=== This is the CP2K Performance-Test ===
Already up to date.
Current branch master is up to date.
Already up to date.
Current branch master is up to date.
GIT Revision: 53a226acd6080c9664e5d7b371c6f14b2e21dcc0
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1,
# Cray-FFTW 3.3.8.10, COSMA 2.6.6, ELPA 2023.05.001,
# HDF5 1.14.2, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.2.2, LIBVORI 220621, LIBXSMM 1.17,
# PLUMED 2.9.0, SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Last update: 04.10.2023
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2023.05.001
USE_HDF5 := 1.14.2
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.9.0
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(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))
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__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 += $(LIBVDWXC_LIB)/libvdwxc.a
endif
ifneq ($(USE_HDF5),)
USE_HDF5 := $(strip $(USE_HDF5))
HDF5_INC := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/include
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/lib
CFLAGS += -I$(HDF5_INC)
DFLAGS += -D__HDF5
LIBS += $(HDF5_LIB)/libhdf5_fortran.a
LIBS += $(HDF5_LIB)/libhdf5_hl.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/01
job id: 49356755
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/02
job id: 49356756
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/03
job id: 49356757
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/04
job id: 49356758
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/05
job id: 49356759
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/06
job id: 49356760
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/07
job id: 49356761
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/08
job id: 49356762
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/09
job id: 49356765
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/10
job id: 49356766
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/11
job id: 49356767
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/12
job id: 49356768
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/13
job id: 49356769
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/14
job id: 49356770
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/15
job id: 49356771
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/16
job id: 49356772
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/17
job id: 49356773
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/18
job id: 49356774
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/19
job id: 49356775
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/20
job id: 49356776
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/21
job id: 49356777
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/22
job id: 49356778
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/23
job id: 49356779
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/24
job id: 49356780
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/25
job id: 49356781
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/26
job id: 49356782
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/27
job id: 49356783
--- 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/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 424 8.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.035 133.032 133.033
farming_run 1 2.0 132.519 132.520 133.005 133.008
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.489846E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1134128.
MP_Allreduce 571 1938539.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.005 0.024 114.767 114.767
qs_energies 1 2.0 0.000 0.000 114.574 114.576
mp2_main 1 3.0 0.000 0.000 112.518 112.521
mp2_gpw_main 1 4.0 0.028 0.034 111.532 111.534
mp2_ri_gpw_compute_in 1 5.0 0.171 0.173 92.685 92.991
mp2_ri_gpw_compute_in_loop 1 6.0 0.003 0.004 54.666 54.973
mp2_eri_3c_integrate_gpw 272 7.0 0.151 0.164 41.069 46.074
get_2c_integrals 1 6.0 0.008 0.009 36.990 37.846
integrate_v_rspace 273 8.0 0.438 0.454 24.632 29.530
pw_transfer 6555 10.6 0.379 0.393 26.827 27.375
fft_wrap_pw1pw2 5465 11.4 0.045 0.047 25.451 25.968
grid_integrate_task_list 273 9.0 20.536 25.916 20.536 25.916
fft_wrap_pw1pw2_100 2178 12.4 1.123 1.393 23.014 23.535
compute_2c_integrals 1 7.0 0.002 0.003 19.148 19.149
rpa_ri_compute_en 1 5.0 0.019 0.022 18.730 18.930
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.623 18.871
mp2_eri_2c_integrate_gpw 1 9.0 2.385 2.427 18.621 18.870
cp_fm_cholesky_decompose 12 8.2 17.768 18.632 17.768 18.632
cholesky_decomp 1 7.0 0.000 0.000 16.694 17.553
fft3d_s 5443 13.4 16.134 16.427 16.157 16.449
ao_to_mo_and_store_B_mult_1 272 7.0 10.763 15.307 10.763 15.307
calculate_wavefunction 272 8.0 5.418 5.556 12.337 12.998
rpa_num_int 1 6.0 0.000 0.001 10.582 10.591
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.569 10.578
calc_mat_Q 8 8.0 0.000 0.000 9.453 9.538
contract_S_to_Q 8 9.0 0.000 0.000 8.875 8.958
calc_potential_gpw 544 9.5 0.005 0.005 8.238 8.682
parallel_gemm_fm 14 9.1 0.000 0.000 8.469 8.558
parallel_gemm_fm_cosma 14 10.1 8.469 8.558 8.469 8.558
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.080 8.316
potential_pw2rs 545 10.0 0.107 0.110 7.494 8.145
create_integ_mat 1 6.0 0.014 0.028 7.749 7.759
collocate_single_gaussian 272 10.0 0.039 0.042 7.339 7.549
array2fm 1 7.0 0.000 0.000 6.737 7.155
pw_scatter_s 2720 13.7 4.353 4.523 4.353 4.523
pw_gather_s 2722 13.2 3.442 3.877 3.442 3.877
array2fm_buffer_send 1 8.0 2.999 3.161 2.999 3.161
pw_poisson_solve 545 10.5 1.115 1.181 2.195 2.322
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=111.534250, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2807.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 424 9.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.024 0.035 420.953 420.955
farming_run 1 2.0 420.340 420.343 420.917 420.920
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.239200E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 757 478553.
MP_Allreduce 2021 21391.
MP_Sync 37
MP_Alltoall 77 6314750.
MP_SendRecv 2876 2171486.
MP_ISendRecv 1034 172620.
MP_Wait 1346
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.031 217.029 217.030
qs_energies 1 2.0 0.000 0.000 216.806 216.812
scf_env_do_scf 1 3.0 0.000 0.000 114.846 114.846
qs_ks_update_qs_env 5 5.0 0.000 0.000 113.889 113.897
rebuild_ks_matrix 4 6.0 0.000 0.000 113.888 113.896
qs_ks_build_kohn_sham_matrix 4 7.0 0.054 0.062 113.888 113.896
hfx_ks_matrix 4 8.0 0.001 0.001 113.532 113.536
integrate_four_center 4 9.0 0.152 0.463 113.531 113.536
integrate_four_center_main 4 10.0 0.107 0.472 102.095 105.184
integrate_four_center_bin 266 11.0 101.988 104.790 101.988 104.790
mp2_main 1 3.0 0.000 0.000 101.652 101.658
mp2_gpw_main 1 4.0 0.047 0.066 100.755 100.766
init_scf_loop 1 4.0 0.000 0.000 96.997 96.998
mp2_ri_gpw_compute_in 1 5.0 0.063 0.064 73.918 75.072
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 53.874 55.027
mp2_eri_3c_integrate_gpw 91 7.0 0.143 0.158 41.345 46.543
integrate_v_rspace 95 8.0 0.395 0.561 27.807 32.823
pw_transfer 2240 10.6 0.146 0.173 29.472 29.942
fft_wrap_pw1pw2 1868 11.4 0.018 0.021 28.495 28.957
mp2_ri_gpw_compute_en 1 5.0 0.055 0.063 26.675 28.762
grid_integrate_task_list 95 9.0 23.216 28.468 23.216 28.468
fft_wrap_pw1pw2_100 730 12.4 1.260 1.444 26.259 26.706
ao_to_mo_and_store_B_mult_1 91 7.0 10.850 26.613 10.850 26.613
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.837 1.885 24.869 24.882
get_2c_integrals 1 6.0 0.000 0.000 19.956 19.979
fft3d_s 1823 13.4 18.469 19.000 18.483 19.013
compute_2c_integrals 1 7.0 0.002 0.003 18.941 18.947
compute_2c_integrals_loop_lm 1 8.0 0.001 0.001 18.543 18.814
mp2_eri_2c_integrate_gpw 1 9.0 1.743 1.883 18.541 18.812
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 17.847 17.847
calculate_wavefunction 91 8.0 2.026 2.057 9.635 9.824
mp2_ri_gpw_compute_en_expansio 172 7.0 0.558 0.586 8.753 9.383
potential_pw2rs 186 10.0 0.033 0.035 8.399 8.983
local_gemm 172 8.0 8.195 8.825 8.195 8.825
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.100 8.511
mp2_ri_gpw_compute_en_comm 22 7.0 0.500 0.523 7.890 8.429
calc_potential_gpw 182 9.5 0.002 0.002 7.886 8.149
collocate_single_gaussian 91 10.0 0.017 0.022 7.717 7.955
integrate_four_center_load 4 10.0 0.000 0.000 6.776 6.780
hfx_load_balance 1 11.0 0.000 0.000 6.776 6.780
mp_sendrecv_dm3 2068 8.0 5.918 6.477 5.918 6.477
mp2_ri_gpw_compute_en_ener 172 7.0 6.345 6.449 6.345 6.449
mp_sync 37 10.5 3.389 5.807 3.389 5.807
pw_gather_s 912 13.2 4.502 5.013 4.502 5.013
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=100.759238, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1499.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 452.689920E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62379.
MP_Allreduce 10329 270.
MP_Sync 530
MP_Alltoall 2083 592243.
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 0.006 0.026 53.686 53.688
qs_mol_dyn_low 1 2.0 0.003 0.003 53.454 53.463
qs_forces 11 3.9 0.006 0.007 53.391 53.393
qs_energies 11 4.9 0.001 0.002 51.892 51.918
scf_env_do_scf 11 5.9 0.000 0.001 45.785 45.785
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 43.694 43.694
dbcsr_multiply_generic 2286 12.5 0.095 0.099 33.689 34.232
qs_scf_new_mos 108 7.5 0.000 0.001 33.718 33.973
qs_scf_loop_do_ot 108 8.5 0.000 0.001 33.717 33.973
ot_scf_mini 108 9.5 0.002 0.002 32.004 32.188
multiply_cannon 2286 13.5 0.181 0.194 26.597 28.535
multiply_cannon_loop 2286 14.5 1.825 1.930 25.949 27.911
velocity_verlet 10 3.0 0.001 0.001 27.140 27.141
ot_mini 108 10.5 0.001 0.001 19.257 19.505
qs_ot_get_derivative 108 11.5 0.001 0.001 16.217 16.423
mp_waitall_1 245248 16.5 8.584 14.528 8.584 14.528
multiply_cannon_metrocomm3 54864 15.5 0.072 0.078 6.186 13.164
multiply_cannon_multrec 54864 15.5 3.642 5.726 7.792 11.224
qs_ot_get_p 119 10.4 0.001 0.001 8.003 8.265
rebuild_ks_matrix 119 8.3 0.000 0.000 7.829 7.966
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 7.828 7.966
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.897 7.013
mp_sum_l 7287 12.8 5.125 6.709 5.125 6.709
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 6.004 6.482
multiply_cannon_sync_h2d 54864 15.5 5.139 6.097 5.139 6.097
dbcsr_mm_accdrv_process 76910 16.1 1.841 2.907 4.064 5.793
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.067 5.107
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 4.793 4.891
init_scf_run 11 5.9 0.000 0.001 4.834 4.834
scf_env_initial_rho_setup 11 6.9 0.000 0.001 4.833 4.834
sum_up_and_integrate 119 10.3 0.001 0.001 4.435 4.442
integrate_v_rspace 119 11.3 0.002 0.003 4.424 4.433
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.332 4.333
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.091 4.199
calculate_rho_elec 119 8.7 0.011 0.017 4.091 4.198
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.141 4.141
cp_fm_redistribute_end 50 14.0 2.114 4.117 2.119 4.119
cp_fm_diag_elpa_base 50 14.0 1.995 4.015 1.998 4.022
calculate_dm_sparse 119 9.5 0.000 0.000 3.069 3.219
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.937 3.136
apply_single 119 13.6 0.000 0.000 2.937 3.136
multiply_cannon_metrocomm1 54864 15.5 0.056 0.062 1.771 2.897
acc_transpose_blocks 54864 15.5 0.230 0.258 2.310 2.871
jit_kernel_multiply 13 15.8 2.161 2.821 2.161 2.821
ot_diis_step 108 11.5 0.006 0.006 2.743 2.743
calculate_first_density_matrix 1 7.0 0.000 0.000 2.630 2.631
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.527 2.583
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.355 2.357
density_rs2pw 119 9.7 0.004 0.004 2.096 2.183
wfi_extrapolate 11 7.9 0.001 0.001 2.139 2.139
grid_integrate_task_list 119 12.3 2.036 2.123 2.036 2.123
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.058 2.117
init_scf_loop 11 6.9 0.000 0.000 2.073 2.073
mp_sum_d 4135 12.0 1.385 1.938 1.385 1.938
potential_pw2rs 119 12.3 0.004 0.004 1.796 1.808
pw_transfer 1439 11.6 0.051 0.056 1.632 1.700
make_m2s 4572 13.5 0.053 0.055 1.647 1.699
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 1.557 1.625
make_images 4572 14.5 0.133 0.138 1.565 1.616
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.528 1.568
mp_alltoall_d11v 2130 13.8 1.314 1.468 1.314 1.468
acc_transpose_blocks_sync 164592 16.5 1.215 1.465 1.215 1.465
transfer_rs2pw 487 10.6 0.005 0.006 1.334 1.425
grid_collocate_task_list 119 9.7 1.354 1.423 1.354 1.423
mp_waitany 12084 13.8 1.249 1.365 1.249 1.365
fft3d_ps 1201 14.6 0.368 0.472 1.271 1.333
transfer_pw2rs 487 13.2 0.006 0.006 1.302 1.310
fft_wrap_pw1pw2_140 487 13.2 0.141 0.156 1.205 1.275
dbcsr_dot_sd 1205 11.9 0.048 0.060 0.766 1.146
acc_transpose_blocks_kernels 54864 16.5 0.249 0.385 0.820 1.099
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=53.688000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=431.545455, yerr=0.782030
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 488.247296E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10306 303.
MP_Sync 54
MP_Alltoall 2060 1416841.
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 0.008 0.026 38.458 38.459
qs_mol_dyn_low 1 2.0 0.003 0.004 38.271 38.278
qs_forces 11 3.9 0.002 0.005 38.205 38.206
qs_energies 11 4.9 0.002 0.004 36.533 36.538
scf_env_do_scf 11 5.9 0.001 0.001 31.533 31.534
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 29.131 29.132
dbcsr_multiply_generic 2286 12.5 0.102 0.104 21.949 22.297
qs_scf_new_mos 108 7.5 0.001 0.001 20.577 20.803
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.576 20.803
ot_scf_mini 108 9.5 0.002 0.003 19.663 19.827
velocity_verlet 10 3.0 0.001 0.001 18.325 18.326
multiply_cannon 2286 13.5 0.207 0.217 16.808 18.284
multiply_cannon_loop 2286 14.5 1.191 1.247 15.659 17.262
ot_mini 108 10.5 0.001 0.001 12.150 12.380
mp_waitall_1 200699 16.5 5.464 10.697 5.464 10.697
qs_ot_get_derivative 108 11.5 0.001 0.001 9.775 9.939
multiply_cannon_metrocomm3 27432 15.5 0.071 0.073 3.988 9.469
multiply_cannon_multrec 27432 15.5 1.828 4.200 6.133 8.900
rebuild_ks_matrix 119 8.3 0.000 0.000 6.732 6.885
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.732 6.884
dbcsr_mm_accdrv_process 47894 16.0 3.465 5.602 4.228 6.291
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.917 6.054
qs_ot_get_p 119 10.4 0.001 0.001 4.592 4.816
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.643 4.554
mp_sum_l 7287 12.8 2.092 4.122 2.092 4.122
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.080 4.074
apply_single 119 13.6 0.000 0.000 3.080 4.074
init_scf_run 11 5.9 0.000 0.001 3.787 3.787
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.787 3.787
sum_up_and_integrate 119 10.3 0.001 0.001 3.659 3.665
integrate_v_rspace 119 11.3 0.002 0.003 3.645 3.652
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.593 3.627
calculate_rho_elec 119 8.7 0.021 0.024 3.593 3.626
qs_ot_p2m_diag 50 11.0 0.008 0.012 2.956 2.974
make_m2s 4572 13.5 0.052 0.053 2.594 2.921
make_images 4572 14.5 0.205 0.243 2.505 2.836
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.565 2.566
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.311 2.396
init_scf_loop 11 6.9 0.000 0.002 2.383 2.384
ot_diis_step 108 11.5 0.011 0.011 2.326 2.327
calculate_first_density_matrix 1 7.0 0.000 0.002 2.274 2.277
multiply_cannon_sync_h2d 27432 15.5 1.704 2.269 1.704 2.269
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.255 2.255
cp_fm_redistribute_end 50 14.0 1.143 2.231 1.146 2.234
cp_fm_diag_elpa_base 50 14.0 1.056 2.137 1.084 2.180
calculate_dm_sparse 119 9.5 0.000 0.001 2.052 2.134
density_rs2pw 119 9.7 0.004 0.004 1.892 1.962
jit_kernel_multiply 10 16.3 0.704 1.945 0.704 1.945
acc_transpose_blocks 27432 15.5 0.114 0.119 1.610 1.940
grid_integrate_task_list 119 12.3 1.840 1.918 1.840 1.918
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.865 1.867
pw_transfer 1439 11.6 0.065 0.069 1.824 1.856
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.732 1.767
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.668 1.708
make_images_data 4572 15.5 0.048 0.054 1.231 1.567
prepare_preconditioner 11 7.9 0.000 0.000 1.492 1.518
make_preconditioner 11 8.9 0.000 0.001 1.492 1.518
wfi_extrapolate 11 7.9 0.001 0.001 1.454 1.454
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.394 1.451
hybrid_alltoall_any 4725 16.4 0.053 0.115 1.090 1.425
fft_wrap_pw1pw2_140 487 13.2 0.160 0.168 1.380 1.413
potential_pw2rs 119 12.3 0.006 0.007 1.402 1.408
fft3d_ps 1201 14.6 0.517 0.572 1.369 1.399
mp_alltoall_d11v 2130 13.8 1.192 1.394 1.192 1.394
grid_collocate_task_list 119 9.7 1.290 1.334 1.290 1.334
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.245 1.255
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.194 1.238
transfer_rs2pw 487 10.6 0.005 0.006 1.087 1.164
mp_sum_d 4135 12.0 0.548 1.030 0.548 1.030
mp_allgather_i34 2286 14.5 0.574 0.972 0.574 0.972
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.950 0.951
acc_transpose_blocks_sync 82296 16.5 0.817 0.939 0.817 0.939
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.873 0.887
transfer_pw2rs 487 13.2 0.005 0.006 0.871 0.875
acc_transpose_blocks_kernels 27432 16.5 0.188 0.277 0.653 0.858
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=38.459000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.090909, yerr=1.378705
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 521.543680E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63490.
MP_Allreduce 10155 305.
MP_Sync 54
MP_Alltoall 1821 1607811.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.028 31.693 31.695
qs_mol_dyn_low 1 2.0 0.003 0.003 31.474 31.482
qs_forces 11 3.9 0.002 0.002 31.392 31.392
qs_energies 11 4.9 0.001 0.001 29.848 29.850
scf_env_do_scf 11 5.9 0.000 0.001 25.281 25.281
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 22.862 22.862
dbcsr_multiply_generic 2286 12.5 0.097 0.098 16.769 16.841
qs_scf_new_mos 108 7.5 0.001 0.001 15.252 15.266
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.251 15.265
velocity_verlet 10 3.0 0.001 0.001 15.082 15.084
ot_scf_mini 108 9.5 0.002 0.002 14.499 14.511
multiply_cannon 2286 13.5 0.193 0.197 13.544 14.268
multiply_cannon_loop 2286 14.5 0.856 0.906 12.779 13.516
ot_mini 108 10.5 0.001 0.001 8.980 8.993
qs_ot_get_derivative 108 11.5 0.001 0.001 7.503 7.515
multiply_cannon_multrec 18288 15.5 1.883 2.836 6.931 7.247
rebuild_ks_matrix 119 8.3 0.000 0.000 5.933 5.941
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.932 5.940
dbcsr_mm_accdrv_process 38222 16.0 4.892 5.684 4.956 5.750
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.226 5.234
mp_waitall_1 158411 16.6 2.547 3.733 2.547 3.733
sum_up_and_integrate 119 10.3 0.001 0.001 3.515 3.530
integrate_v_rspace 119 11.3 0.002 0.003 3.502 3.517
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.792 3.408
init_scf_run 11 5.9 0.000 0.001 3.386 3.386
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.385 3.386
qs_ot_get_p 119 10.4 0.001 0.001 3.296 3.321
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.231 3.235
calculate_rho_elec 119 8.7 0.031 0.031 3.230 3.234
multiply_cannon_metrocomm3 18288 15.5 0.047 0.049 1.441 2.425
init_scf_loop 11 6.9 0.000 0.000 2.403 2.404
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.020 2.330
apply_single 119 13.6 0.000 0.000 2.020 2.330
qs_ot_p2m_diag 50 11.0 0.012 0.012 2.201 2.206
calculate_first_density_matrix 1 7.0 0.000 0.000 2.163 2.164
make_m2s 4572 13.5 0.044 0.045 1.791 1.933
density_rs2pw 119 9.7 0.004 0.004 1.802 1.912
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.911 1.911
grid_integrate_task_list 119 12.3 1.796 1.909 1.796 1.909
make_images 4572 14.5 0.191 0.204 1.706 1.846
calculate_dm_sparse 119 9.5 0.000 0.001 1.835 1.846
pw_transfer 1439 11.6 0.065 0.069 1.802 1.811
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.709 1.720
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.685 1.694
acc_transpose_blocks 18288 15.5 0.080 0.082 1.613 1.693
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.667 1.676
cp_fm_diag_elpa_base 50 14.0 1.645 1.658 1.666 1.674
prepare_preconditioner 11 7.9 0.000 0.000 1.652 1.653
make_preconditioner 11 8.9 0.000 0.000 1.652 1.653
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.649 1.650
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.510 1.593
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.498 1.504
ot_diis_step 108 11.5 0.011 0.011 1.463 1.463
mp_sum_l 7287 12.8 1.037 1.413 1.037 1.413
fft_wrap_pw1pw2_140 487 13.2 0.212 0.217 1.387 1.397
potential_pw2rs 119 12.3 0.007 0.009 1.320 1.324
grid_collocate_task_list 119 9.7 1.235 1.302 1.235 1.302
fft3d_ps 1201 14.6 0.525 0.543 1.278 1.289
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.178 1.182
wfi_extrapolate 11 7.9 0.001 0.001 1.170 1.170
multiply_cannon_sync_h2d 18288 15.5 1.044 1.144 1.044 1.144
transfer_rs2pw 487 10.6 0.005 0.005 0.966 1.084
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.965 0.965
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.937 0.954
make_images_data 4572 15.5 0.048 0.052 0.766 0.921
hybrid_alltoall_any 4725 16.4 0.058 0.116 0.662 0.856
acc_transpose_blocks_sync 54864 16.5 0.736 0.815 0.736 0.815
transfer_pw2rs 487 13.2 0.004 0.004 0.808 0.811
acc_transpose_blocks_kernels 18288 16.5 0.217 0.225 0.778 0.797
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.775 0.777
mp_alltoall_d11v 2130 13.8 0.648 0.774 0.648 0.774
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.678 0.738
cp_fm_cholesky_invert 11 10.9 0.704 0.708 0.704 0.708
mp_alltoall_z22v 1201 16.6 0.606 0.678 0.606 0.678
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=31.695000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.272727, yerr=2.004128
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 560.091136E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63489.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.006 0.027 36.357 36.358
qs_mol_dyn_low 1 2.0 0.003 0.003 36.162 36.170
qs_forces 11 3.9 0.002 0.002 36.093 36.094
qs_energies 11 4.9 0.001 0.001 34.384 34.390
scf_env_do_scf 11 5.9 0.000 0.001 29.387 29.388
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 26.013 26.013
dbcsr_multiply_generic 2286 12.5 0.100 0.103 19.823 19.967
velocity_verlet 10 3.0 0.014 0.017 18.491 18.493
qs_scf_new_mos 108 7.5 0.001 0.001 17.906 17.961
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.906 17.960
ot_scf_mini 108 9.5 0.002 0.003 16.875 16.931
multiply_cannon 2286 13.5 0.220 0.229 15.918 16.392
multiply_cannon_loop 2286 14.5 1.497 1.588 14.957 15.385
ot_mini 108 10.5 0.001 0.001 10.355 10.419
multiply_cannon_multrec 27432 15.5 2.445 3.103 8.972 9.254
qs_ot_get_derivative 108 11.5 0.001 0.001 8.492 8.549
dbcsr_mm_accdrv_process 47916 15.9 5.869 7.651 6.427 7.736
rebuild_ks_matrix 119 8.3 0.000 0.000 6.329 6.380
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.329 6.380
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.605 5.650
qs_ot_get_p 119 10.4 0.001 0.001 3.662 3.742
init_scf_run 11 5.9 0.000 0.001 3.662 3.662
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.662 3.662
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.066 3.548
sum_up_and_integrate 119 10.3 0.001 0.001 3.504 3.513
integrate_v_rspace 119 11.3 0.003 0.003 3.493 3.502
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.380 3.417
calculate_rho_elec 119 8.7 0.040 0.046 3.380 3.416
init_scf_loop 11 6.9 0.000 0.000 3.355 3.355
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.199 2.645
apply_single 119 13.6 0.000 0.000 2.199 2.645
prepare_preconditioner 11 7.9 0.000 0.000 2.529 2.538
make_preconditioner 11 8.9 0.000 0.000 2.529 2.538
acc_transpose_blocks 27432 15.5 0.119 0.124 2.348 2.523
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.135 2.462
make_m2s 4572 13.5 0.054 0.056 2.332 2.446
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.353 2.383
make_images 4572 14.5 0.273 0.334 2.224 2.336
mp_waitall_1 137007 16.6 1.716 2.310 1.716 2.310
calculate_first_density_matrix 1 7.0 0.000 0.000 2.268 2.270
qs_ot_p2m_diag 50 11.0 0.015 0.022 2.257 2.269
calculate_dm_sparse 119 9.5 0.000 0.000 2.139 2.190
pw_transfer 1439 11.6 0.065 0.069 1.943 1.979
density_rs2pw 119 9.7 0.004 0.004 1.808 1.895
grid_integrate_task_list 119 12.3 1.831 1.893 1.831 1.893
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.851 1.890
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.868 1.869
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.857 1.858
ot_diis_step 108 11.5 0.012 0.012 1.821 1.822
mp_sum_l 7287 12.8 1.139 1.662 1.139 1.662
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.625 1.639
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.570 1.579
cp_fm_diag_elpa_base 50 14.0 1.538 1.555 1.569 1.578
fft_wrap_pw1pw2_140 487 13.2 0.245 0.259 1.525 1.566
acc_transpose_blocks_sync 82296 16.5 1.387 1.553 1.387 1.553
fft3d_ps 1201 14.6 0.553 0.601 1.371 1.395
wfi_extrapolate 11 7.9 0.001 0.001 1.342 1.342
grid_collocate_task_list 119 9.7 1.248 1.330 1.248 1.330
jit_kernel_multiply 7 16.0 0.493 1.315 0.493 1.315
potential_pw2rs 119 12.3 0.009 0.010 1.307 1.311
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.288 1.310
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.236 1.246
multiply_cannon_metrocomm3 27432 15.5 0.040 0.041 0.696 1.220
cp_fm_upper_to_full 72 14.2 0.813 1.157 0.813 1.157
dbcsr_complete_redistribute 329 12.2 0.197 0.224 0.841 1.116
qs_energies_init_hamiltonians 11 5.9 0.002 0.005 1.074 1.075
make_images_data 4572 15.5 0.048 0.052 0.871 1.002
transfer_rs2pw 487 10.6 0.004 0.005 0.900 0.992
hybrid_alltoall_any 4725 16.4 0.065 0.155 0.773 0.974
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.636 0.904
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.799 0.877
mp_alltoall_d11v 2130 13.8 0.803 0.872 0.803 0.872
acc_transpose_blocks_kernels 27432 16.5 0.269 0.276 0.814 0.828
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.814 0.819
transfer_pw2rs 487 13.2 0.004 0.004 0.747 0.751
mp_alltoall_i22 627 13.8 0.438 0.745 0.438 0.745
cp_fm_cholesky_invert 11 10.9 0.730 0.733 0.730 0.733
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=36.358000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=532.181818, yerr=3.270201
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 613.150720E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63488.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.027 27.994 27.996
qs_mol_dyn_low 1 2.0 0.003 0.003 27.834 27.842
qs_forces 11 3.9 0.002 0.002 27.775 27.776
qs_energies 11 4.9 0.001 0.001 26.079 26.081
scf_env_do_scf 11 5.9 0.000 0.001 21.460 21.460
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.065 19.065
velocity_verlet 10 3.0 0.001 0.001 14.454 14.457
dbcsr_multiply_generic 2286 12.5 0.094 0.099 12.835 12.925
qs_scf_new_mos 108 7.5 0.001 0.001 11.551 11.581
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.550 11.580
ot_scf_mini 108 9.5 0.002 0.002 10.846 10.872
multiply_cannon 2286 13.5 0.224 0.232 10.332 10.851
multiply_cannon_loop 2286 14.5 0.636 0.656 9.419 9.566
ot_mini 108 10.5 0.001 0.001 6.245 6.275
multiply_cannon_multrec 9144 15.5 1.655 1.958 5.910 6.184
rebuild_ks_matrix 119 8.3 0.000 0.000 5.689 5.710
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.689 5.710
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.060 5.079
qs_ot_get_derivative 108 11.5 0.001 0.001 4.970 4.997
dbcsr_mm_accdrv_process 12550 15.8 3.524 4.172 4.146 4.213
sum_up_and_integrate 119 10.3 0.001 0.001 3.386 3.391
integrate_v_rspace 119 11.3 0.003 0.003 3.376 3.381
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.271 3.277
calculate_rho_elec 119 8.7 0.060 0.061 3.271 3.276
init_scf_run 11 5.9 0.000 0.001 3.156 3.156
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.155 3.155
qs_ot_get_p 119 10.4 0.001 0.001 2.734 2.767
init_scf_loop 11 6.9 0.000 0.000 2.376 2.378
calculate_first_density_matrix 1 7.0 0.000 0.000 2.038 2.039
pw_transfer 1439 11.6 0.065 0.068 1.933 1.942
grid_integrate_task_list 119 12.3 1.878 1.931 1.878 1.931
make_m2s 4572 13.5 0.033 0.035 1.757 1.916
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.840 1.850
make_images 4572 14.5 0.269 0.298 1.668 1.825
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.781 1.782
calculate_dm_sparse 119 9.5 0.000 0.000 1.735 1.751
density_rs2pw 119 9.7 0.003 0.004 1.647 1.748
mp_waitall_1 115863 16.7 1.316 1.740 1.316 1.740
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.700 1.723
prepare_preconditioner 11 7.9 0.000 0.000 1.667 1.671
make_preconditioner 11 8.9 0.000 0.000 1.667 1.671
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.558 1.585
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.542 1.543
fft_wrap_pw1pw2_140 487 13.2 0.323 0.331 1.523 1.534
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.512 1.513
acc_transpose_blocks 9144 15.5 0.041 0.042 1.411 1.440
grid_collocate_task_list 119 9.7 1.296 1.392 1.296 1.392
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.355 1.364
jit_kernel_multiply 8 15.4 0.582 1.352 0.582 1.352
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.312 1.327
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.265 1.274
cp_fm_diag_elpa_base 50 14.0 1.239 1.257 1.264 1.272
ot_diis_step 108 11.5 0.013 0.013 1.264 1.264
fft3d_ps 1201 14.6 0.557 0.570 1.253 1.260
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.254 1.254
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.210 1.241
apply_single 119 13.6 0.000 0.000 1.210 1.240
potential_pw2rs 119 12.3 0.010 0.011 1.195 1.196
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.126 1.132
wfi_extrapolate 11 7.9 0.001 0.001 1.067 1.067
hybrid_alltoall_any 4725 16.4 0.065 0.176 0.758 0.989
make_images_data 4572 15.5 0.042 0.046 0.758 0.935
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.878 0.928
mp_alltoall_d11v 2130 13.8 0.776 0.838 0.776 0.838
cp_fm_cholesky_invert 11 10.9 0.827 0.830 0.827 0.830
transfer_rs2pw 487 10.6 0.004 0.005 0.741 0.828
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.802 0.809
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.750 0.753
qs_env_update_s_mstruct 11 6.9 0.000 0.001 0.701 0.750
multiply_cannon_metrocomm3 9144 15.5 0.020 0.021 0.366 0.747
acc_transpose_blocks_sync 27432 16.5 0.714 0.742 0.714 0.742
mp_allgather_i34 2286 14.5 0.231 0.653 0.231 0.653
acc_transpose_blocks_kernels 9144 16.5 0.117 0.120 0.640 0.642
transfer_pw2rs 487 13.2 0.003 0.004 0.636 0.637
mp_alltoall_z22v 1201 16.6 0.570 0.604 0.570 0.604
qs_create_task_list 11 7.9 0.000 0.001 0.539 0.564
generate_qs_task_list 11 8.9 0.188 0.210 0.539 0.564
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=27.996000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=583.818182, yerr=3.432947
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 772.390912E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63723.
MP_Allreduce 10154 429.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.030 42.952 42.952
qs_mol_dyn_low 1 2.0 0.003 0.003 42.714 42.722
qs_forces 11 3.9 0.002 0.002 42.614 42.616
qs_energies 11 4.9 0.001 0.001 40.618 40.622
scf_env_do_scf 11 5.9 0.001 0.001 34.972 34.972
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 27.102 27.104
velocity_verlet 10 3.0 0.035 0.036 24.181 24.188
dbcsr_multiply_generic 2286 12.5 0.101 0.104 19.186 19.299
qs_scf_new_mos 108 7.5 0.001 0.001 17.648 17.742
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.647 17.741
ot_scf_mini 108 9.5 0.002 0.002 16.486 16.583
multiply_cannon 2286 13.5 0.305 0.314 15.137 15.918
multiply_cannon_loop 2286 14.5 0.866 0.890 13.845 14.662
ot_mini 108 10.5 0.001 0.001 10.132 10.243
multiply_cannon_multrec 9144 15.5 3.454 4.916 8.812 8.942
qs_ot_get_derivative 108 11.5 0.001 0.001 8.048 8.144
init_scf_loop 11 6.9 0.000 0.000 7.842 7.845
rebuild_ks_matrix 119 8.3 0.000 0.001 7.141 7.278
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.141 7.278
prepare_preconditioner 11 7.9 0.000 0.000 6.882 6.896
make_preconditioner 11 8.9 0.000 0.000 6.882 6.895
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.432 6.772
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.434 6.559
dbcsr_mm_accdrv_process 12550 15.8 4.481 5.992 5.224 6.524
cp_fm_upper_to_full 72 14.2 3.164 4.555 3.164 4.555
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.067 4.075
calculate_rho_elec 119 8.7 0.118 0.121 4.067 4.074
sum_up_and_integrate 119 10.3 0.001 0.001 3.819 3.825
integrate_v_rspace 119 11.3 0.003 0.004 3.808 3.815
init_scf_run 11 5.9 0.000 0.001 3.590 3.590
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.590 3.590
qs_ot_get_p 119 10.4 0.001 0.001 3.389 3.522
mp_waitall_1 94719 16.7 2.368 3.275 2.368 3.275
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.802 3.227
dbcsr_complete_redistribute 329 12.2 0.292 0.297 2.009 2.828
pw_transfer 1439 11.6 0.068 0.069 2.598 2.603
make_m2s 4572 13.5 0.037 0.038 2.413 2.589
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.255 2.547
apply_single 119 13.6 0.000 0.000 2.255 2.547
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.500 2.505
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.684 2.488
make_images 4572 14.5 0.351 0.383 2.293 2.469
mp_alltoall_i22 627 13.8 1.422 2.254 1.422 2.254
calculate_dm_sparse 119 9.5 0.000 0.000 2.235 2.252
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.445 2.243
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.387 2.195
grid_integrate_task_list 119 12.3 2.108 2.149 2.108 2.149
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.084 2.134
fft_wrap_pw1pw2_140 487 13.2 0.574 0.579 2.103 2.111
density_rs2pw 119 9.7 0.003 0.004 2.066 2.108
calculate_first_density_matrix 1 7.0 0.000 0.000 2.102 2.103
ot_diis_step 108 11.5 0.014 0.015 2.060 2.060
qs_ot_p2m_diag 50 11.0 0.042 0.043 1.993 1.994
acc_transpose_blocks 9144 15.5 0.044 0.045 1.813 1.853
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.820 1.821
qs_energies_init_hamiltonians 11 5.9 0.003 0.004 1.776 1.777
mp_sum_l 7287 12.8 1.091 1.754 1.091 1.754
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.635 1.636
fft3d_ps 1201 14.6 0.594 0.607 1.598 1.602
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.557 1.597
grid_collocate_task_list 119 9.7 1.542 1.592 1.542 1.592
cp_fm_cholesky_invert 11 10.9 1.467 1.470 1.467 1.470
wfi_extrapolate 11 7.9 0.001 0.001 1.426 1.426
hybrid_alltoall_any 4725 16.4 0.089 0.148 1.105 1.375
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.326 1.345
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.342 1.342
cp_fm_diag_elpa_base 50 14.0 1.198 1.250 1.341 1.341
potential_pw2rs 119 12.3 0.014 0.015 1.281 1.282
make_images_data 4572 15.5 0.046 0.049 1.044 1.273
mp_alltoall_d11v 2130 13.8 1.211 1.266 1.211 1.266
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.237 1.245
acc_transpose_blocks_sync 27432 16.5 1.109 1.146 1.109 1.146
qs_env_update_s_mstruct 11 6.9 0.001 0.002 1.103 1.124
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 0.995 1.025
qs_create_task_list 11 7.9 0.000 0.001 0.928 0.940
generate_qs_task_list 11 8.9 0.367 0.387 0.927 0.940
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.917 0.930
mp_alltoall_z22v 1201 16.6 0.869 0.885 0.869 0.885
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=42.952000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=722.636364, yerr=15.040991
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/09/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 198.287135E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 8410880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 117.0
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 501.710848E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66212.
MP_Allreduce 9776 488.
MP_Sync 52
MP_Alltoall 1938 2123256.
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 0.006 0.025 83.491 83.492
qs_mol_dyn_low 1 2.0 0.003 0.003 83.223 83.236
qs_forces 11 3.9 0.003 0.003 83.149 83.151
qs_energies 11 4.9 0.001 0.002 80.236 80.257
scf_env_do_scf 11 5.9 0.000 0.001 71.376 71.380
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 65.692 65.693
dbcsr_multiply_generic 2055 12.4 0.110 0.111 51.477 51.753
qs_scf_new_mos 99 7.5 0.001 0.001 48.188 48.319
qs_scf_loop_do_ot 99 8.5 0.000 0.001 48.188 48.319
ot_scf_mini 99 9.5 0.002 0.002 45.795 45.895
velocity_verlet 10 3.0 0.001 0.002 44.406 44.407
multiply_cannon 2055 13.4 0.178 0.185 42.779 43.889
multiply_cannon_loop 2055 14.4 1.794 1.839 41.748 42.915
ot_mini 99 10.5 0.001 0.001 27.038 27.163
qs_ot_get_derivative 99 11.5 0.001 0.001 20.165 20.333
multiply_cannon_multrec 49320 15.4 11.351 12.011 17.344 17.958
rebuild_ks_matrix 110 8.3 0.000 0.000 14.825 14.963
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.824 14.963
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.012 13.134
mp_waitall_1 220248 16.4 11.211 12.100 11.211 12.100
qs_ot_get_p 110 10.4 0.001 0.001 10.134 10.252
multiply_cannon_sync_h2d 49320 15.4 9.550 10.156 9.550 10.156
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.890 8.424
multiply_cannon_metrocomm3 49320 15.4 0.082 0.085 6.705 7.850
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.227 7.780
apply_single 110 13.6 0.000 0.000 7.227 7.780
sum_up_and_integrate 110 10.3 0.002 0.003 7.215 7.228
integrate_v_rspace 110 11.3 0.003 0.003 7.190 7.209
qs_ot_p2m_diag 48 11.0 0.012 0.018 6.859 6.885
init_scf_run 11 5.9 0.000 0.001 6.788 6.788
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.788 6.788
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.569 6.696
calculate_rho_elec 110 8.6 0.021 0.025 6.569 6.696
ot_diis_step 99 11.5 0.006 0.006 6.627 6.627
dbcsr_mm_accdrv_process 87628 16.1 3.055 3.133 5.864 6.122
cp_dbcsr_syevd 48 12.0 0.002 0.003 6.030 6.030
init_scf_loop 11 6.9 0.000 0.000 5.658 5.658
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.417 5.449
cp_fm_diag_elpa_base 48 14.0 5.402 5.436 5.415 5.447
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.332 5.388
mp_sum_l 6594 12.7 4.222 4.983 4.222 4.983
make_m2s 4110 13.4 0.060 0.065 4.009 4.167
wfi_extrapolate 11 7.9 0.001 0.001 4.112 4.112
make_images 4110 14.4 0.177 0.190 3.913 4.075
calculate_dm_sparse 110 9.5 0.001 0.001 3.755 3.853
density_rs2pw 110 9.6 0.004 0.005 3.597 3.763
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.606 3.610
multiply_cannon_metrocomm1 49320 15.4 0.064 0.066 2.480 3.556
grid_integrate_task_list 110 12.3 3.255 3.431 3.255 3.431
prepare_preconditioner 11 7.9 0.000 0.000 3.394 3.416
make_preconditioner 11 8.9 0.000 0.000 3.394 3.416
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.329 3.385
pw_transfer 1331 11.6 0.055 0.065 3.285 3.382
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.196 3.297
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.222 3.261
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.174 3.224
fft_wrap_pw1pw2_140 451 13.1 0.389 0.428 2.707 2.822
potential_pw2rs 110 12.3 0.005 0.006 2.786 2.818
acc_transpose_blocks 49320 15.4 0.214 0.222 2.559 2.612
calculate_first_density_matrix 1 7.0 0.000 0.000 2.572 2.581
fft3d_ps 1111 14.6 0.793 0.881 2.491 2.562
jit_kernel_multiply 13 15.9 2.531 2.548 2.531 2.548
mp_alltoall_d11v 2046 13.8 2.140 2.537 2.140 2.537
grid_collocate_task_list 110 9.6 2.159 2.274 2.159 2.274
transfer_rs2pw 451 10.6 0.005 0.006 2.040 2.199
mp_waitany 14300 13.8 1.838 2.077 1.838 2.077
mp_sum_d 3889 11.9 1.483 1.963 1.483 1.963
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.943 1.961
make_images_data 4110 15.4 0.044 0.047 1.810 1.952
transfer_pw2rs 451 13.1 0.006 0.007 1.895 1.920
cp_fm_cholesky_invert 11 10.9 1.874 1.878 1.874 1.878
hybrid_alltoall_any 4261 16.3 0.085 0.487 1.567 1.835
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.649 1.678
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=83.492000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=476.181818, yerr=3.069444
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 591.958016E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66430.
MP_Allreduce 9775 566.
MP_Sync 52
MP_Alltoall 1717 3451230.
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 0.006 0.026 67.946 67.947
qs_mol_dyn_low 1 2.0 0.003 0.003 67.723 67.733
qs_forces 11 3.9 0.003 0.003 67.657 67.658
qs_energies 11 4.9 0.001 0.002 64.354 64.358
scf_env_do_scf 11 5.9 0.000 0.001 55.919 55.923
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 48.500 48.500
dbcsr_multiply_generic 2055 12.4 0.118 0.123 37.770 37.947
velocity_verlet 10 3.0 0.001 0.001 35.714 35.715
qs_scf_new_mos 99 7.5 0.001 0.001 32.817 32.981
qs_scf_loop_do_ot 99 8.5 0.001 0.001 32.817 32.980
multiply_cannon 2055 13.4 0.223 0.243 31.276 32.368
ot_scf_mini 99 9.5 0.003 0.003 31.151 31.299
multiply_cannon_loop 2055 14.4 1.163 1.186 30.028 30.871
ot_mini 99 10.5 0.001 0.001 18.244 18.397
multiply_cannon_multrec 24660 15.4 7.024 9.052 14.035 15.984
rebuild_ks_matrix 110 8.3 0.000 0.000 13.428 13.593
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 13.428 13.593
qs_ot_get_derivative 99 11.5 0.001 0.001 12.489 12.635
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.812 11.955
mp_waitall_1 176588 16.5 7.400 9.895 7.400 9.895
multiply_cannon_metrocomm3 24660 15.4 0.072 0.075 5.097 7.771
multiply_cannon_sync_h2d 24660 15.4 6.390 7.686 6.390 7.686
init_scf_loop 11 6.9 0.000 0.000 7.385 7.386
dbcsr_mm_accdrv_process 52282 16.1 5.541 6.154 6.842 7.142
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.452 7.090
apply_single 110 13.6 0.000 0.001 6.452 7.090
qs_ot_get_p 110 10.4 0.001 0.001 6.267 6.471
sum_up_and_integrate 110 10.3 0.001 0.002 6.292 6.307
integrate_v_rspace 110 11.3 0.002 0.003 6.265 6.278
init_scf_run 11 5.9 0.000 0.001 6.079 6.080
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.079 6.079
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.795 5.806
calculate_rho_elec 110 8.6 0.039 0.047 5.794 5.805
ot_diis_step 99 11.5 0.010 0.010 5.707 5.707
prepare_preconditioner 11 7.9 0.000 0.000 5.396 5.419
make_preconditioner 11 8.9 0.000 0.000 5.396 5.419
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.693 5.419
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.961 5.114
make_m2s 4110 13.4 0.057 0.059 4.129 4.605
make_images 4110 14.4 0.408 0.469 4.020 4.493
qs_ot_p2m_diag 48 11.0 0.028 0.043 4.408 4.427
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.950 3.950
pw_transfer 1331 11.6 0.066 0.070 3.467 3.610
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.360 3.501
wfi_extrapolate 11 7.9 0.001 0.001 3.500 3.500
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.400 3.412
cp_fm_diag_elpa_base 48 14.0 3.354 3.377 3.397 3.409
grid_integrate_task_list 110 12.3 3.174 3.353 3.174 3.353
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.252 3.333
density_rs2pw 110 9.6 0.004 0.005 3.042 3.315
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.246 3.248
fft_wrap_pw1pw2_140 451 13.1 0.460 0.474 2.900 3.045
calculate_dm_sparse 110 9.5 0.001 0.001 2.997 3.024
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.901 2.965
hybrid_alltoall_any 4261 16.3 0.106 0.447 1.973 2.764
make_images_data 4110 15.4 0.049 0.053 2.251 2.743
fft3d_ps 1111 14.6 1.110 1.337 2.457 2.608
cp_fm_cholesky_invert 11 10.9 2.488 2.495 2.488 2.495
calculate_first_density_matrix 1 7.0 0.000 0.000 2.483 2.486
mp_sum_l 6594 12.7 1.727 2.367 1.727 2.367
grid_collocate_task_list 110 9.6 2.171 2.313 2.171 2.313
potential_pw2rs 110 12.3 0.009 0.009 2.175 2.183
acc_transpose_blocks 24660 15.4 0.116 0.120 1.980 2.010
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.951 1.972
mp_alltoall_d11v 2046 13.8 1.728 1.899 1.728 1.899
jit_kernel_multiply 10 16.3 0.936 1.888 0.936 1.888
qs_energies_init_hamiltonians 11 5.9 0.000 0.000 1.850 1.850
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.771 1.789
transfer_rs2pw 451 10.6 0.006 0.007 1.381 1.734
multiply_cannon_metrocomm4 22605 15.4 0.079 0.085 0.773 1.692
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.567 1.577
mp_irecv_dv 57340 16.2 0.642 1.569 0.642 1.569
mp_allgather_i34 2055 14.4 0.557 1.564 0.557 1.564
mp_waitany 10164 13.8 1.156 1.528 1.156 1.528
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.389 1.490
dbcsr_complete_redistribute 325 12.2 0.242 0.334 1.121 1.385
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=67.947000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=560.181818, yerr=6.846607
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 665.538560E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66421.
MP_Allreduce 9774 562.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.028 59.219 59.220
qs_mol_dyn_low 1 2.0 0.003 0.005 58.959 58.969
qs_forces 11 3.9 0.003 0.003 58.892 58.893
qs_energies 11 4.9 0.002 0.006 55.741 55.745
scf_env_do_scf 11 5.9 0.001 0.001 47.823 47.823
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 39.330 39.332
velocity_verlet 10 3.0 0.001 0.002 32.194 32.195
dbcsr_multiply_generic 2055 12.4 0.109 0.114 28.522 28.814
qs_scf_new_mos 99 7.5 0.001 0.001 24.862 24.980
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.861 24.979
ot_scf_mini 99 9.5 0.002 0.003 23.619 23.759
multiply_cannon 2055 13.4 0.216 0.226 22.357 23.383
multiply_cannon_loop 2055 14.4 0.814 0.844 21.312 22.378
ot_mini 99 10.5 0.001 0.001 13.621 13.759
rebuild_ks_matrix 110 8.3 0.000 0.000 12.049 12.217
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.032 12.048 12.216
multiply_cannon_multrec 16440 15.4 3.606 4.698 9.803 10.850
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.633 10.788
mp_waitall_1 139946 16.5 6.905 10.174 6.905 10.174
qs_ot_get_derivative 99 11.5 0.001 0.001 9.175 9.317
init_scf_loop 11 6.9 0.004 0.033 8.458 8.460
multiply_cannon_metrocomm3 16440 15.4 0.046 0.048 4.407 7.562
prepare_preconditioner 11 7.9 0.000 0.000 6.644 6.662
make_preconditioner 11 8.9 0.000 0.001 6.644 6.662
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.988 6.340
dbcsr_mm_accdrv_process 34862 16.1 5.324 5.767 6.044 6.242
sum_up_and_integrate 110 10.3 0.002 0.003 6.220 6.235
integrate_v_rspace 110 11.3 0.003 0.003 6.194 6.209
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.666 5.676
calculate_rho_elec 110 8.6 0.059 0.059 5.665 5.676
init_scf_run 11 5.9 0.000 0.001 5.465 5.466
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.465 5.465
qs_ot_get_p 110 10.4 0.001 0.001 5.204 5.386
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.901 5.358
apply_single 110 13.6 0.000 0.000 4.901 5.358
ot_diis_step 99 11.5 0.014 0.040 4.416 4.417
make_m2s 4110 13.4 0.049 0.051 4.090 4.363
make_images 4110 14.4 0.399 0.521 3.975 4.248
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.151 3.811
multiply_cannon_sync_h2d 16440 15.4 3.252 3.771 3.252 3.771
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.595 3.599
pw_transfer 1331 11.6 0.066 0.073 3.509 3.520
grid_integrate_task_list 110 12.3 3.206 3.462 3.206 3.462
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.402 3.416
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.228 3.228
density_rs2pw 110 9.6 0.004 0.005 2.909 3.162
fft_wrap_pw1pw2_140 451 13.1 0.579 0.586 2.955 2.968
wfi_extrapolate 11 7.9 0.001 0.003 2.930 2.930
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.805 2.807
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.708 2.718
cp_fm_diag_elpa_base 48 14.0 2.644 2.672 2.707 2.717
make_images_data 4110 15.4 0.046 0.051 2.295 2.649
hybrid_alltoall_any 4261 16.3 0.109 0.381 2.082 2.633
cp_fm_cholesky_invert 11 10.9 2.555 2.561 2.555 2.561
calculate_dm_sparse 110 9.5 0.001 0.001 2.506 2.536
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.478 2.530
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.406 2.485
multiply_cannon_metrocomm4 14385 15.4 0.049 0.054 0.879 2.459
calculate_first_density_matrix 1 7.0 0.000 0.002 2.446 2.447
grid_collocate_task_list 110 9.6 2.214 2.398 2.214 2.398
fft3d_ps 1111 14.6 1.089 1.096 2.345 2.360
mp_irecv_dv 48980 15.7 0.802 2.324 0.802 2.324
potential_pw2rs 110 12.3 0.011 0.012 2.091 2.096
mp_sum_l 6594 12.7 1.449 2.049 1.449 2.049
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 2.020 2.021
mp_alltoall_d11v 2046 13.8 1.754 1.978 1.754 1.978
dbcsr_complete_redistribute 325 12.2 0.328 0.354 1.408 1.869
cp_fm_upper_to_full 70 14.2 1.402 1.771 1.402 1.771
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.634 1.650
acc_transpose_blocks 16440 15.4 0.077 0.079 1.568 1.611
cp_fm_cholesky_decompose 22 10.9 1.542 1.563 1.542 1.563
transfer_rs2pw 451 10.6 0.005 0.006 1.253 1.501
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.372 1.469
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.448 1.462
mp_waitany 17072 13.8 1.133 1.422 1.133 1.422
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.967 1.417
mp_allgather_i34 2055 14.4 0.355 1.384 0.355 1.384
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.275 1.282
qs_env_update_s_mstruct 11 6.9 0.012 0.025 1.109 1.198
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=59.220000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=630.545455, yerr=8.489176
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 723.087360E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66419.
MP_Allreduce 9774 603.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.008 0.029 64.907 64.908
qs_mol_dyn_low 1 2.0 0.003 0.003 64.685 64.697
qs_forces 11 3.9 0.003 0.003 64.613 64.614
qs_energies 11 4.9 0.001 0.002 61.236 61.242
scf_env_do_scf 11 5.9 0.000 0.001 53.072 53.075
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 41.634 41.635
velocity_verlet 10 3.0 0.001 0.001 36.909 36.911
dbcsr_multiply_generic 2055 12.4 0.116 0.122 30.413 30.630
qs_scf_new_mos 99 7.5 0.001 0.001 27.230 27.315
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.230 27.315
ot_scf_mini 99 9.5 0.002 0.003 25.546 25.635
multiply_cannon 2055 13.4 0.237 0.253 23.402 24.848
multiply_cannon_loop 2055 14.4 1.391 1.480 22.060 22.704
ot_mini 99 10.5 0.001 0.001 14.732 14.852
multiply_cannon_multrec 24660 15.4 4.079 6.759 12.990 13.985
rebuild_ks_matrix 110 8.3 0.000 0.000 11.809 11.934
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.809 11.934
init_scf_loop 11 6.9 0.000 0.000 11.399 11.399
qs_ot_get_derivative 99 11.5 0.001 0.001 10.541 10.633
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.425 10.537
dbcsr_mm_accdrv_process 52304 16.0 7.765 9.236 8.756 9.728
prepare_preconditioner 11 7.9 0.000 0.000 9.686 9.702
make_preconditioner 11 8.9 0.000 0.000 9.686 9.702
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.954 9.389
sum_up_and_integrate 110 10.3 0.001 0.002 6.160 6.173
integrate_v_rspace 110 11.3 0.003 0.003 6.134 6.146
mp_waitall_1 121746 16.5 4.133 6.016 4.133 6.016
qs_ot_get_p 110 10.4 0.001 0.001 5.734 5.873
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.755 5.765
calculate_rho_elec 110 8.6 0.078 0.081 5.754 5.764
make_m2s 4110 13.4 0.059 0.061 5.369 5.630
make_images 4110 14.4 0.577 0.700 5.229 5.486
init_scf_run 11 5.9 0.000 0.001 5.450 5.450
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.450 5.450
cp_fm_upper_to_full 70 14.2 3.329 4.811 3.329 4.811
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.063 4.176
apply_single 110 13.6 0.000 0.000 4.063 4.176
ot_diis_step 99 11.5 0.011 0.011 4.151 4.151
qs_ot_p2m_diag 48 11.0 0.054 0.063 3.890 3.910
dbcsr_complete_redistribute 325 12.2 0.421 0.471 2.588 3.703
pw_transfer 1331 11.6 0.065 0.073 3.597 3.631
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.490 3.526
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.484 3.522
grid_integrate_task_list 110 12.3 3.289 3.423 3.289 3.423
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.338 3.399
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.328 3.328
multiply_cannon_metrocomm3 24660 15.4 0.038 0.039 1.397 3.239
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.077 3.179
fft_wrap_pw1pw2_140 451 13.1 0.610 0.632 3.011 3.051
density_rs2pw 110 9.6 0.004 0.004 2.876 3.020
calculate_dm_sparse 110 9.5 0.001 0.001 2.982 3.017
wfi_extrapolate 11 7.9 0.001 0.001 2.992 2.992
make_images_data 4110 15.4 0.049 0.053 2.645 2.945
hybrid_alltoall_any 4261 16.3 0.122 0.457 2.286 2.902
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.724 2.805
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.796 2.805
cp_fm_diag_elpa_base 48 14.0 2.651 2.708 2.793 2.803
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.791 2.793
mp_alltoall_i22 605 13.7 1.631 2.790 1.631 2.790
cp_fm_cholesky_invert 11 10.9 2.589 2.596 2.589 2.596
acc_transpose_blocks 24660 15.4 0.113 0.117 2.387 2.564
multiply_cannon_sync_h2d 24660 15.4 2.364 2.545 2.364 2.545
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.436 2.476
grid_collocate_task_list 110 9.6 2.265 2.427 2.265 2.427
fft3d_ps 1111 14.6 1.083 1.115 2.372 2.388
calculate_first_density_matrix 1 7.0 0.000 0.001 2.364 2.367
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 2.220 2.221
potential_pw2rs 110 12.3 0.013 0.014 1.986 1.989
mp_alltoall_d11v 2046 13.8 1.727 1.890 1.727 1.890
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.794 1.837
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.612 1.714
cp_fm_cholesky_decompose 22 10.9 1.619 1.656 1.619 1.656
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.624 1.635
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.544 1.562
acc_transpose_blocks_sync 73980 16.4 1.371 1.535 1.371 1.535
mp_allgather_i34 2055 14.4 0.426 1.512 0.426 1.512
mp_sum_l 6594 12.7 0.951 1.475 0.951 1.475
multiply_cannon_metrocomm4 20550 15.4 0.063 0.066 0.866 1.408
mp_irecv_dv 62702 16.1 0.759 1.326 0.759 1.326
transfer_rs2pw 451 10.6 0.005 0.005 1.150 1.306
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=64.908000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=687.909091, yerr=5.727994
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 853.258240E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66417.
MP_Allreduce 9774 644.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.028 54.591 54.592
qs_mol_dyn_low 1 2.0 0.003 0.003 54.349 54.380
qs_forces 11 3.9 0.003 0.003 54.279 54.279
qs_energies 11 4.9 0.046 0.050 50.641 50.646
scf_env_do_scf 11 5.9 0.000 0.001 42.330 42.330
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 34.732 34.733
velocity_verlet 10 3.0 0.001 0.001 30.799 30.803
dbcsr_multiply_generic 2055 12.4 0.105 0.109 23.100 23.284
qs_scf_new_mos 99 7.5 0.001 0.001 20.378 20.423
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.377 20.423
ot_scf_mini 99 9.5 0.002 0.002 19.134 19.153
multiply_cannon 2055 13.4 0.238 0.247 17.597 19.063
multiply_cannon_loop 2055 14.4 0.597 0.616 16.335 16.731
rebuild_ks_matrix 110 8.3 0.000 0.000 11.501 11.517
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.500 11.516
ot_mini 99 10.5 0.001 0.001 10.644 10.657
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.228 10.242
multiply_cannon_multrec 8220 15.4 3.172 4.316 7.572 8.423
init_scf_loop 11 6.9 0.000 0.000 7.552 7.553
mp_waitall_1 103326 16.6 5.812 7.369 5.812 7.369
qs_ot_get_derivative 99 11.5 0.001 0.001 6.970 6.988
sum_up_and_integrate 110 10.3 0.001 0.002 6.136 6.150
integrate_v_rspace 110 11.3 0.003 0.003 6.109 6.123
prepare_preconditioner 11 7.9 0.000 0.000 5.928 5.934
make_preconditioner 11 8.9 0.000 0.000 5.928 5.934
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.870 5.881
calculate_rho_elec 110 8.6 0.114 0.114 5.870 5.881
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.500 5.576
dbcsr_mm_accdrv_process 17442 15.9 3.067 3.951 4.262 5.137
init_scf_run 11 5.9 0.000 0.001 5.115 5.115
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.115 5.115
qs_ot_get_p 110 10.4 0.001 0.001 4.563 4.579
make_m2s 4110 13.4 0.038 0.039 4.173 4.465
multiply_cannon_metrocomm3 8220 15.4 0.019 0.019 3.022 4.422
make_images 4110 14.4 0.648 0.709 4.043 4.335
pw_transfer 1331 11.6 0.066 0.073 3.775 3.793
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.715 3.771
apply_single 110 13.6 0.000 0.000 3.715 3.770
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.667 3.687
ot_diis_step 99 11.5 0.012 0.012 3.652 3.652
grid_integrate_task_list 110 12.3 3.372 3.474 3.372 3.474
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.231 3.236
fft_wrap_pw1pw2_140 451 13.1 0.778 0.788 3.180 3.206
density_rs2pw 110 9.6 0.004 0.004 2.865 3.008
cp_dbcsr_syevd 48 12.0 0.003 0.003 2.892 2.893
cp_fm_cholesky_invert 11 10.9 2.798 2.801 2.798 2.801
hybrid_alltoall_any 4261 16.3 0.200 0.846 2.262 2.784
wfi_extrapolate 11 7.9 0.001 0.001 2.688 2.689
qs_energies_init_hamiltonians 11 5.9 0.011 0.016 2.682 2.683
make_images_data 4110 15.4 0.041 0.047 2.242 2.623
grid_collocate_task_list 110 9.6 2.368 2.572 2.368 2.572
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.559 2.560
calculate_dm_sparse 110 9.5 0.001 0.001 2.461 2.504
multiply_cannon_sync_h2d 8220 15.4 2.382 2.467 2.382 2.467
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.344 2.352
cp_fm_diag_elpa_base 48 14.0 2.288 2.310 2.343 2.350
fft3d_ps 1111 14.6 1.133 1.164 2.333 2.345
calculate_first_density_matrix 1 7.0 0.000 0.000 2.332 2.333
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.197 2.223
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.788 2.011
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 1.985 2.008
potential_pw2rs 110 12.3 0.015 0.016 1.927 1.930
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.886 1.894
mp_alltoall_d11v 2046 13.8 1.698 1.850 1.698 1.850
cp_fm_cholesky_decompose 22 10.9 1.666 1.691 1.666 1.691
qs_env_update_s_mstruct 11 6.9 0.000 0.001 1.492 1.604
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.598 1.601
mp_allgather_i34 2055 14.4 0.433 1.536 0.433 1.536
dbcsr_complete_redistribute 325 12.2 0.570 0.594 1.430 1.533
jit_kernel_multiply 9 15.6 0.883 1.521 0.883 1.521
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.420 1.433
acc_transpose_blocks 8220 15.4 0.039 0.040 1.344 1.379
qs_create_task_list 11 7.9 0.000 0.001 1.209 1.307
generate_qs_task_list 11 8.9 0.373 0.441 1.208 1.307
transfer_rs2pw 451 10.6 0.005 0.005 1.094 1.268
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.195 1.234
mp_waitany 9240 13.8 1.045 1.209 1.045 1.209
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=54.592000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=802.818182, yerr=9.232533
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 1.429819E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67098.
MP_Allreduce 9752 812.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.032 86.599 86.599
qs_mol_dyn_low 1 2.0 0.003 0.003 86.299 86.309
qs_forces 11 3.9 0.003 0.003 86.229 86.230
qs_energies 11 4.9 0.002 0.002 82.130 82.131
scf_env_do_scf 11 5.9 0.000 0.001 72.156 72.156
velocity_verlet 10 3.0 0.002 0.002 55.480 55.487
scf_env_do_scf_inner_loop 99 6.5 0.020 0.023 43.860 43.862
dbcsr_multiply_generic 2055 12.4 0.120 0.126 29.462 29.567
init_scf_loop 11 6.9 0.000 0.000 28.224 28.228
qs_scf_new_mos 99 7.5 0.001 0.001 26.706 26.749
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.705 26.748
prepare_preconditioner 11 7.9 0.000 0.000 26.238 26.247
make_preconditioner 11 8.9 0.000 0.000 26.238 26.247
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.357 25.725
ot_scf_mini 99 9.5 0.002 0.002 24.901 24.940
multiply_cannon 2055 13.4 0.332 0.354 22.489 23.118
multiply_cannon_loop 2055 14.4 0.820 0.849 20.725 21.012
cp_fm_upper_to_full 70 14.2 12.709 18.328 12.709 18.328
ot_mini 99 10.5 0.001 0.001 14.097 14.136
rebuild_ks_matrix 110 8.3 0.001 0.001 13.482 13.536
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 13.482 13.536
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.225 12.273
dbcsr_complete_redistribute 325 12.2 1.031 1.064 7.442 10.686
multiply_cannon_multrec 8220 15.4 4.045 4.212 9.764 9.889
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.432 9.679
qs_ot_get_derivative 99 11.5 0.001 0.001 9.468 9.507
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.865 9.075
mp_alltoall_i22 605 13.7 5.502 8.775 5.502 8.775
mp_waitall_1 84994 16.7 7.463 8.293 7.463 8.293
qs_rho_update_rho_low 110 7.6 0.000 0.001 7.038 7.072
calculate_rho_elec 110 8.6 0.225 0.225 7.038 7.072
sum_up_and_integrate 110 10.3 0.002 0.002 6.678 6.693
integrate_v_rspace 110 11.3 0.004 0.004 6.649 6.665
dbcsr_mm_accdrv_process 11614 15.7 3.795 4.117 5.571 5.781
init_scf_run 11 5.9 0.000 0.001 5.719 5.719
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.719 5.719
make_m2s 4110 13.4 0.043 0.044 5.218 5.681
qs_ot_get_p 110 10.4 0.001 0.001 5.522 5.569
make_images 4110 14.4 0.890 0.942 5.030 5.492
multiply_cannon_metrocomm3 8220 15.4 0.019 0.020 4.937 5.235
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.765 5.217
apply_single 110 13.6 0.000 0.000 4.765 5.216
cp_fm_cholesky_invert 11 10.9 5.161 5.166 5.161 5.166
pw_transfer 1331 11.6 0.074 0.075 4.866 4.874
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 4.748 4.757
ot_diis_step 99 11.5 0.015 0.016 4.609 4.609
fft_wrap_pw1pw2_140 451 13.1 1.279 1.287 4.154 4.162
qs_ot_p2m_diag 48 11.0 0.150 0.155 3.947 3.957
grid_integrate_task_list 110 12.3 3.677 3.737 3.677 3.737
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 3.651 3.653
density_rs2pw 110 9.6 0.004 0.004 3.478 3.491
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.993 3.459
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.433 3.434
hybrid_alltoall_any 4261 16.3 0.262 0.563 2.794 3.396
wfi_extrapolate 11 7.9 0.001 0.001 3.293 3.293
make_images_data 4110 15.4 0.046 0.049 2.687 3.285
calculate_dm_sparse 110 9.5 0.001 0.001 3.183 3.209
multiply_cannon_sync_h2d 8220 15.4 3.131 3.165 3.131 3.165
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.913 2.917
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.850 2.850
cp_fm_diag_elpa_base 48 14.0 2.310 2.510 2.848 2.848
fft3d_ps 1111 14.6 1.295 1.303 2.802 2.811
grid_collocate_task_list 110 9.6 2.669 2.680 2.669 2.680
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.641 2.663
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.410 2.428
calculate_first_density_matrix 1 7.0 0.000 0.000 2.313 2.313
potential_pw2rs 110 12.3 0.022 0.022 2.228 2.230
qs_env_update_s_mstruct 11 6.9 0.001 0.001 2.166 2.218
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.112 2.206
mp_alltoall_d11v 2046 13.8 2.027 2.083 2.027 2.083
cp_fm_cholesky_decompose 22 10.9 1.953 1.969 1.953 1.969
qs_create_task_list 11 7.9 0.000 0.001 1.879 1.920
generate_qs_task_list 11 8.9 0.730 0.783 1.878 1.920
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.856 1.862
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.765 1.808
jit_kernel_multiply 10 15.3 1.580 1.785 1.580 1.785
acc_transpose_blocks 8220 15.4 0.040 0.041 1.712 1.771
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=86.599000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1273.818182, yerr=66.702621
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 629.194752E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175954870160
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 57905.
MP_Allreduce 11059 797.
MP_Sync 87
MP_Alltoall 2226 3006175.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.026 206.224 206.322
qs_mol_dyn_low 1 2.0 0.003 0.003 205.793 205.808
qs_forces 11 3.9 0.005 0.005 205.706 205.707
qs_energies 11 4.9 0.002 0.002 200.108 200.125
scf_env_do_scf 11 5.9 0.001 0.001 183.573 183.577
scf_env_do_scf_inner_loop 117 6.6 0.002 0.008 163.298 163.299
dbcsr_multiply_generic 2507 12.6 0.182 0.185 125.682 126.149
qs_scf_new_mos 117 7.6 0.001 0.001 124.151 124.450
qs_scf_loop_do_ot 117 8.6 0.001 0.001 124.151 124.449
velocity_verlet 10 3.0 0.001 0.001 122.876 122.877
ot_scf_mini 117 9.6 0.003 0.003 117.457 117.741
multiply_cannon 2507 13.6 0.255 0.290 101.598 103.356
multiply_cannon_loop 2507 14.6 2.401 2.462 99.471 101.011
ot_mini 117 10.6 0.001 0.001 66.292 66.581
multiply_cannon_multrec 60168 15.6 31.840 34.151 41.583 43.159
qs_ot_get_derivative 117 11.6 0.001 0.001 41.508 41.770
rebuild_ks_matrix 128 8.3 0.001 0.001 33.422 33.689
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 33.421 33.688
mp_waitall_1 267128 16.5 28.806 32.327 28.806 32.327
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.001 30.226
qs_ot_get_p 128 10.4 0.001 0.001 28.885 29.195
multiply_cannon_sync_h2d 60168 15.6 26.353 27.745 26.353 27.745
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.310 25.001
apply_single 128 13.6 0.001 0.001 24.309 25.001
ot_diis_step 117 11.6 0.008 0.008 24.524 24.526
qs_ot_p2m_diag 83 11.4 0.079 0.091 22.375 22.443
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.085 20.270
init_scf_loop 11 6.9 0.000 0.001 20.202 20.203
cp_dbcsr_syevd 83 12.4 0.005 0.005 19.519 19.520
multiply_cannon_metrocomm3 60168 15.6 0.117 0.123 16.190 17.956
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.593 16.627
cp_fm_diag_elpa_base 83 14.4 16.527 16.566 16.589 16.624
prepare_preconditioner 11 7.9 0.000 0.000 15.655 15.689
make_preconditioner 11 8.9 0.000 0.000 15.655 15.689
make_full_inverse_cholesky 11 9.9 0.000 0.000 14.871 15.037
make_m2s 5014 13.6 0.105 0.111 13.749 14.226
make_images 5014 14.6 0.396 0.415 13.566 14.051
sum_up_and_integrate 128 10.3 0.002 0.004 13.879 13.894
integrate_v_rspace 128 11.3 0.003 0.004 13.820 13.838
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.004 13.164
calculate_rho_elec 128 8.7 0.045 0.064 13.004 13.163
init_scf_run 11 5.9 0.000 0.001 12.421 12.421
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.421 12.421
mp_sum_l 7950 12.9 9.080 10.551 9.080 10.551
dbcsr_mm_accdrv_process 124484 16.2 4.718 4.861 9.311 9.836
wfi_extrapolate 11 7.9 0.001 0.001 9.038 9.038
cp_fm_cholesky_invert 11 10.9 8.902 8.910 8.902 8.910
calculate_dm_sparse 128 9.5 0.001 0.001 8.501 8.599
qs_ot_get_orbitals 117 10.6 0.001 0.001 8.160 8.287
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.041 8.201
multiply_cannon_metrocomm1 60168 15.6 0.093 0.097 6.240 8.197
pw_transfer 1547 11.6 0.073 0.088 7.535 7.713
make_images_data 5014 15.6 0.066 0.073 6.724 7.701
fft_wrap_pw1pw2 1291 12.7 0.010 0.012 7.333 7.512
grid_integrate_task_list 128 12.3 7.080 7.458 7.080 7.458
density_rs2pw 128 9.7 0.006 0.007 6.639 7.263
hybrid_alltoall_any 5200 16.5 0.294 2.253 5.915 7.222
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.636 6.647
fft_wrap_pw1pw2_140 523 13.2 1.129 1.167 6.468 6.633
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.793 5.925
fft3d_ps 1291 14.7 2.191 2.851 5.337 5.679
grid_collocate_task_list 128 9.7 4.838 5.244 4.838 5.244
mp_alltoall_d11v 2415 14.1 4.293 5.197 4.293 5.197
cp_fm_cholesky_decompose 22 10.9 4.565 4.582 4.565 4.582
potential_pw2rs 128 12.3 0.009 0.010 4.385 4.403
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=206.322000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=595.818182, yerr=5.424158
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 2.183246E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5975232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.7
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 838.565888E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100942E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703955E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70860 2317615104
32768 < size <= 131072 722992 55511613440
131072 < size <= 4194304 1375664 1398181724160
4194304 < size <= 16777216 154704 1463834332048
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58357.
MP_Allreduce 11058 960.
MP_Sync 87
MP_Alltoall 1969 6429344.
MP_SendRecv 12032 47072.
MP_ISendRecv 12032 47072.
MP_Wait 25916
MP_ISend 11748 212467.
MP_IRecv 11748 212467.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.029 0.100 188.898 188.898
qs_mol_dyn_low 1 2.0 0.003 0.005 188.332 188.376
qs_forces 11 3.9 0.006 0.028 188.215 188.221
qs_energies 11 4.9 0.013 0.094 181.452 181.473
scf_env_do_scf 11 5.9 0.008 0.032 164.906 164.919
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 132.326 132.329
velocity_verlet 10 3.0 0.001 0.002 118.580 118.582
dbcsr_multiply_generic 2507 12.6 0.190 0.197 97.312 98.501
qs_scf_new_mos 117 7.6 0.001 0.001 93.859 94.434
qs_scf_loop_do_ot 117 8.6 0.001 0.001 93.858 94.433
ot_scf_mini 117 9.6 0.004 0.005 89.038 89.647
multiply_cannon 2507 13.6 0.504 0.561 77.226 82.192
multiply_cannon_loop 2507 14.6 1.568 1.636 73.826 76.511
ot_mini 117 10.6 0.001 0.001 50.070 50.647
mp_waitall_1 214728 16.6 23.904 37.779 23.904 37.779
multiply_cannon_multrec 30084 15.6 21.032 25.129 31.586 36.137
rebuild_ks_matrix 128 8.3 0.001 0.001 32.282 33.040
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.021 32.281 33.040
init_scf_loop 11 6.9 0.001 0.004 32.484 32.496
qs_ks_update_qs_env 128 7.6 0.001 0.001 29.033 29.729
qs_ot_get_derivative 117 11.6 0.001 0.002 28.084 28.687
prepare_preconditioner 11 7.9 0.000 0.000 28.124 28.201
make_preconditioner 11 8.9 0.000 0.001 28.124 28.201
multiply_cannon_metrocomm3 30084 15.6 0.097 0.102 15.525 28.150
make_full_inverse_cholesky 11 9.9 0.000 0.000 26.815 27.357
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 22.062 23.136
apply_single 128 13.6 0.001 0.001 22.061 23.135
ot_diis_step 117 11.6 0.017 0.032 21.808 21.810
qs_ot_get_p 128 10.4 0.001 0.001 20.957 21.715
multiply_cannon_sync_h2d 30084 15.6 18.072 20.984 18.072 20.984
cp_fm_cholesky_invert 11 10.9 16.395 16.408 16.395 16.408
qs_ot_p2m_diag 83 11.4 0.191 0.235 16.095 16.131
make_m2s 5014 13.6 0.089 0.095 14.140 15.851
make_images 5014 14.6 1.173 1.376 13.929 15.640
cp_dbcsr_syevd 83 12.4 0.005 0.006 14.897 14.901
sum_up_and_integrate 128 10.3 0.002 0.004 13.922 13.954
integrate_v_rspace 128 11.3 0.003 0.004 13.862 13.898
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.350 13.391
calculate_rho_elec 128 8.7 0.087 0.104 13.350 13.390
cp_fm_diag_elpa 83 13.4 0.000 0.000 11.748 11.780
cp_fm_diag_elpa_base 83 14.4 11.497 11.586 11.742 11.770
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 11.144 11.617
init_scf_run 11 5.9 0.000 0.001 11.598 11.600
scf_env_initial_rho_setup 11 6.9 0.008 0.061 11.598 11.600
multiply_cannon_metrocomm4 27577 15.6 0.107 0.120 3.865 10.935
dbcsr_mm_accdrv_process 62242 16.2 5.437 6.223 10.002 10.583
mp_irecv_dv 69486 16.3 3.659 10.534 3.659 10.534
make_images_data 5014 15.6 0.067 0.077 8.371 10.354
hybrid_alltoall_any 5200 16.5 0.352 1.534 7.069 9.618
pw_transfer 1547 11.6 0.085 0.093 8.493 8.547
wfi_extrapolate 11 7.9 0.006 0.044 8.398 8.398
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.270 8.327
grid_integrate_task_list 128 12.3 7.187 7.561 7.187 7.561
density_rs2pw 128 9.7 0.006 0.007 7.005 7.492
fft_wrap_pw1pw2_140 523 13.2 1.214 1.235 7.348 7.419
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.227 6.966
cp_fm_cholesky_decompose 22 10.9 6.832 6.920 6.832 6.920
calculate_dm_sparse 128 9.5 0.001 0.001 6.506 6.651
mp_sum_l 7950 12.9 4.150 6.095 4.150 6.095
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.080 6.089
fft3d_ps 1291 14.7 2.841 3.020 5.829 5.871
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.365 5.449
grid_collocate_task_list 128 9.7 5.008 5.404 5.008 5.404
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.216 5.389
mp_allgather_i34 2507 14.6 1.917 5.169 1.917 5.169
mp_alltoall_d11v 2415 14.1 4.108 4.602 4.108 4.602
potential_pw2rs 128 12.3 0.016 0.019 4.445 4.460
mp_sum_d 4465 12.1 2.659 3.940 2.659 3.940
dbcsr_complete_redistribute 395 12.7 0.777 0.861 3.125 3.927
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=188.898000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=799.909091, yerr=0.287480
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 959.848448E+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 58221.
MP_Allreduce 11080 1000.
MP_Sync 86
MP_Alltoall 1700 9383497.
MP_SendRecv 7874 75008.
MP_ISendRecv 7874 75008.
MP_Wait 21654
MP_ISend 11660 275234.
MP_IRecv 11660 275234.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.027 173.637 173.638
qs_mol_dyn_low 1 2.0 0.003 0.003 173.213 173.227
qs_forces 11 3.9 0.004 0.004 173.118 173.122
qs_energies 11 4.9 0.001 0.002 166.584 166.593
scf_env_do_scf 11 5.9 0.001 0.001 151.094 151.095
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 116.290 116.291
velocity_verlet 10 3.0 0.001 0.002 111.128 111.130
dbcsr_multiply_generic 2485 12.5 0.181 0.186 81.492 82.676
qs_scf_new_mos 116 7.6 0.001 0.001 79.622 79.932
qs_scf_loop_do_ot 116 8.6 0.001 0.001 79.621 79.931
ot_scf_mini 116 9.6 0.003 0.004 75.399 75.778
multiply_cannon 2485 13.5 0.505 0.523 61.848 66.852
multiply_cannon_loop 2485 14.5 1.118 1.172 58.864 61.408
ot_mini 116 10.6 0.001 0.001 42.259 42.609
init_scf_loop 11 6.9 0.000 0.000 34.705 34.705
mp_waitall_1 169034 16.6 24.507 34.578 24.507 34.578
prepare_preconditioner 11 7.9 0.000 0.000 30.723 30.765
make_preconditioner 11 8.9 0.000 0.000 30.723 30.765
rebuild_ks_matrix 127 8.3 0.001 0.001 29.920 30.358
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.017 29.920 30.358
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.264 29.716
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.938 27.332
multiply_cannon_multrec 19880 15.5 13.056 15.735 22.341 25.087
multiply_cannon_metrocomm3 19880 15.5 0.062 0.065 15.108 24.466
qs_ot_get_derivative 116 11.6 0.002 0.002 22.910 23.295
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.427 20.525
apply_single 127 13.6 0.001 0.001 19.426 20.524
ot_diis_step 116 11.6 0.018 0.018 19.243 19.244
qs_ot_get_p 127 10.4 0.001 0.001 18.606 19.061
make_m2s 4970 13.5 0.078 0.084 14.668 15.899
make_images 4970 14.5 1.145 1.250 14.436 15.665
multiply_cannon_sync_h2d 19880 15.5 13.592 15.026 13.592 15.026
qs_ot_p2m_diag 82 11.4 0.261 0.269 14.305 14.314
cp_fm_cholesky_invert 11 10.9 14.155 14.164 14.155 14.164
sum_up_and_integrate 127 10.3 0.002 0.003 13.799 13.822
integrate_v_rspace 127 11.3 0.003 0.004 13.740 13.763
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.421 13.451
calculate_rho_elec 127 8.7 0.129 0.144 13.420 13.451
cp_dbcsr_syevd 82 12.4 0.005 0.005 13.270 13.271
init_scf_run 11 5.9 0.000 0.001 10.487 10.487
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.486 10.487
make_images_data 4970 15.5 0.062 0.072 8.777 10.402
cp_fm_diag_elpa 82 13.4 0.000 0.000 10.135 10.153
cp_fm_diag_elpa_base 82 14.4 9.730 9.886 10.132 10.149
hybrid_alltoall_any 5155 16.4 0.444 2.017 7.606 9.983
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.017 9.292
multiply_cannon_metrocomm4 17395 15.5 0.067 0.077 3.412 8.970
dbcsr_mm_accdrv_process 41158 16.2 5.663 5.975 8.743 8.929
mp_irecv_dv 49801 16.2 3.283 8.711 3.283 8.711
pw_transfer 1535 11.6 0.084 0.100 8.587 8.690
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 8.365 8.474
grid_integrate_task_list 127 12.3 7.263 7.713 7.263 7.713
fft_wrap_pw1pw2_140 519 13.2 1.294 1.323 7.437 7.558
wfi_extrapolate 11 7.9 0.001 0.001 7.451 7.451
cp_fm_upper_to_full 104 14.8 5.810 7.396 5.810 7.396
cp_fm_cholesky_decompose 22 10.9 7.228 7.273 7.228 7.273
density_rs2pw 127 9.7 0.006 0.006 6.816 7.162
dbcsr_complete_redistribute 393 12.7 1.183 1.213 4.661 6.454
calculate_dm_sparse 127 9.5 0.001 0.001 5.820 5.926
fft3d_ps 1281 14.7 2.739 2.965 5.761 5.838
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.519 5.527
grid_collocate_task_list 127 9.7 5.117 5.467 5.117 5.467
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.651 5.294
copy_fm_to_dbcsr 208 11.6 0.002 0.002 3.492 5.280
mp_alltoall_d11v 2401 14.1 4.318 4.823 4.318 4.823
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.635 4.771
mp_allgather_i34 2485 14.5 1.488 4.598 1.488 4.598
mp_sum_l 7884 12.9 3.148 4.573 3.148 4.573
potential_pw2rs 127 12.3 0.021 0.023 4.275 4.283
transfer_fm_to_dbcsr 11 9.9 0.019 0.024 2.440 4.203
mp_alltoall_i22 712 14.1 2.011 3.944 2.011 3.944
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.885 3.919
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.719 3.720
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.475 3.519
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=173.638000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=911.363636, yerr=14.661469
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430454546432 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1973537472512 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 1992006770688 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 11613065416704 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239182565376 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239182565376 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 94.243766E+12 0.0% 0.0% 100.0%
flops max/rank 4.387745E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806547424 0.0% 0.0% 100.0%
number of processed stacks 6032640 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1128.3
marketing flops 145.662591E+12
-------------------------------------------------------------------------------
# multiplications 2534
max memory usage/rank 1.148957E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1155504
MPI messages size (bytes):
total size 2.039825E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.765312E+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 323336 36533174272
131072 < size <= 4194304 723016 792311627776
4194304 < size <= 16777216 70800 669922351760
16777216 < size 30960 541065216000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 57626.
MP_Allreduce 11184 1079.
MP_Sync 88
MP_Alltoall 1724 12509545.
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 0.011 0.032 187.322 187.324
qs_mol_dyn_low 1 2.0 0.003 0.003 186.936 186.952
qs_forces 11 3.9 0.004 0.004 186.808 186.817
qs_energies 11 4.9 0.002 0.002 179.712 179.727
scf_env_do_scf 11 5.9 0.001 0.001 163.182 163.193
velocity_verlet 10 3.0 0.002 0.002 124.941 124.945
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 117.084 117.085
dbcsr_multiply_generic 2534 12.6 0.190 0.196 81.050 81.666
qs_scf_new_mos 118 7.6 0.001 0.001 80.741 81.053
qs_scf_loop_do_ot 118 8.6 0.001 0.001 80.741 81.052
ot_scf_mini 118 9.6 0.003 0.004 76.132 76.486
multiply_cannon 2534 13.6 0.561 0.604 56.188 59.469
multiply_cannon_loop 2534 14.6 1.833 1.945 52.361 54.030
init_scf_loop 11 6.9 0.000 0.000 45.976 45.977
ot_mini 118 10.6 0.001 0.001 43.135 43.481
prepare_preconditioner 11 7.9 0.000 0.000 41.839 41.868
make_preconditioner 11 8.9 0.000 0.000 41.839 41.868
make_full_inverse_cholesky 11 9.9 0.011 0.023 35.465 40.467
multiply_cannon_multrec 30408 15.6 13.693 19.350 26.652 32.009
rebuild_ks_matrix 129 8.3 0.001 0.001 29.405 29.724
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.020 29.405 29.724
mp_waitall_1 149452 16.7 17.285 27.740 17.285 27.740
qs_ks_update_qs_env 129 7.6 0.001 0.001 26.573 26.857
qs_ot_get_derivative 118 11.6 0.002 0.002 23.303 23.662
make_m2s 5068 13.6 0.097 0.103 20.516 21.774
make_images 5068 14.6 1.966 2.299 20.209 21.469
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.231 19.724
apply_single 129 13.6 0.001 0.001 19.231 19.724
ot_diis_step 118 11.6 0.018 0.018 19.707 19.708
qs_ot_get_p 129 10.4 0.001 0.001 19.013 19.387
cp_fm_upper_to_full 106 14.8 11.419 16.822 11.419 16.822
cp_fm_cholesky_invert 11 10.9 15.920 15.929 15.920 15.929
multiply_cannon_metrocomm3 30408 15.6 0.050 0.053 6.372 15.209
qs_ot_p2m_diag 84 11.4 0.347 0.394 14.861 14.912
sum_up_and_integrate 129 10.3 0.002 0.003 13.957 13.981
integrate_v_rspace 129 11.3 0.003 0.004 13.897 13.922
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.670 13.717
calculate_rho_elec 129 8.7 0.174 0.191 13.669 13.717
cp_dbcsr_syevd 84 12.4 0.005 0.006 13.446 13.448
dbcsr_mm_accdrv_process 62840 16.2 8.450 9.265 12.525 13.060
make_images_data 5068 15.6 0.067 0.074 11.094 12.994
dbcsr_complete_redistribute 397 12.7 1.529 1.649 9.102 12.824
hybrid_alltoall_any 5255 16.5 0.534 2.232 9.697 12.254
copy_fm_to_dbcsr 210 11.7 0.001 0.002 7.745 11.469
multiply_cannon_sync_h2d 30408 15.6 10.539 11.437 10.539 11.437
init_scf_run 11 5.9 0.000 0.001 10.639 10.641
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.639 10.640
cp_fm_diag_elpa 84 13.4 0.000 0.000 10.221 10.234
cp_fm_diag_elpa_base 84 14.4 9.267 9.591 10.214 10.226
qs_ot_get_derivative_diag 78 12.4 0.002 0.002 9.821 10.086
transfer_fm_to_dbcsr 11 9.9 0.002 0.011 6.352 9.996
mp_alltoall_i22 720 14.1 5.605 9.418 5.605 9.418
pw_transfer 1559 11.6 0.086 0.102 8.919 8.984
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 8.691 8.764
grid_integrate_task_list 129 12.3 7.582 8.001 7.582 8.001
fft_wrap_pw1pw2_140 527 13.2 1.459 1.490 7.767 7.852
wfi_extrapolate 11 7.9 0.001 0.001 7.620 7.620
multiply_cannon_metrocomm4 25340 15.6 0.086 0.096 2.882 7.571
cp_fm_cholesky_decompose 22 10.9 7.408 7.511 7.408 7.511
mp_irecv_dv 76866 16.2 2.723 7.286 2.723 7.286
density_rs2pw 129 9.7 0.006 0.007 6.710 6.952
calculate_dm_sparse 129 9.5 0.001 0.001 6.297 6.385
fft3d_ps 1301 14.7 2.873 2.967 5.846 5.905
grid_collocate_task_list 129 9.7 5.322 5.755 5.322 5.755
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.437 5.468
mp_alltoall_d11v 2429 14.1 4.843 5.341 4.843 5.341
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.476 4.574
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.367 4.441
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 4.389 4.390
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.251 4.325
potential_pw2rs 129 12.3 0.023 0.024 4.199 4.207
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=187.324000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1085.090909, yerr=19.523667
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420243808256 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528908111872 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514772E+12 0.0% 0.0% 100.0%
flops max/rank 5.865089E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755942624 0.0% 0.0% 100.0%
number of processed stacks 1960712 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3445.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.523646E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 240672
MPI messages size (bytes):
total size 1.331455E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.532238E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 113904 59718500352
4194304 < size <= 16777216 104976 550376570880
16777216 < size 20208 721350290592
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8931 51.
MP_Alltoall 9654 799394.
MP_ISend 40068 2102573.
MP_IRecv 40068 2101676.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 57947.
MP_Allreduce 11127 1165.
MP_Sync 87
MP_Alltoall 1712 18838222.
MP_SendRecv 3840 122880.
MP_ISendRecv 3840 122880.
MP_Wait 16122
MP_ISend 10680 423556.
MP_IRecv 10680 423556.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.039 0.111 167.110 167.133
qs_mol_dyn_low 1 2.0 0.003 0.003 166.658 166.691
qs_forces 11 3.9 0.004 0.004 166.561 166.567
qs_energies 11 4.9 0.002 0.002 159.162 159.170
scf_env_do_scf 11 5.9 0.001 0.001 142.073 142.084
velocity_verlet 10 3.0 0.001 0.001 110.135 110.140
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 106.666 106.667
dbcsr_multiply_generic 2507 12.6 0.182 0.193 71.377 71.757
qs_scf_new_mos 117 7.6 0.001 0.001 70.273 70.351
qs_scf_loop_do_ot 117 8.6 0.001 0.001 70.273 70.351
ot_scf_mini 117 9.6 0.003 0.004 65.876 65.924
multiply_cannon 2507 13.6 0.562 0.600 53.021 56.449
multiply_cannon_loop 2507 14.6 0.803 0.840 50.129 50.918
ot_mini 117 10.6 0.001 0.001 37.144 37.195
init_scf_loop 11 6.9 0.000 0.000 35.259 35.260
prepare_preconditioner 11 7.9 0.000 0.000 31.467 31.502
make_preconditioner 11 8.9 0.000 0.000 31.467 31.502
mp_waitall_1 125778 16.7 24.436 30.689 24.436 30.689
make_full_inverse_cholesky 11 9.9 0.014 0.028 29.406 29.677
rebuild_ks_matrix 128 8.3 0.001 0.001 28.810 28.921
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.018 28.810 28.920
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.277 26.377
multiply_cannon_multrec 10028 15.6 10.396 14.302 17.931 20.844
qs_ot_get_derivative 117 11.6 0.001 0.002 20.237 20.292
multiply_cannon_metrocomm3 10028 15.6 0.025 0.027 12.405 19.353
cp_fm_cholesky_invert 11 10.9 17.940 17.946 17.940 17.946
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 16.930 17.073
apply_single 128 13.6 0.001 0.001 16.930 17.072
ot_diis_step 117 11.6 0.020 0.020 16.837 16.837
qs_ot_get_p 128 10.4 0.001 0.001 16.089 16.184
make_m2s 5014 13.6 0.066 0.071 14.592 15.396
make_images 5014 14.6 2.169 2.596 14.285 15.086
sum_up_and_integrate 128 10.3 0.002 0.002 13.966 14.009
integrate_v_rspace 128 11.3 0.004 0.004 13.906 13.951
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.826 13.849
calculate_rho_elec 128 8.7 0.256 0.267 13.825 13.849
qs_ot_p2m_diag 83 11.4 0.495 0.500 12.437 12.454
cp_dbcsr_syevd 83 12.4 0.005 0.006 11.286 11.286
multiply_cannon_sync_h2d 10028 15.6 10.885 11.104 10.885 11.104
init_scf_run 11 5.9 0.000 0.001 10.244 10.245
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.244 10.245
hybrid_alltoall_any 5200 16.5 0.843 3.795 8.276 9.698
make_images_data 5014 15.6 0.055 0.065 8.410 9.625
pw_transfer 1547 11.6 0.085 0.093 9.131 9.160
fft_wrap_pw1pw2 1291 12.7 0.010 0.010 8.910 8.947
cp_fm_diag_elpa 83 13.4 0.000 0.000 8.465 8.478
cp_fm_diag_elpa_base 83 14.4 8.219 8.308 8.459 8.472
grid_integrate_task_list 128 12.3 7.765 8.087 7.765 8.087
cp_fm_cholesky_decompose 22 10.9 7.892 7.987 7.892 7.987
fft_wrap_pw1pw2_140 523 13.2 1.800 1.850 7.929 7.972
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 7.920 7.964
dbcsr_mm_accdrv_process 20762 16.1 3.138 4.442 7.183 7.876
multiply_cannon_metrocomm1 10028 15.6 0.030 0.031 4.414 7.325
wfi_extrapolate 11 7.9 0.001 0.001 7.298 7.298
density_rs2pw 128 9.7 0.006 0.006 6.560 6.917
calculate_dm_sparse 128 9.5 0.001 0.001 6.048 6.130
grid_collocate_task_list 128 9.7 5.565 5.949 5.565 5.949
fft3d_ps 1291 14.7 2.763 2.838 5.644 5.666
dbcsr_complete_redistribute 395 12.7 2.134 2.166 5.110 5.499
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 5.296 5.297
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.091 5.103
mp_alltoall_d11v 2415 14.1 4.442 4.835 4.442 4.835
mp_allgather_i34 2507 14.6 1.073 4.487 1.073 4.487
potential_pw2rs 128 12.3 0.027 0.028 4.149 4.155
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.015 4.053
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.600 3.924
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.545 3.609
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.284 3.583
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.542 3.573
multiply_cannon_metrocomm4 7521 15.6 0.027 0.029 1.732 3.535
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.415 3.502
mp_irecv_dv 28860 15.9 1.693 3.458 1.693 3.458
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.435 3.445
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=167.133000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1422.272727, yerr=32.515985
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_performance_tests/20/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 11.606412E+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 1947808 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3442.6
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 3.184464E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 99400
MPI messages size (bytes):
total size 1.127422E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.342272E+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 44768 34745614336
4194304 < size <= 16777216 43984 376564613120
16777216 < size 10032 716108490000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3991 59293.
MP_Allreduce 11055 1504.
MP_Sync 86
MP_Alltoall 1700 36954339.
MP_SendRecv 1778 218624.
MP_ISendRecv 1778 218624.
MP_Wait 9728
MP_ISend 6360 1080477.
MP_IRecv 6360 1080477.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.070 288.348 288.360
qs_mol_dyn_low 1 2.0 0.003 0.003 287.753 287.777
qs_forces 11 3.9 0.005 0.005 287.656 287.659
qs_energies 11 4.9 0.005 0.005 278.848 278.858
scf_env_do_scf 11 5.9 0.001 0.001 256.644 256.650
velocity_verlet 10 3.0 0.002 0.002 208.387 208.395
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 132.407 132.410
init_scf_loop 11 6.9 0.000 0.000 123.961 123.966
prepare_preconditioner 11 7.9 0.000 0.000 119.202 119.225
make_preconditioner 11 8.9 0.000 0.000 119.202 119.225
make_full_inverse_cholesky 11 9.9 0.038 0.039 95.570 116.294
qs_scf_new_mos 116 7.6 0.001 0.001 89.437 89.506
qs_scf_loop_do_ot 116 8.6 0.001 0.001 89.436 89.505
ot_scf_mini 116 9.6 0.004 0.004 84.563 84.597
dbcsr_multiply_generic 2485 12.5 0.212 0.226 81.313 81.775
cp_fm_upper_to_full 104 14.8 53.047 75.555 53.047 75.555
multiply_cannon 2485 13.5 0.678 0.709 58.721 59.734
multiply_cannon_loop 2485 14.5 1.039 1.072 54.747 55.942
ot_mini 116 10.6 0.001 0.001 43.982 44.016
dbcsr_complete_redistribute 393 12.7 4.045 4.083 29.552 42.374
copy_fm_to_dbcsr 208 11.6 0.001 0.001 26.032 38.902
transfer_fm_to_dbcsr 11 9.9 0.030 0.031 23.590 36.412
mp_alltoall_i22 712 14.1 21.316 34.123 21.316 34.123
rebuild_ks_matrix 127 8.3 0.001 0.001 33.593 33.618
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.017 33.592 33.617
cp_fm_cholesky_invert 11 10.9 33.368 33.374 33.368 33.374
mp_waitall_1 102768 16.8 27.761 32.128 27.761 32.128
qs_ks_update_qs_env 127 7.6 0.001 0.001 31.169 31.198
qs_ot_get_p 127 10.4 0.001 0.001 25.658 25.713
qs_ot_get_derivative 116 11.6 0.002 0.002 24.480 24.516
qs_ot_p2m_diag 82 11.4 0.868 0.875 21.630 21.662
multiply_cannon_metrocomm3 9940 15.5 0.025 0.026 19.053 20.200
cp_dbcsr_syevd 82 12.4 0.006 0.006 19.845 19.847
make_m2s 4970 13.5 0.075 0.079 18.045 19.481
ot_diis_step 116 11.6 0.022 0.022 19.473 19.473
make_images 4970 14.5 3.045 3.247 17.566 19.006
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 18.812 18.926
apply_single 127 13.6 0.001 0.001 18.812 18.926
multiply_cannon_multrec 9940 15.5 10.206 12.052 17.923 17.987
cp_fm_diag_elpa 82 13.4 0.000 0.000 16.668 16.669
cp_fm_diag_elpa_base 82 14.4 12.319 13.829 16.665 16.665
qs_rho_update_rho_low 127 7.7 0.001 0.001 16.332 16.347
calculate_rho_elec 127 8.7 0.477 0.478 16.331 16.346
sum_up_and_integrate 127 10.3 0.002 0.002 15.945 16.040
integrate_v_rspace 127 11.3 0.005 0.005 15.883 15.978
multiply_cannon_sync_h2d 9940 15.5 14.190 14.226 14.190 14.226
hybrid_alltoall_any 5155 16.4 1.301 3.032 10.168 12.227
make_images_data 4970 15.5 0.064 0.070 10.000 12.101
init_scf_run 11 5.9 0.000 0.001 11.978 11.979
scf_env_initial_rho_setup 11 6.9 0.000 0.001 11.978 11.979
pw_transfer 1535 11.6 0.094 0.095 11.193 11.204
fft_wrap_pw1pw2 1281 12.7 0.011 0.012 10.957 10.966
fft_wrap_pw1pw2_140 519 13.2 3.014 3.045 9.768 9.784
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.660 9.693
dbcsr_mm_accdrv_process 20590 16.0 4.154 6.094 7.468 9.320
cp_fm_cholesky_decompose 22 10.9 8.912 8.935 8.912 8.935
wfi_extrapolate 11 7.9 0.001 0.001 8.890 8.890
grid_integrate_task_list 127 12.3 8.489 8.689 8.489 8.689
qs_energies_init_hamiltonians 11 5.9 0.001 0.005 8.104 8.105
density_rs2pw 127 9.7 0.005 0.005 7.418 7.460
mp_alltoall_d11v 2401 14.1 6.906 7.043 6.906 7.043
calculate_dm_sparse 127 9.5 0.001 0.001 6.735 6.798
grid_collocate_task_list 127 9.7 6.365 6.439 6.365 6.439
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.329 6.371
fft3d_ps 1281 14.7 2.828 2.838 6.249 6.268
copy_dbcsr_to_fm 185 11.7 0.004 0.004 6.169 6.226
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=288.360000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2889.454545, yerr=155.842207
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 1.260270E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255646.
MP_Allreduce 3139 6114.
MP_Sync 4
MP_Alltoall 54 4713940.
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 0.015 0.031 84.860 84.862
qs_energies 1 2.0 0.000 0.000 84.300 84.312
ls_scf 1 3.0 0.000 0.000 83.408 83.420
dbcsr_multiply_generic 111 6.7 0.014 0.015 72.460 72.579
multiply_cannon 111 7.7 0.017 0.020 55.808 56.810
multiply_cannon_loop 111 8.7 0.226 0.242 52.455 53.712
ls_scf_main 1 4.0 0.000 0.000 52.222 52.223
density_matrix_trs4 2 5.0 0.002 0.003 46.721 46.805
ls_scf_init_scf 1 4.0 0.000 0.000 28.159 28.160
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.038 27.092
mp_waitall_1 11031 10.9 22.466 25.113 22.466 25.113
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 24.959 24.976
multiply_cannon_multrec 2664 9.7 8.137 8.931 15.507 17.291
multiply_cannon_sync_h2d 2664 9.7 13.541 14.734 13.541 14.734
make_m2s 222 7.7 0.008 0.011 13.040 13.490
make_images 222 8.7 0.098 0.108 13.018 13.471
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.737 12.083
multiply_cannon_metrocomm3 2664 9.7 0.009 0.011 5.457 8.795
make_images_data 222 9.7 0.004 0.005 7.606 8.247
dbcsr_mm_accdrv_process 4760 10.4 0.592 0.710 6.990 7.956
hybrid_alltoall_any 227 10.6 0.216 1.846 6.526 7.832
dbcsr_mm_accdrv_process_sort 4760 11.4 6.196 7.076 6.196 7.076
calculate_norms 4752 9.8 5.506 6.183 5.506 6.183
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.008 5.117
mp_sum_l 887 5.1 3.084 4.893 3.084 4.893
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.307 3.616
make_images_sizes 222 9.7 0.000 0.000 0.750 3.580
mp_alltoall_i44 222 10.7 0.750 3.580 0.750 3.580
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.050 3.365
mp_irecv_dv 6231 10.9 2.034 3.333 2.034 3.333
arnoldi_extremal 4 6.8 0.000 0.000 3.215 3.238
arnoldi_normal_ev 4 7.8 0.001 0.002 3.215 3.238
build_subspace 16 8.4 0.009 0.012 3.126 3.128
ls_scf_post 1 4.0 0.000 0.000 3.026 3.039
ls_scf_store_result 1 5.0 0.000 0.000 2.852 2.890
dbcsr_special_finalize 555 9.7 0.005 0.006 2.332 2.764
dbcsr_merge_single_wm 555 10.7 0.457 0.571 2.324 2.755
make_images_pack 222 9.7 2.205 2.630 2.207 2.632
dbcsr_matrix_vector_mult 304 9.0 0.006 0.013 2.330 2.550
dbcsr_matrix_vector_mult_local 304 10.0 2.065 2.454 2.067 2.456
dbcsr_sort_data 658 11.4 2.123 2.455 2.123 2.455
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.280 2.322
buffer_matrices_ensure_size 222 8.7 1.756 2.067 1.756 2.067
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.757 1.759
rebuild_ks_matrix 3 7.3 0.000 0.000 1.748 1.749
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.748 1.749
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.862000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1131.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 2.141573E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 10075.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.039 0.055 89.754 89.756
qs_energies 1 2.0 0.000 0.000 89.286 89.289
ls_scf 1 3.0 0.000 0.000 87.955 87.958
dbcsr_multiply_generic 111 6.7 0.015 0.016 74.078 74.419
multiply_cannon 111 7.7 0.028 0.042 52.479 57.425
ls_scf_main 1 4.0 0.005 0.066 54.589 54.596
multiply_cannon_loop 111 8.7 0.136 0.148 49.691 53.313
density_matrix_trs4 2 5.0 0.003 0.018 48.914 49.165
ls_scf_init_scf 1 4.0 0.000 0.001 29.672 29.673
mp_waitall_1 9105 10.9 20.487 29.387 20.487 29.387
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.448 28.535
multiply_cannon_multrec 1332 9.7 13.288 16.575 22.540 26.968
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.115 26.124
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.307 20.576
make_m2s 222 7.7 0.006 0.007 14.986 15.522
make_images 222 8.7 1.370 1.703 14.955 15.493
dbcsr_mm_accdrv_process 4041 10.4 0.330 0.512 8.849 10.349
dbcsr_mm_accdrv_process_sort 4041 11.4 8.366 9.825 8.366 9.825
hybrid_alltoall_any 227 10.6 0.543 2.594 7.958 9.658
make_images_data 222 9.7 0.004 0.005 8.694 9.601
mp_sum_l 887 5.1 5.142 8.250 5.142 8.250
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.217 7.867
mp_irecv_dv 3311 11.0 3.197 7.802 3.197 7.802
calculate_norms 2376 9.8 6.077 6.665 6.077 6.665
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.905 6.568
multiply_cannon_sync_h2d 1332 9.7 4.924 6.005 4.924 6.005
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.915 5.159
arnoldi_extremal 4 6.8 0.000 0.000 4.611 4.623
arnoldi_normal_ev 4 7.8 0.001 0.005 4.611 4.623
build_subspace 16 8.4 0.014 0.021 4.362 4.365
ls_scf_post 1 4.0 0.000 0.000 3.694 3.698
ls_scf_store_result 1 5.0 0.000 0.000 3.404 3.500
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.142 3.369
dbcsr_matrix_vector_mult_local 304 10.0 2.749 3.234 2.751 3.236
mp_allgather_i34 111 8.7 0.774 3.015 0.774 3.015
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.606 2.705
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.135 2.452
dbcsr_data_new 4174 10.1 2.119 2.409 2.119 2.409
make_images_pack 222 9.7 1.820 2.130 1.823 2.133
dbcsr_sort_data 436 11.2 1.815 2.045 1.815 2.045
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.885 1.888
rebuild_ks_matrix 3 7.3 0.000 0.000 1.872 1.875
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.872 1.875
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=89.756000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1774.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 2.848772E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265448.
MP_Allreduce 3138 10896.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.043 91.555 91.556
qs_energies 1 2.0 0.000 0.000 91.029 91.033
ls_scf 1 3.0 0.000 0.000 89.620 89.623
dbcsr_multiply_generic 111 6.7 0.016 0.016 74.414 74.672
multiply_cannon 111 7.7 0.033 0.061 51.488 56.118
ls_scf_main 1 4.0 0.000 0.000 55.918 55.924
multiply_cannon_loop 111 8.7 0.117 0.134 48.748 52.393
density_matrix_trs4 2 5.0 0.002 0.003 49.997 50.129
mp_waitall_1 7281 11.0 23.005 32.225 23.005 32.225
ls_scf_init_scf 1 4.0 0.000 0.000 29.970 29.973
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.775 28.854
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.404 26.420
multiply_cannon_multrec 888 9.7 12.750 15.315 21.309 24.444
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 10.589 21.758
make_m2s 222 7.7 0.006 0.007 16.461 17.157
make_images 222 8.7 1.583 1.854 16.423 17.119
make_images_data 222 9.7 0.004 0.004 9.560 10.637
hybrid_alltoall_any 227 10.6 0.642 2.923 9.032 10.165
dbcsr_mm_accdrv_process 3754 10.4 0.329 0.495 8.073 9.311
dbcsr_mm_accdrv_process_sort 3754 11.4 7.639 8.815 7.639 8.815
mp_sum_l 887 5.1 4.744 7.974 4.744 7.974
multiply_cannon_sync_h2d 888 9.7 6.067 7.124 6.067 7.124
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.453 7.101
mp_irecv_dv 2335 11.1 2.437 7.031 2.437 7.031
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.501 6.934
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.602 6.402
arnoldi_extremal 4 6.8 0.000 0.000 5.067 5.088
arnoldi_normal_ev 4 7.8 0.001 0.005 5.067 5.088
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.813 5.051
calculate_norms 1584 9.8 4.368 4.951 4.368 4.951
build_subspace 16 8.4 0.014 0.020 4.768 4.774
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.445 3.777
mp_allgather_i34 111 8.7 0.855 3.762 0.855 3.762
ls_scf_post 1 4.0 0.000 0.000 3.732 3.736
dbcsr_matrix_vector_mult_local 304 10.0 3.034 3.611 3.036 3.614
ls_scf_store_result 1 5.0 0.000 0.000 3.475 3.558
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.844 2.953
dbcsr_data_new 4116 9.9 2.105 2.466 2.105 2.466
make_images_sizes 222 9.7 0.000 0.000 1.064 2.290
mp_alltoall_i44 222 10.7 1.063 2.290 1.063 2.290
dbcsr_sort_data 325 11.1 1.880 2.176 1.880 2.176
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.890 1.892
make_images_pack 222 9.7 1.639 1.878 1.642 1.881
rebuild_ks_matrix 3 7.3 0.000 0.000 1.872 1.874
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.872 1.874
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=91.556000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2245.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 3.333407E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 13030.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.026 0.043 96.852 96.853
qs_energies 1 2.0 0.000 0.000 96.263 96.271
ls_scf 1 3.0 0.000 0.000 94.612 94.621
dbcsr_multiply_generic 111 6.7 0.017 0.017 78.363 78.610
ls_scf_main 1 4.0 0.000 0.000 58.775 58.776
multiply_cannon 111 7.7 0.053 0.139 51.500 56.363
density_matrix_trs4 2 5.0 0.002 0.003 52.664 52.796
multiply_cannon_loop 111 8.7 0.152 0.168 46.437 49.985
ls_scf_init_scf 1 4.0 0.000 0.000 32.626 32.628
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.423 31.492
mp_waitall_1 6369 11.0 22.407 29.343 22.407 29.343
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.884 28.897
multiply_cannon_multrec 1332 9.7 14.173 17.286 22.061 24.657
make_m2s 222 7.7 0.006 0.008 21.048 22.543
make_images 222 8.7 3.144 3.610 20.998 22.495
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 9.133 17.200
make_images_data 222 9.7 0.004 0.004 11.703 13.307
hybrid_alltoall_any 227 10.6 0.801 3.817 11.047 12.797
dbcsr_mm_accdrv_process 3641 10.4 0.303 0.478 7.523 9.042
dbcsr_mm_accdrv_process_sort 3641 11.4 7.082 8.544 7.082 8.544
mp_sum_l 887 5.1 4.215 7.457 4.215 7.457
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.094 6.184
multiply_cannon_sync_h2d 1332 9.7 5.472 6.137 5.472 6.137
mp_irecv_dv 3229 10.9 2.069 6.103 2.069 6.103
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.224 5.813
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.406 5.249
arnoldi_extremal 4 6.8 0.000 0.000 5.202 5.220
arnoldi_normal_ev 4 7.8 0.001 0.005 5.202 5.219
build_subspace 16 8.4 0.014 0.021 4.871 4.879
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.611 4.811
mp_allgather_i34 111 8.7 2.176 4.677 2.176 4.677
calculate_norms 2376 9.8 4.177 4.529 4.177 4.529
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.603 3.894
dbcsr_matrix_vector_mult_local 304 10.0 3.207 3.711 3.210 3.713
dbcsr_sort_data 658 11.4 3.118 3.439 3.118 3.439
ls_scf_post 1 4.0 0.000 0.000 3.211 3.219
dbcsr_special_finalize 555 9.7 0.006 0.007 2.860 3.216
dbcsr_merge_single_wm 555 10.7 0.540 0.656 2.851 3.208
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.006 3.049
ls_scf_store_result 1 5.0 0.000 0.000 2.929 2.991
dbcsr_data_release 10477 10.7 1.592 2.430 1.592 2.430
make_images_pack 222 9.7 1.636 1.996 1.638 1.998
dbcsr_finalize 304 7.8 0.049 0.061 1.810 1.964
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=96.853000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2750.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 4.759384E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265536.
MP_Allreduce 3129 15263.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.067 0.072 91.612 91.613
qs_energies 1 2.0 0.000 0.000 90.857 90.861
ls_scf 1 3.0 0.000 0.000 88.896 88.900
dbcsr_multiply_generic 111 6.7 0.018 0.018 70.468 70.669
ls_scf_main 1 4.0 0.000 0.000 56.247 56.248
multiply_cannon 111 7.7 0.088 0.181 52.216 55.839
multiply_cannon_loop 111 8.7 0.088 0.094 49.688 51.204
density_matrix_trs4 2 5.0 0.002 0.003 49.305 49.375
ls_scf_init_scf 1 4.0 0.000 0.000 29.246 29.249
mp_waitall_1 5436 11.0 24.254 28.744 24.254 28.744
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.007 28.039
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.923 25.938
multiply_cannon_multrec 444 9.7 13.709 16.275 20.761 23.231
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 10.729 15.528
make_m2s 222 7.7 0.005 0.005 13.566 14.468
make_images 222 8.7 2.039 2.480 13.498 14.397
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 6.055 14.217
hybrid_alltoall_any 227 10.6 0.802 3.839 8.080 9.702
make_images_data 222 9.7 0.003 0.004 8.285 9.547
multiply_cannon_sync_h2d 444 9.7 6.667 8.124 6.667 8.124
dbcsr_mm_accdrv_process 3003 10.4 0.353 0.407 6.755 7.857
dbcsr_mm_accdrv_process_sort 3003 11.4 6.391 7.455 6.391 7.455
arnoldi_extremal 4 6.8 0.000 0.000 5.877 5.893
arnoldi_normal_ev 4 7.8 0.001 0.005 5.877 5.893
build_subspace 16 8.4 0.015 0.019 5.481 5.490
mp_sum_l 887 5.1 2.687 4.660 2.687 4.660
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.398 4.604
dbcsr_matrix_vector_mult 304 9.0 0.011 0.022 4.198 4.406
dbcsr_matrix_vector_mult_local 304 10.0 3.737 4.213 3.739 4.215
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.542 4.059
mp_irecv_dv 1241 11.2 1.531 4.035 1.531 4.035
calculate_norms 792 9.8 3.603 3.763 3.603 3.763
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.930 3.733
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.514 3.577
mp_allgather_i34 111 8.7 1.103 3.565 1.103 3.565
ls_scf_post 1 4.0 0.000 0.000 3.403 3.407
make_images_sizes 222 9.7 0.000 0.000 0.887 3.331
mp_alltoall_i44 222 10.7 0.887 3.331 0.887 3.331
ls_scf_store_result 1 5.0 0.000 0.000 3.196 3.229
dbcsr_data_new 4608 9.7 1.782 2.302 1.782 2.302
dbcsr_finalize 304 7.8 0.062 0.076 2.193 2.271
dbcsr_merge_all 275 8.9 0.477 0.522 2.051 2.108
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.017 2.018
rebuild_ks_matrix 3 7.3 0.000 0.000 1.985 1.986
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.985 1.986
qs_energies_init_hamiltonians 1 3.0 0.000 0.000 1.945 1.945
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=91.613000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3749.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 8.796508E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284089.
MP_Allreduce 3123 21388.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.071 0.090 104.764 104.764
qs_energies 1 2.0 0.000 0.000 103.400 103.406
ls_scf 1 3.0 0.000 0.000 100.483 100.489
dbcsr_multiply_generic 111 6.7 0.024 0.028 74.091 74.237
ls_scf_main 1 4.0 0.000 0.000 63.311 63.311
density_matrix_trs4 2 5.0 0.002 0.003 54.331 54.396
multiply_cannon 111 7.7 0.145 0.236 48.149 50.124
multiply_cannon_loop 111 8.7 0.099 0.101 45.161 45.854
ls_scf_init_scf 1 4.0 0.000 0.000 33.414 33.415
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.915 31.943
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.225 29.233
mp_waitall_1 4527 11.1 21.265 25.043 21.265 25.043
make_m2s 222 7.7 0.005 0.005 22.268 23.345
multiply_cannon_multrec 444 9.7 17.840 18.596 22.465 23.241
make_images 222 8.7 3.577 3.871 22.161 23.236
hybrid_alltoall_any 227 10.6 1.655 3.614 12.529 15.247
make_images_data 222 9.7 0.003 0.004 12.691 14.816
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 9.999 10.414
multiply_cannon_sync_h2d 444 9.7 8.787 8.844 8.787 8.844
arnoldi_extremal 4 6.8 0.000 0.000 7.472 7.481
arnoldi_normal_ev 4 7.8 0.003 0.009 7.472 7.481
build_subspace 16 8.4 0.026 0.036 6.928 6.939
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.586 5.737
dbcsr_matrix_vector_mult_local 304 10.0 5.139 5.454 5.142 5.457
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.094 5.188
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.858 5.106
dbcsr_mm_accdrv_process 1814 10.4 0.266 0.356 4.440 4.567
dbcsr_mm_accdrv_process_sort 1814 11.4 4.104 4.237 4.104 4.237
ls_scf_post 1 4.0 0.000 0.000 3.759 3.766
make_images_sizes 222 9.7 0.000 0.000 1.446 3.551
mp_alltoall_i44 222 10.7 1.446 3.550 1.446 3.550
ls_scf_store_result 1 5.0 0.000 0.000 3.483 3.519
mp_allgather_i34 111 8.7 1.109 3.488 1.109 3.488
calculate_norms 792 9.8 3.231 3.272 3.231 3.272
dbcsr_finalize 304 7.8 0.082 0.089 3.090 3.190
dbcsr_merge_all 275 8.9 0.886 0.914 2.859 2.942
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 2.885 2.885
dbcsr_complete_redistribute 5 7.6 1.450 1.486 2.762 2.877
matrix_ls_to_qs 2 6.0 0.000 0.000 2.422 2.544
dbcsr_sort_data 325 11.1 2.437 2.505 2.437 2.505
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.376 2.379
dbcsr_data_new 6591 9.6 1.861 2.330 1.861 2.330
rebuild_ks_matrix 3 7.3 0.000 0.000 2.311 2.313
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.311 2.313
dbcsr_new_transposed 4 7.5 0.247 0.264 2.289 2.301
dbcsr_frobenius_norm 74 6.6 2.057 2.136 2.208 2.242
dbcsr_add_d 103 6.2 0.000 0.000 2.150 2.236
dbcsr_add_anytype 103 7.2 0.859 0.891 2.150 2.236
dbcsr_data_release 12724 10.6 1.999 2.204 1.999 2.204
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=104.764000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6961.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/53a226acd6080c9664e5d7b371c6f14b2e21dcc0_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 588.816384E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 78072.
MP_Allreduce 13232 2081.
MP_Sync 1064
MP_Alltoall 2588 4805783.
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 0.013 0.032 231.225 231.227
qs_mol_dyn_low 1 2.0 0.003 0.014 230.260 230.287
qs_forces 5 3.8 0.004 0.004 230.176 230.178
qs_energies 5 4.8 0.001 0.002 227.223 227.249
scf_env_do_scf 5 5.8 0.000 0.001 211.938 211.941
scf_env_do_scf_inner_loop 105 6.6 0.002 0.006 185.578 185.579
qs_scf_new_mos 105 7.6 0.000 0.001 144.966 145.169
qs_scf_loop_do_ot 105 8.6 0.001 0.001 144.965 145.169
ot_scf_mini 105 9.6 0.003 0.003 135.119 135.281
dbcsr_multiply_generic 1445 12.2 0.126 0.133 134.114 134.782
multiply_cannon 1445 13.2 0.274 0.286 115.311 118.146
multiply_cannon_loop 1445 14.2 2.850 2.996 113.555 114.979
velocity_verlet 4 3.0 0.004 0.013 106.435 106.436
ot_mini 105 10.6 0.001 0.001 60.276 60.409
qs_ot_get_p 112 10.4 0.001 0.001 44.960 45.290
multiply_cannon_multrec 69360 15.2 29.744 34.749 39.544 44.689
mp_waitall_1 488190 16.1 34.879 41.124 34.879 41.124
qs_ot_get_derivative 55 11.6 0.001 0.001 38.561 38.728
qs_ot_p2m_diag 40 11.0 0.020 0.030 33.998 34.080
multiply_cannon_sync_h2d 69360 15.2 29.119 33.436 29.119 33.436
multiply_cannon_metrocomm3 69360 15.2 0.199 0.209 25.688 32.447
cp_dbcsr_syevd 40 12.0 0.002 0.002 30.819 30.820
rebuild_ks_matrix 110 8.4 0.000 0.000 28.928 29.086
qs_ks_build_kohn_sham_matrix 110 9.4 0.013 0.041 28.927 29.085
qs_ks_update_qs_env 112 7.6 0.001 0.001 26.546 26.694
init_scf_loop 7 6.6 0.000 0.000 26.328 26.329
cp_fm_syevd 40 13.0 0.000 0.001 25.726 25.859
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 23.030 23.440
apply_single 62 13.6 0.000 0.000 23.030 23.440
prepare_preconditioner 7 7.6 0.000 0.000 21.475 21.503
make_preconditioner 7 8.6 0.000 0.000 21.475 21.503
ot_new_cg_direction 55 11.6 0.001 0.001 21.042 21.042
cp_fm_redistribute_end 40 14.0 10.486 20.923 10.491 20.925
cp_fm_syevd_base 40 14.0 10.428 20.867 10.428 20.867
qs_rho_update_rho_low 110 7.6 0.001 0.001 17.438 17.723
calculate_rho_elec 110 8.6 0.030 0.032 17.438 17.723
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.729 14.954
make_full_inverse_cholesky 7 9.6 0.000 0.000 14.419 14.480
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 13.911 14.000
mp_sum_l 4764 12.2 11.970 12.707 11.970 12.707
init_scf_run 5 5.8 0.000 0.000 12.696 12.697
scf_env_initial_rho_setup 5 6.8 0.001 0.001 12.696 12.696
density_rs2pw 110 9.6 0.005 0.007 11.775 12.097
pw_transfer 1645 12.4 0.079 0.098 11.493 11.693
calculate_dm_sparse 110 9.5 0.000 0.001 11.397 11.594
fft_wrap_pw1pw2 1425 13.5 0.012 0.015 11.355 11.556
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 10.426 10.495
dbcsr_mm_accdrv_process 154766 15.8 6.211 6.400 9.668 10.448
qs_vxc_create 110 10.4 0.002 0.003 10.322 10.374
fft_wrap_pw1pw2_240 915 15.0 1.144 1.240 9.970 10.178
cp_fm_cholesky_invert 7 10.6 10.128 10.136 10.128 10.136
check_diag 80 13.5 8.684 8.937 9.547 9.683
acc_transpose_blocks 69360 15.2 0.353 0.366 8.713 9.080
transfer_rs2pw 445 10.6 0.007 0.008 8.494 8.839
sum_up_and_integrate 60 10.3 0.001 0.002 8.628 8.653
integrate_v_rspace 60 11.3 0.001 0.002 8.611 8.635
fft3d_pb 915 16.0 2.403 2.669 8.160 8.455
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.001 8.377 8.393
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 8.017 8.074
calculate_first_density_matrix 1 7.0 0.000 0.001 7.850 7.865
xc_rho_set_and_dset_create 110 12.4 0.077 0.098 7.253 7.504
multiply_cannon_metrocomm1 69360 15.2 0.095 0.103 4.541 7.326
xc_vxc_pw_create 60 11.3 0.039 0.049 6.886 6.934
make_m2s 2890 13.2 0.078 0.087 6.342 6.881
make_full_single_inverse 7 9.6 0.001 0.001 6.802 6.831
make_images 2890 14.2 0.239 0.259 6.236 6.775
acc_transpose_blocks_kernels 69360 16.2 0.847 0.889 5.946 6.216
xc_pw_derive 510 13.4 0.005 0.007 6.097 6.162
mp_alltoall_z22v 2340 17.7 5.168 5.473 5.168 5.473
jit_kernel_transpose 5 15.0 5.098 5.337 5.098 5.337
potential_pw2rs 60 12.3 0.003 0.003 4.900 4.972
multiply_cannon_metrocomm4 67915 15.2 0.185 0.202 2.019 4.780
mp_waitany 7680 13.5 4.333 4.757 4.333 4.757
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=231.227000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=561.000000, yerr=2.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 53a226acd6080c9664e5d7b371c6f14b2e21dcc0
Summary: empty
Status: OK