.616 5.212
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.857 4.983
mp_alltoall_i22 712 14.1 2.612 4.978 2.612 4.978
dbcsr_dot_sd 1305 12.0 1.040 1.055 2.992 4.967
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.744 4.929
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=245.013000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=886.363636, yerr=6.064393
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 1.160212E+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 4043 57632.
MP_Allreduce 11104 1087.
MP_Sync 88
MP_Alltoall 1724 12509575.
MP_SendRecv 5934 75008.
MP_ISendRecv 5934 75008.
MP_Wait 22612
MP_ISend 15064 244788.
MP_IRecv 15064 244788.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.031 227.026 227.027
qs_mol_dyn_low 1 2.0 0.003 0.004 226.582 226.595
qs_forces 11 3.9 0.004 0.004 226.355 226.362
qs_energies 11 4.9 0.001 0.002 218.420 218.429
scf_env_do_scf 11 5.9 0.002 0.004 199.431 199.438
velocity_verlet 10 3.0 0.002 0.002 150.562 150.569
scf_env_do_scf_inner_loop 118 6.6 0.003 0.009 141.195 141.197
qs_scf_new_mos 118 7.6 0.001 0.001 97.251 97.496
qs_scf_loop_do_ot 118 8.6 0.001 0.001 97.250 97.495
ot_scf_mini 118 9.6 0.003 0.004 92.446 92.716
dbcsr_multiply_generic 2529 12.6 0.187 0.193 87.091 87.665
multiply_cannon 2529 13.6 0.552 0.587 57.260 59.901
init_scf_loop 11 6.9 0.000 0.000 58.086 58.087
multiply_cannon_loop 2529 14.6 1.193 1.219 53.045 54.486
prepare_preconditioner 11 7.9 0.000 0.000 53.623 53.654
make_preconditioner 11 8.9 0.000 0.000 53.623 53.654
make_full_inverse_cholesky 11 9.9 0.000 0.000 46.492 52.236
ot_mini 118 10.6 0.001 0.001 46.989 47.258
rebuild_ks_matrix 129 8.3 0.001 0.001 34.258 34.465
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.018 34.257 34.464
mp_waitall_1 149172 16.7 23.587 33.289 23.587 33.289
multiply_cannon_multrec 30348 15.6 14.242 19.099 26.514 30.975
qs_ks_update_qs_env 129 7.6 0.001 0.001 30.731 30.915
qs_ot_get_p 129 10.4 0.001 0.001 30.474 30.728
qs_ot_get_derivative 118 11.6 0.002 0.004 26.955 27.221
make_m2s 5058 13.6 0.098 0.102 24.864 26.004
make_images 5058 14.6 1.949 2.286 24.551 25.693
qs_ot_p2m_diag 84 11.4 0.348 0.396 24.362 24.415
cp_dbcsr_syevd 84 12.4 0.005 0.005 22.770 22.778
cp_fm_cholesky_invert 11 10.9 21.654 21.666 21.654 21.666
ot_diis_step 118 11.6 0.018 0.019 19.887 19.890
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.299 19.709
apply_single 129 13.6 0.001 0.001 19.299 19.709
cp_fm_diag_elpa 84 13.4 0.000 0.000 18.601 18.620
cp_fm_diag_elpa_base 84 14.4 17.351 17.753 18.592 18.608
cp_fm_upper_to_full 106 14.8 12.336 18.552 12.336 18.552
sum_up_and_integrate 129 10.3 0.142 0.154 18.007 18.041
integrate_v_rspace 129 11.3 0.004 0.004 17.865 17.903
qs_rho_update_rho_low 129 7.7 0.001 0.001 17.777 17.819
calculate_rho_elec 129 8.7 0.177 0.192 17.776 17.818
make_images_data 5058 15.6 0.063 0.069 14.084 16.123
multiply_cannon_metrocomm3 30348 15.6 0.047 0.050 6.466 14.865
dbcsr_complete_redistribute 397 12.7 1.494 1.613 10.590 14.669
hybrid_alltoall_any 5245 16.5 0.530 2.211 12.107 14.566
init_scf_run 11 5.9 0.000 0.001 12.809 12.811
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.809 12.810
copy_fm_to_dbcsr 210 11.7 0.002 0.002 8.676 12.747
dbcsr_mm_accdrv_process 62780 16.2 7.542 8.561 11.849 12.375
multiply_cannon_sync_h2d 30348 15.6 11.666 12.178 11.666 12.178
cp_fm_cholesky_decompose 22 10.9 12.002 12.101 12.002 12.101
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 11.381 11.572
density_rs2pw 129 9.7 0.006 0.006 9.744 11.331
pw_transfer 1559 11.6 0.090 0.107 11.207 11.302
fft_wrap_pw1pw2 1301 12.7 0.011 0.012 10.976 11.075
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 7.108 11.058
mp_alltoall_i22 720 14.1 6.364 10.462 6.364 10.462
rs_pw_transfer 1054 12.0 0.014 0.014 8.297 9.934
fft_wrap_pw1pw2_140 527 13.2 0.480 0.490 9.380 9.499
wfi_extrapolate 11 7.9 0.001 0.001 9.350 9.351
fft3d_ps 1301 14.7 2.797 2.876 9.196 9.258
grid_integrate_task_list 129 12.3 7.557 7.959 7.557 7.959
mp_alltoall_d11v 2429 14.1 7.017 7.679 7.017 7.679
multiply_cannon_metrocomm4 25290 15.6 0.084 0.095 2.779 7.018
potential_pw2rs 129 12.3 0.023 0.023 6.951 7.004
mp_irecv_dv 76751 16.2 2.630 6.753 2.630 6.753
calculate_dm_sparse 129 9.5 0.001 0.001 6.643 6.745
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.349 6.380
grid_collocate_task_list 129 9.7 5.108 5.903 5.108 5.903
make_images_sizes 5058 15.6 0.007 0.007 2.502 5.400
mp_alltoall_i44 5058 16.6 2.495 5.393 2.495 5.393
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 5.181 5.261
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.851 4.903
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.549 4.613
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=227.027000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1085.272727, yerr=26.205797
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/19/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 5.909858E+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 1976192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3444.2
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 1.522688E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 242592
MPI messages size (bytes):
total size 1.341705E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.530704E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 114816 60196651008
4194304 < size <= 16777216 105840 554906419200
16777216 < size 20352 726591807936
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 9003 51.
MP_Alltoall 9708 795382.
MP_ISend 40388 2101966.
MP_IRecv 40388 2101070.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4030 57813.
MP_Allreduce 11072 1172.
MP_Sync 87
MP_Alltoall 1724 18848449.
MP_SendRecv 3870 122880.
MP_ISendRecv 3870 122880.
MP_Wait 16244
MP_ISend 10760 423501.
MP_IRecv 10760 423501.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.044 0.058 215.747 215.748
qs_mol_dyn_low 1 2.0 0.003 0.004 215.225 215.236
qs_forces 11 3.9 0.004 0.006 215.087 215.092
qs_energies 11 4.9 0.002 0.002 206.504 206.521
scf_env_do_scf 11 5.9 0.002 0.013 186.752 186.755
velocity_verlet 10 3.0 0.009 0.012 141.011 141.015
scf_env_do_scf_inner_loop 118 6.6 0.003 0.009 139.376 139.378
qs_scf_new_mos 118 7.6 0.001 0.001 93.225 93.373
qs_scf_loop_do_ot 118 8.6 0.001 0.001 93.225 93.373
ot_scf_mini 118 9.6 0.003 0.004 88.330 88.391
dbcsr_multiply_generic 2527 12.6 0.182 0.191 83.040 83.693
multiply_cannon 2527 13.6 0.596 0.631 58.166 63.786
multiply_cannon_loop 2527 14.6 0.448 0.456 52.956 54.122
init_scf_loop 11 6.9 0.000 0.000 47.022 47.025
ot_mini 118 10.6 0.001 0.001 45.028 45.092
prepare_preconditioner 11 7.9 0.000 0.000 42.599 42.627
make_preconditioner 11 8.9 0.000 0.000 42.599 42.627
make_full_inverse_cholesky 11 9.9 0.000 0.000 40.150 40.667
mp_waitall_1 126780 16.7 32.953 40.341 32.953 40.341
rebuild_ks_matrix 129 8.3 0.001 0.001 35.329 35.401
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 35.328 35.400
qs_ks_update_qs_env 129 7.6 0.001 0.001 31.639 31.708
qs_ot_get_p 129 10.4 0.001 0.001 29.204 29.286
qs_ot_get_derivative 118 11.6 0.002 0.002 24.754 24.815
cp_fm_cholesky_invert 11 10.9 24.346 24.353 24.346 24.353
qs_ot_p2m_diag 83 11.4 0.496 0.502 23.383 23.404
make_m2s 5054 13.6 0.067 0.072 20.440 22.860
make_images 5054 14.6 2.302 2.838 20.133 22.555
cp_dbcsr_syevd 83 12.4 0.005 0.005 21.700 21.708
multiply_cannon_multrec 10108 15.6 10.363 15.092 17.699 20.875
ot_diis_step 118 11.6 0.020 0.020 20.188 20.188
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.781 20.004
apply_single 129 13.6 0.001 0.001 19.780 20.004
qs_rho_update_rho_low 129 7.7 0.001 0.001 19.234 19.289
calculate_rho_elec 129 8.7 0.261 0.271 19.233 19.288
sum_up_and_integrate 129 10.3 0.183 0.193 19.229 19.279
integrate_v_rspace 129 11.3 0.004 0.005 19.046 19.099
multiply_cannon_metrocomm3 10108 15.6 0.023 0.024 11.529 18.455
cp_fm_diag_elpa 83 13.4 0.000 0.000 17.729 17.736
cp_fm_diag_elpa_base 83 14.4 17.248 17.562 17.721 17.727
make_images_data 5054 15.6 0.051 0.060 12.641 15.918
hybrid_alltoall_any 5240 16.5 0.826 3.622 11.875 15.027
pw_transfer 1559 11.6 0.088 0.096 13.045 13.104
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 12.819 12.884
multiply_cannon_sync_h2d 10108 15.6 11.548 12.181 11.548 12.181
cp_fm_cholesky_decompose 22 10.9 12.057 12.161 12.057 12.161
init_scf_run 11 5.9 0.000 0.001 12.079 12.079
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.078 12.079
multiply_cannon_metrocomm1 10108 15.6 0.028 0.029 8.414 11.889
fft3d_ps 1301 14.7 2.733 2.821 10.976 11.024
density_rs2pw 129 9.7 0.005 0.006 9.928 10.814
fft_wrap_pw1pw2_140 527 13.2 0.494 0.517 10.712 10.790
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 10.503 10.555
mp_alltoall_d11v 2423 14.1 8.313 9.582 8.313 9.582
wfi_extrapolate 11 7.9 0.001 0.001 8.951 8.951
mp_allgather_i34 2527 14.6 3.386 8.772 3.386 8.772
grid_integrate_task_list 129 12.3 7.749 8.291 7.749 8.291
rs_pw_transfer 1054 12.0 0.013 0.014 7.098 8.002
dbcsr_mm_accdrv_process 20926 16.1 2.690 3.490 6.976 7.611
potential_pw2rs 129 12.3 0.026 0.027 7.240 7.281
dbcsr_complete_redistribute 395 12.7 2.098 2.147 6.497 7.022
calculate_dm_sparse 129 9.5 0.001 0.001 6.594 6.737
grid_collocate_task_list 129 9.7 5.441 6.138 5.441 6.138
mp_alltoall_z22v 1301 16.7 5.470 5.693 5.470 5.693
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.522 5.524
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.478 5.486
make_images_sizes 5054 15.6 0.007 0.008 2.406 5.253
mp_alltoall_i44 5054 16.6 2.399 5.246 2.399 5.246
multiply_cannon_metrocomm4 7581 15.6 0.024 0.028 1.991 4.856
mp_irecv_dv 29086 15.9 1.954 4.769 1.954 4.769
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 4.600 4.606
copy_dbcsr_to_fm 186 11.8 0.004 0.004 4.501 4.581
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 4.509 4.520
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.973 4.461
yz_to_x 656 15.2 1.347 1.399 4.322 4.421
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=215.748000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1423.454545, yerr=48.355433
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 11.786061E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 1980288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3437.1
marketing flops 145.650931E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 3.016901E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101160
MPI messages size (bytes):
total size 1.144970E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.318403E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 44 2883584
131072 < size <= 4194304 45648 35433480192
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592466352
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 58564.
MP_Allreduce 11104 1511.
MP_Sync 88
MP_Alltoall 1724 36993632.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.041 0.066 304.523 304.528
qs_mol_dyn_low 1 2.0 0.003 0.005 303.937 303.956
qs_forces 11 3.9 0.005 0.005 303.830 303.836
qs_energies 11 4.9 0.002 0.002 294.277 294.284
scf_env_do_scf 11 5.9 0.004 0.005 271.481 271.494
velocity_verlet 10 3.0 0.002 0.003 219.328 219.335
scf_env_do_scf_inner_loop 118 6.6 0.004 0.009 146.170 146.172
init_scf_loop 11 6.9 0.000 0.000 125.052 125.056
prepare_preconditioner 11 7.9 0.000 0.000 119.913 119.934
make_preconditioner 11 8.9 0.000 0.000 119.913 119.934
make_full_inverse_cholesky 11 9.9 0.000 0.000 96.287 117.037
qs_scf_new_mos 118 7.6 0.001 0.001 93.872 93.920
qs_scf_loop_do_ot 118 8.6 0.001 0.001 93.872 93.919
ot_scf_mini 118 9.6 0.004 0.004 88.957 88.976
dbcsr_multiply_generic 2529 12.6 0.212 0.225 85.128 85.517
cp_fm_upper_to_full 106 14.8 52.879 75.476 52.879 75.476
multiply_cannon 2529 13.6 0.713 0.750 59.780 61.227
multiply_cannon_loop 2529 14.6 0.475 0.482 55.613 56.831
ot_mini 118 10.6 0.001 0.001 45.894 45.948
dbcsr_complete_redistribute 397 12.7 4.002 4.085 29.518 42.159
rebuild_ks_matrix 129 8.3 0.001 0.001 39.608 39.652
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.018 39.607 39.651
copy_fm_to_dbcsr 210 11.7 0.002 0.002 25.933 38.599
qs_ks_update_qs_env 129 7.6 0.001 0.001 36.441 36.479
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 23.581 36.124
cp_fm_cholesky_invert 11 10.9 34.025 34.031 34.025 34.031
mp_alltoall_i22 720 14.1 21.391 33.977 21.391 33.977
mp_waitall_1 104580 16.8 29.211 32.987 29.211 32.987
qs_ot_get_p 129 10.4 0.001 0.001 27.126 27.227
qs_ot_get_derivative 118 11.6 0.003 0.006 24.855 24.883
qs_ot_p2m_diag 84 11.4 0.889 0.893 22.859 22.890
make_m2s 5058 13.6 0.078 0.081 20.613 22.044
make_images 5058 14.6 3.794 3.908 20.130 21.568
qs_rho_update_rho_low 129 7.7 0.001 0.001 21.164 21.179
calculate_rho_elec 129 8.7 0.487 0.488 21.163 21.178
cp_dbcsr_syevd 84 12.4 0.006 0.006 21.087 21.102
ot_diis_step 118 11.6 0.022 0.023 20.995 20.996
sum_up_and_integrate 129 10.3 0.324 0.327 20.483 20.558
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 20.299 20.431
apply_single 129 13.6 0.001 0.001 20.299 20.430
integrate_v_rspace 129 11.3 0.004 0.004 20.159 20.234
multiply_cannon_metrocomm3 10116 15.6 0.023 0.025 19.164 20.152
multiply_cannon_multrec 10116 15.6 10.621 12.445 18.085 18.142
cp_fm_diag_elpa 84 13.4 0.000 0.000 17.775 17.776
cp_fm_diag_elpa_base 84 14.4 13.253 14.816 17.768 17.770
multiply_cannon_sync_h2d 10116 15.6 15.771 15.793 15.771 15.793
hybrid_alltoall_any 5245 16.5 1.311 3.059 11.415 13.701
make_images_data 5058 15.6 0.060 0.065 11.233 13.486
pw_transfer 1559 11.6 0.094 0.096 12.717 12.727
init_scf_run 11 5.9 0.000 0.001 12.666 12.667
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.666 12.667
fft_wrap_pw1pw2 1301 12.7 0.011 0.012 12.479 12.490
fft_wrap_pw1pw2_140 527 13.2 0.547 0.549 11.092 11.102
fft3d_ps 1301 14.7 2.769 2.781 10.497 10.518
mp_alltoall_d11v 2429 14.1 9.031 9.870 9.031 9.870
wfi_extrapolate 11 7.9 0.001 0.001 9.515 9.515
qs_ot_get_derivative_diag 78 12.4 0.002 0.002 9.447 9.464
cp_fm_cholesky_decompose 22 10.9 9.138 9.198 9.138 9.198
dbcsr_mm_accdrv_process 20934 16.1 3.920 5.660 7.223 9.011
grid_integrate_task_list 129 12.3 8.641 8.821 8.641 8.821
density_rs2pw 129 9.7 0.005 0.005 8.553 8.712
qs_energies_init_hamiltonians 11 5.9 0.002 0.007 8.120 8.122
calculate_dm_sparse 129 9.5 0.001 0.001 6.808 6.843
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.517 6.646
copy_dbcsr_to_fm 187 11.8 0.004 0.004 6.403 6.606
rs_scatter_matrices 140 9.7 3.647 4.520 6.240 6.481
grid_collocate_task_list 129 9.7 6.408 6.447 6.408 6.447
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=304.528000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2683.636364, yerr=173.791554
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 1.262125E+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 6805335.
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 0.035 0.107 86.739 86.755
qs_energies 1 2.0 0.000 0.001 86.036 86.051
ls_scf 1 3.0 0.000 0.000 85.119 85.135
dbcsr_multiply_generic 111 6.7 0.014 0.015 73.333 73.475
multiply_cannon 111 7.7 0.018 0.021 56.233 57.369
multiply_cannon_loop 111 8.7 0.205 0.217 52.701 54.372
ls_scf_main 1 4.0 0.000 0.000 53.080 53.092
density_matrix_trs4 2 5.0 0.002 0.003 47.415 47.488
ls_scf_init_scf 1 4.0 0.000 0.000 28.955 28.957
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.769 27.821
mp_waitall_1 11031 10.9 23.091 25.775 23.091 25.775
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.680 25.698
multiply_cannon_multrec 2664 9.7 8.160 8.963 15.459 17.303
multiply_cannon_sync_h2d 2664 9.7 13.422 15.216 13.422 15.216
make_m2s 222 7.7 0.008 0.010 13.230 13.813
make_images 222 8.7 0.099 0.111 13.208 13.794
multiply_cannon_metrocomm1 2664 9.7 0.009 0.010 10.172 12.445
make_images_data 222 9.7 0.004 0.005 7.749 8.378
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.489 7.932
dbcsr_mm_accdrv_process 4760 10.4 0.508 0.636 6.917 7.908
hybrid_alltoall_any 227 10.6 0.216 1.840 6.586 7.904
dbcsr_mm_accdrv_process_sort 4760 11.4 6.210 7.091 6.210 7.091
calculate_norms 4752 9.8 5.499 6.168 5.499 6.168
mp_sum_l 807 5.4 3.488 5.305 3.488 5.305
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.036 5.171
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.562 3.936
arnoldi_extremal 4 6.8 0.000 0.000 3.813 3.833
arnoldi_normal_ev 4 7.8 0.001 0.004 3.813 3.833
build_subspace 16 8.4 0.009 0.012 3.710 3.713
make_images_sizes 222 9.7 0.000 0.000 0.849 3.386
mp_alltoall_i44 222 10.7 0.848 3.386 0.848 3.386
multiply_cannon_metrocomm4 2442 9.7 0.011 0.014 2.058 3.381
mp_irecv_dv 6231 10.9 2.040 3.353 2.040 3.353
ls_scf_post 1 4.0 0.000 0.000 3.083 3.099
ls_scf_store_result 1 5.0 0.000 0.000 2.878 2.929
dbcsr_matrix_vector_mult 304 9.0 0.006 0.013 2.555 2.780
dbcsr_special_finalize 555 9.7 0.005 0.006 2.294 2.742
dbcsr_merge_single_wm 555 10.7 0.463 0.622 2.286 2.733
make_images_pack 222 9.7 2.192 2.599 2.194 2.601
dbcsr_sort_data 658 11.4 2.082 2.487 2.082 2.487
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.427 2.480
dbcsr_matrix_vector_mult_local 304 10.0 2.008 2.415 2.010 2.417
buffer_matrices_ensure_size 222 8.7 1.765 2.171 1.765 2.171
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.865 1.866
rebuild_ks_matrix 3 7.3 0.000 0.000 1.856 1.857
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.856 1.857
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=86.755000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1143.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 2.111107E+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 0.050 0.067 90.909 90.909
qs_energies 1 2.0 0.000 0.000 90.400 90.403
ls_scf 1 3.0 0.000 0.000 89.071 89.074
dbcsr_multiply_generic 111 6.7 0.017 0.034 74.956 75.382
multiply_cannon 111 7.7 0.040 0.065 53.147 56.783
ls_scf_main 1 4.0 0.000 0.000 54.599 54.605
multiply_cannon_loop 111 8.7 0.116 0.123 49.885 53.186
density_matrix_trs4 2 5.0 0.002 0.003 48.789 49.003
ls_scf_init_scf 1 4.0 0.000 0.000 30.925 30.927
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.709 29.836
mp_waitall_1 9105 10.9 20.735 29.806 20.735 29.806
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.302 27.312
multiply_cannon_multrec 1332 9.7 13.203 16.299 22.425 26.603
multiply_cannon_metrocomm3 1332 9.7 0.006 0.007 11.622 20.759
make_m2s 222 7.7 0.006 0.008 15.043 15.694
make_images 222 8.7 1.575 1.923 15.013 15.664
dbcsr_mm_accdrv_process 4041 10.4 0.307 0.448 8.821 10.369
dbcsr_mm_accdrv_process_sort 4041 11.4 8.409 9.921 8.409 9.921
make_images_data 222 9.7 0.004 0.004 8.478 9.329
hybrid_alltoall_any 227 10.6 0.521 2.476 7.899 8.851
mp_sum_l 807 5.4 5.457 8.473 5.457 8.473
multiply_cannon_metrocomm4 1221 9.7 0.006 0.008 3.232 7.739
mp_irecv_dv 3311 11.0 3.212 7.667 3.212 7.667
calculate_norms 2376 9.8 6.006 6.743 6.006 6.743
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.230 6.706
multiply_cannon_sync_h2d 1332 9.7 4.750 6.247 4.750 6.247
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.122 5.343
arnoldi_extremal 4 6.8 0.000 0.000 4.737 4.753
arnoldi_normal_ev 4 7.8 0.008 0.018 4.737 4.753
build_subspace 16 8.4 0.014 0.021 4.456 4.465
ls_scf_post 1 4.0 0.000 0.000 3.547 3.550
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.149 3.372
ls_scf_store_result 1 5.0 0.000 0.000 3.246 3.371
dbcsr_matrix_vector_mult_local 304 10.0 2.739 3.224 2.741 3.225
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 1.265 2.965
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.624 2.725
make_images_pack 222 9.7 2.026 2.438 2.028 2.440
mp_allgather_i34 111 8.7 0.943 2.219 0.943 2.219
dbcsr_sort_data 436 11.2 1.828 2.022 1.828 2.022
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.879 1.881
rebuild_ks_matrix 3 7.3 0.000 0.000 1.866 1.868
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.866 1.868
dbcsr_data_new 4174 10.1 1.610 1.848 1.610 1.848
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=90.909000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1781.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 2.760114E+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 0.037 0.057 97.360 97.361
qs_energies 1 2.0 0.000 0.000 96.737 96.740
ls_scf 1 3.0 0.000 0.000 95.322 95.326
dbcsr_multiply_generic 111 6.7 0.015 0.016 78.705 79.011
ls_scf_main 1 4.0 0.000 0.000 59.723 59.726
multiply_cannon 111 7.7 0.042 0.096 53.766 57.568
multiply_cannon_loop 111 8.7 0.100 0.104 50.171 54.081
density_matrix_trs4 2 5.0 0.002 0.003 52.977 53.169
mp_waitall_1 7281 11.0 25.443 35.602 25.443 35.602
ls_scf_init_scf 1 4.0 0.000 0.000 32.024 32.026
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.655 30.739
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.236 28.248
multiply_cannon_multrec 888 9.7 12.605 15.257 21.113 24.365
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.369 24.290
make_m2s 222 7.7 0.006 0.007 17.932 19.188
make_images 222 8.7 1.960 2.277 17.894 19.147
hybrid_alltoall_any 227 10.6 0.620 2.866 9.904 11.744
make_images_data 222 9.7 0.003 0.004 10.360 11.568
dbcsr_mm_accdrv_process 3754 10.4 0.246 0.415 8.041 9.256
mp_sum_l 807 5.4 5.458 9.012 5.458 9.012
dbcsr_mm_accdrv_process_sort 3754 11.4 7.672 8.841 7.672 8.841
multiply_cannon_sync_h2d 888 9.7 5.991 7.276 5.991 7.276
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 4.322 7.260
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.328 7.200
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.469 6.714
mp_irecv_dv 2335 11.1 2.456 6.671 2.456 6.671
arnoldi_extremal 4 6.8 0.000 0.000 5.385 5.403
arnoldi_normal_ev 4 7.8 0.001 0.006 5.385 5.402
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.972 5.170
build_subspace 16 8.4 0.014 0.020 5.081 5.090
calculate_norms 1584 9.8 4.282 4.684 4.282 4.684
mp_allgather_i34 111 8.7 1.420 3.946 1.420 3.946
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.566 3.884
dbcsr_matrix_vector_mult_local 304 10.0 3.010 3.590 3.012 3.592
ls_scf_post 1 4.0 0.000 0.000 3.575 3.578
ls_scf_store_result 1 5.0 0.000 0.000 3.330 3.402
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.257 3.351
make_images_sizes 222 9.7 0.000 0.000 1.352 2.887
mp_alltoall_i44 222 10.7 1.352 2.887 1.352 2.887
make_images_pack 222 9.7 1.811 2.117 1.814 2.120
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.096 2.098
dbcsr_sort_data 325 11.1 1.819 2.091 1.819 2.091
rebuild_ks_matrix 3 7.3 0.000 0.000 2.078 2.080
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.078 2.080
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=97.361000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2171.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 3.326034E+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 0.054 0.077 97.599 97.600
qs_energies 1 2.0 0.000 0.000 96.885 96.892
ls_scf 1 3.0 0.001 0.010 95.228 95.235
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.435 78.681
ls_scf_main 1 4.0 0.002 0.020 58.829 58.831
multiply_cannon 111 7.7 0.062 0.127 51.395 56.209
density_matrix_trs4 2 5.0 0.003 0.020 52.630 52.758
multiply_cannon_loop 111 8.7 0.115 0.127 46.297 49.840
ls_scf_init_scf 1 4.0 0.000 0.002 33.014 33.015
ls_scf_init_matrix_S 1 5.0 0.000 0.001 31.576 31.650
mp_waitall_1 6369 11.0 22.536 29.792 22.536 29.792
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.120 29.134
multiply_cannon_multrec 1332 9.7 14.237 17.964 22.087 25.645
make_m2s 222 7.7 0.006 0.008 21.268 22.735
make_images 222 8.7 3.144 3.610 21.218 22.686
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.020 17.284
make_images_data 222 9.7 0.004 0.004 11.887 13.552
hybrid_alltoall_any 227 10.6 0.797 3.807 11.202 12.925
dbcsr_mm_accdrv_process 3641 10.4 0.231 0.421 7.486 8.998
dbcsr_mm_accdrv_process_sort 3641 11.4 7.108 8.582 7.108 8.582
mp_sum_l 807 5.4 4.198 7.322 4.198 7.322
multiply_cannon_sync_h2d 1332 9.7 5.525 6.228 5.525 6.228
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.078 6.131
mp_irecv_dv 3229 10.9 2.055 6.044 2.055 6.044
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.191 5.772
arnoldi_extremal 4 6.8 0.000 0.000 5.447 5.462
arnoldi_normal_ev 4 7.8 0.001 0.005 5.447 5.462
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.459 5.385
build_subspace 16 8.4 0.014 0.021 5.098 5.110
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.651 4.836
mp_allgather_i34 111 8.7 2.217 4.664 2.217 4.664
calculate_norms 2376 9.8 4.180 4.572 4.180 4.572
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.644 3.919
dbcsr_matrix_vector_mult_local 304 10.0 3.182 3.686 3.184 3.688
dbcsr_sort_data 658 11.4 3.068 3.485 3.068 3.485
ls_scf_post 1 4.0 0.000 0.000 3.383 3.389
dbcsr_special_finalize 555 9.7 0.006 0.007 2.810 3.255
dbcsr_merge_single_wm 555 10.7 0.531 0.663 2.802 3.246
ls_scf_store_result 1 5.0 0.000 0.000 3.071 3.131
ls_scf_dm_to_ks 2 5.0 0.001 0.015 3.067 3.110
dbcsr_data_release 10477 10.7 1.589 2.399 1.589 2.399
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.045 2.046
rebuild_ks_matrix 3 7.3 0.000 0.000 2.005 2.007
qs_ks_build_kohn_sham_matrix 3 8.3 0.002 0.017 2.005 2.007
dbcsr_finalize 304 7.8 0.049 0.061 1.801 1.990
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.600000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2721.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 4.630757E+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 0.044 0.056 103.012 103.013
qs_energies 1 2.0 0.000 0.000 102.149 102.155
ls_scf 1 3.0 0.000 0.000 100.195 100.206
dbcsr_multiply_generic 111 6.7 0.017 0.018 80.366 80.546
ls_scf_main 1 4.0 0.000 0.000 64.409 64.410
multiply_cannon 111 7.7 0.075 0.152 56.175 61.878
density_matrix_trs4 2 5.0 0.002 0.003 56.680 56.773
multiply_cannon_loop 111 8.7 0.069 0.078 51.740 53.532
mp_waitall_1 5436 11.0 28.474 34.113 28.474 34.113
ls_scf_init_scf 1 4.0 0.000 0.000 32.151 32.156
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.662 30.710
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.476 28.487
multiply_cannon_multrec 444 9.7 13.942 16.443 20.964 23.306
make_m2s 222 7.7 0.004 0.005 19.078 21.475
make_images 222 8.7 3.727 4.437 19.016 21.415
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 12.345 17.312
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.041 13.804
make_images_data 222 9.7 0.003 0.004 10.852 13.606
hybrid_alltoall_any 227 10.6 0.787 3.740 10.528 13.209
multiply_cannon_sync_h2d 444 9.7 6.494 8.173 6.494 8.173
dbcsr_mm_accdrv_process 3003 10.4 0.163 0.341 6.731 7.872
dbcsr_mm_accdrv_process_sort 3003 11.4 6.418 7.530 6.418 7.530
mp_allgather_i34 111 8.7 2.662 7.205 2.662 7.205
arnoldi_extremal 4 6.8 0.000 0.000 6.370 6.379
arnoldi_normal_ev 4 7.8 0.002 0.006 6.369 6.379
build_subspace 16 8.4 0.015 0.019 5.962 5.977
mp_sum_l 807 5.4 3.152 5.534 3.152 5.534
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.671 4.833
make_images_sizes 222 9.7 0.000 0.000 1.575 4.556
mp_alltoall_i44 222 10.7 1.575 4.555 1.575 4.555
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.353 4.543
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.695 4.399
mp_irecv_dv 1241 11.2 1.676 4.371 1.676 4.371
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.231 4.266
dbcsr_matrix_vector_mult_local 304 10.0 3.670 4.139 3.672 4.141
ls_scf_dm_to_ks 2 5.0 0.000 0.000 4.008 4.107
calculate_norms 792 9.8 3.546 3.696 3.546 3.696
ls_scf_post 1 4.0 0.000 0.000 3.636 3.641
ls_scf_store_result 1 5.0 0.000 0.000 3.426 3.461
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.429 2.430
rebuild_ks_matrix 3 7.3 0.000 0.000 2.397 2.398
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.397 2.398
dbcsr_finalize 304 7.8 0.062 0.077 2.199 2.255
dbcsr_merge_all 275 8.9 0.472 0.519 2.048 2.105
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=103.013000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3654.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_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 8.721187E+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 0.083 0.103 108.206 108.206
qs_energies 1 2.0 0.000 0.000 106.769 106.783
ls_scf 1 3.0 0.000 0.000 103.811 103.825
dbcsr_multiply_generic 111 6.7 0.023 0.026 77.075 77.166
ls_scf_main 1 4.0 0.000 0.000 65.832 65.833
density_matrix_trs4 2 5.0 0.002 0.003 56.480 56.527
multiply_cannon 111 7.7 0.128 0.267 49.429 51.620
multiply_cannon_loop 111 8.7 0.067 0.070 46.005 46.651
ls_scf_init_scf 1 4.0 0.000 0.000 34.328 34.329
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.544 32.560
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.722 29.732
mp_waitall_1 4527 11.1 21.881 25.377 21.881 25.377
make_m2s 222 7.7 0.005 0.005 24.147 25.115
make_images 222 8.7 4.593 4.979 24.041 25.006
multiply_cannon_multrec 444 9.7 17.950 18.655 22.553 23.289
hybrid_alltoall_any 227 10.6 1.661 3.624 12.981 16.056
make_images_data 222 9.7 0.003 0.003 13.295 15.767
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.003 10.633
multiply_cannon_sync_h2d 444 9.7 8.848 8.889 8.848 8.889
arnoldi_extremal 4 6.8 0.000 0.000 7.382 7.397
arnoldi_normal_ev 4 7.8 0.004 0.009 7.382 7.397
build_subspace 16 8.4 0.026 0.037 6.818 6.834
dbcsr_matrix_vector_mult 304 9.0 0.017 0.033 5.442 5.601
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.300 5.391
dbcsr_matrix_vector_mult_local 304 10.0 5.010 5.316 5.012 5.318
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.020 5.274
dbcsr_mm_accdrv_process 1814 10.4 0.231 0.319 4.427 4.544
dbcsr_mm_accdrv_process_sort 1814 11.4 4.127 4.251 4.127 4.251
make_images_sizes 222 9.7 0.000 0.000 1.510 3.709
mp_alltoall_i44 222 10.7 1.510 3.709 1.510 3.709
ls_scf_post 1 4.0 0.000 0.000 3.651 3.664
ls_scf_store_result 1 5.0 0.000 0.000 3.402 3.409
mp_allgather_i34 111 8.7 1.068 3.339 1.068 3.339
calculate_norms 792 9.8 3.241 3.279 3.241 3.279
dbcsr_finalize 304 7.8 0.082 0.089 3.090 3.169
dbcsr_merge_all 275 8.9 0.892 0.916 2.876 2.949
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.928 2.928
dbcsr_complete_redistribute 5 7.6 1.430 1.458 2.764 2.883
dbcsr_data_release 12724 10.6 2.344 2.837 2.344 2.837
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.612 2.614
rebuild_ks_matrix 3 7.3 0.000 0.000 2.546 2.548
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.546 2.548
matrix_ls_to_qs 2 6.0 0.000 0.000 2.415 2.534
dbcsr_sort_data 325 11.1 2.443 2.496 2.443 2.496
dbcsr_new_transposed 4 7.5 0.244 0.253 2.287 2.303
dbcsr_frobenius_norm 74 6.6 2.056 2.137 2.199 2.239
dbcsr_add_d 103 6.2 0.000 0.000 2.132 2.202
dbcsr_add_anytype 103 7.2 0.859 0.892 2.132 2.202
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=108.206000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6858.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/b22e23451d9407afd60f6019384990b988ad9a21_performance_tests/27/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 7009386627072 0.0% 0.0% 100.0%
flops 9 x 9 x 32 7335108845568 0.0% 0.0% 100.0%
flops 9 x 22 x 32 9866241589248 0.0% 0.0% 100.0%
flops 22 x 9 x 32 9884108906496 0.0% 0.0% 100.0%
flops 22 x 22 x 32 13354440523776 0.0% 0.0% 100.0%
flops 32 x 32 x 9 20607185977344 0.0% 0.0% 100.0%
flops 32 x 32 x 22 25186560638976 0.0% 0.0% 100.0%
flops 9 x 32 x 32 28458319085568 0.0% 0.0% 100.0%
flops 22 x 32 x 32 34782389993472 0.0% 0.0% 100.0%
flops 9 x 32 x 9 42881542373376 0.0% 0.0% 100.0%
flops 22 x 32 x 9 55680402235392 0.0% 0.0% 100.0%
flops 9 x 32 x 22 55680402235392 0.0% 0.0% 100.0%
flops 22 x 32 x 22 72328573419520 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 383.054662E+12 0.0% 0.0% 100.0%
flops max/rank 733.641090E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 26899403712 0.0% 0.0% 100.0%
number of processed stacks 118860288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 226.3
marketing flops 780.439111E+12
-------------------------------------------------------------------------------
# multiplications 1445
max memory usage/rank 586.162176E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 76559.
MP_Allreduce 13152 2094.
MP_Sync 1064
MP_Alltoall 2588 881903851.
MP_SendRecv 168740 11136.
MP_ISendRecv 92040 11136.
MP_Wait 102830
MP_comm_split 40
MP_ISend 26090 85106.
MP_IRecv 37890 59644.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.051 0.103 351.026 351.028
qs_mol_dyn_low 1 2.0 0.003 0.017 349.878 349.898
qs_forces 5 3.8 0.005 0.029 349.774 349.777
qs_energies 5 4.8 0.003 0.041 345.649 345.693
scf_env_do_scf 5 5.8 0.001 0.011 328.852 328.855
scf_env_do_scf_inner_loop 105 6.6 0.005 0.070 280.604 280.637
qs_scf_new_mos 105 7.6 0.000 0.001 216.810 217.447
qs_scf_loop_do_ot 105 8.6 0.001 0.001 216.809 217.446
ot_scf_mini 105 9.6 0.003 0.004 205.228 205.975
dbcsr_multiply_generic 1445 12.2 0.121 0.128 148.713 152.147
velocity_verlet 4 3.0 0.003 0.016 143.694 143.696
multiply_cannon 1445 13.2 0.281 0.295 120.968 129.810
multiply_cannon_loop 1445 14.2 2.365 2.487 117.139 126.508
qs_ot_get_p 112 10.4 0.001 0.001 102.585 103.052
qs_ot_p2m_diag 40 11.0 0.020 0.030 86.962 87.089
cp_dbcsr_syevd 40 12.0 0.002 0.002 82.213 82.221
cp_fm_syevd 40 13.0 0.000 0.000 73.629 73.800
mp_waitall_1 488190 16.1 50.426 70.057 50.426 70.057
ot_mini 105 10.6 0.001 0.001 69.365 69.622
cp_fm_redistribute_end 40 14.0 33.735 67.276 33.749 67.289
cp_fm_syevd_base 40 14.0 33.511 67.059 33.511 67.059
multiply_cannon_metrocomm3 69360 15.2 0.187 0.199 31.784 54.379
init_scf_loop 7 6.6 0.001 0.048 48.149 48.156
qs_ot_get_derivative 55 11.6 0.002 0.017 46.767 47.013
multiply_cannon_multrec 69360 15.2 30.442 36.398 38.277 43.781
prepare_preconditioner 7 7.6 0.000 0.001 41.400 41.445
make_preconditioner 7 8.6 0.001 0.041 41.400 41.445
rebuild_ks_matrix 110 8.4 0.000 0.000 40.256 40.752
qs_ks_build_kohn_sham_matrix 110 9.4 0.013 0.053 40.256 40.752
qs_ks_update_qs_env 112 7.6 0.001 0.001 36.982 37.446
multiply_cannon_sync_h2d 69360 15.2 28.754 33.782 28.754 33.782
qs_rho_update_rho_low 110 7.6 0.001 0.001 32.728 33.336
calculate_rho_elec 110 8.6 0.029 0.032 32.728 33.335
make_full_inverse_cholesky 7 9.6 0.000 0.000 31.127 31.203
rs_pw_transfer 690 11.5 0.011 0.013 28.894 30.234
density_rs2pw 110 9.6 0.005 0.006 24.433 25.804
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 23.475 23.791
apply_single 62 13.6 0.000 0.000 23.474 23.791
multiply_cannon_metrocomm1 69360 15.2 0.089 0.095 7.360 23.511
ot_new_cg_direction 55 11.6 0.001 0.001 21.686 21.689
cp_fm_cholesky_invert 7 10.6 20.659 20.679 20.659 20.679
qs_ot_get_orbitals 105 10.6 0.001 0.001 17.048 18.000
mp_sum_l 4684 12.4 15.527 17.943 15.527 17.943
pw_transfer 1645 12.4 0.079 0.101 17.283 17.720
fft_wrap_pw1pw2 1425 13.5 0.012 0.014 17.145 17.583
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 17.193 17.405
sum_up_and_integrate 60 10.3 0.028 0.043 15.914 15.925
integrate_v_rspace 60 11.3 0.002 0.003 15.886 15.898
fft_wrap_pw1pw2_240 915 15.0 0.414 0.458 14.455 14.804
fft3d_pb 915 16.0 2.290 2.662 13.303 13.648
check_diag 80 13.5 8.589 8.881 13.447 13.644
init_scf_run 5 5.8 0.000 0.001 13.572 13.573
scf_env_initial_rho_setup 5 6.8 0.000 0.001 13.571 13.573
calculate_dm_sparse 110 9.5 0.000 0.001 13.316 13.566
qs_vxc_create 110 10.4 0.002 0.020 13.502 13.548
rs_pw_transfer_RS2PW_30 110 11.6 1.543 1.641 12.068 12.807
mp_sendrecv_dv 168740 12.6 12.510 12.693 12.510 12.693
make_m2s 2890 13.2 0.075 0.080 11.768 12.618
make_images 2890 14.2 0.247 0.267 11.667 12.512
qs_ot_get_derivative_diag 18 12.0 0.000 0.000 12.311 12.494
potential_pw2rs 60 12.3 0.003 0.003 11.343 11.404
cp_fm_cholesky_decompose 14 10.2 9.822 9.832 9.822 9.832
make_full_single_inverse 7 9.6 0.001 0.050 9.732 9.778
mp_alltoall_z22v 2340 17.7 8.608 9.576 8.608 9.576
xc_rho_set_and_dset_create 110 12.4 0.086 0.200 9.130 9.450
xc_vxc_pw_create 60 11.3 0.041 0.079 9.139 9.186
xc_pw_derive 510 13.4 0.005 0.006 8.868 8.954
mp_alltoall_d11v 1300 13.8 7.761 8.865 7.761 8.865
dbcsr_mm_accdrv_process 154766 15.8 4.007 4.278 7.713 8.679
make_images_sizes 2890 15.2 0.004 0.004 6.211 7.318
mp_alltoall_i44 2890 16.2 6.208 7.314 6.208 7.314
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=351.028000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=558.600000, yerr=1.854724
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: b22e23451d9407afd60f6019384990b988ad9a21
Summary: empty
Status: OK