=== 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: 4e8b7c4a5ed53c7d43bc840588464933da05e6c1
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.2, ELPA 2022.11.001, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.1.0, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.1,
# SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Author: Matthias Krack (03.02.2023)
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2022.11.001
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.1.0
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.1
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
HDF5_VER := $(strip $(HDF5_VER))
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(HDF5_VER)/lib
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__HDF5
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta_prefixed_scalapack.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/01
job id: 45071918
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/02
job id: 45071919
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/03
job id: 45071921
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/04
job id: 45071923
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/05
job id: 45071924
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/06
job id: 45071925
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/07
job id: 45071926
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/08
job id: 45071929
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/09
job id: 45071931
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/10
job id: 45071933
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/11
job id: 45071934
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/12
job id: 45071935
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/13
job id: 45071936
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/14
job id: 45071937
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/15
job id: 45071938
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/16
job id: 45071939
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/17
job id: 45071940
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/18
job id: 45071941
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/19
job id: 45071942
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/20
job id: 45071943
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/21
job id: 45071944
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/22
job id: 45071945
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/23
job id: 45071946
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/24
job id: 45071947
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/25
job id: 45071948
--- 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/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/26
job id: 45071949
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 344 9.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.024 0.034 135.245 135.247
farming_run 1 2.0 134.648 134.649 135.208 135.211
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.467318E+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 231 1098811.
MP_Allreduce 491 2254389.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.032 116.112 116.113
qs_energies 1 2.0 0.000 0.000 115.907 115.909
mp2_main 1 3.0 0.000 0.000 113.729 113.732
mp2_gpw_main 1 4.0 0.020 0.027 112.675 112.677
mp2_ri_gpw_compute_in 1 5.0 0.172 0.175 93.644 94.186
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 55.390 55.932
mp2_eri_3c_integrate_gpw 272 7.0 0.152 0.166 41.686 47.158
get_2c_integrals 1 6.0 0.000 0.000 37.424 38.081
integrate_v_rspace 273 8.0 0.440 0.450 25.086 30.266
pw_transfer 6555 10.6 0.372 0.383 27.303 27.762
grid_integrate_task_list 273 9.0 20.908 26.565 20.908 26.565
fft_wrap_pw1pw2 5465 11.4 0.044 0.046 25.943 26.426
fft_wrap_pw1pw2_100 2178 12.4 1.179 1.414 23.460 23.949
compute_2c_integrals 1 7.0 0.002 0.003 19.726 19.727
compute_2c_integrals_loop_lm 1 8.0 0.002 0.004 18.844 19.446
mp2_eri_2c_integrate_gpw 1 9.0 2.383 2.437 18.841 19.442
rpa_ri_compute_en 1 5.0 0.001 0.002 18.921 19.062
cp_fm_cholesky_decompose 12 8.2 17.676 18.375 17.676 18.375
cholesky_decomp 1 7.0 0.000 0.000 16.542 17.232
fft3d_s 5443 13.4 16.081 16.605 16.103 16.625
ao_to_mo_and_store_B_mult_1 272 7.0 10.858 15.571 10.858 15.571
calculate_wavefunction 272 8.0 5.422 5.579 12.515 13.145
rpa_num_int 1 6.0 0.000 0.001 10.530 10.530
rpa_num_int_RPA_matrix_operati 8 7.0 0.001 0.001 10.519 10.529
calc_mat_Q 8 8.0 0.000 0.000 9.339 9.424
contract_S_to_Q 8 9.0 0.000 0.000 8.762 8.847
calc_potential_gpw 544 9.5 0.005 0.005 8.265 8.691
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.001 8.200 8.587
parallel_gemm_fm 14 9.1 0.000 0.000 8.335 8.428
parallel_gemm_fm_cosma 14 10.1 8.335 8.428 8.335 8.428
potential_pw2rs 545 10.0 0.107 0.109 7.658 8.325
collocate_single_gaussian 272 10.0 0.040 0.042 7.440 7.781
create_integ_mat 1 6.0 0.023 0.027 7.752 7.752
array2fm 1 7.0 0.000 0.000 6.701 7.179
pw_scatter_s 2720 13.7 4.407 4.537 4.407 4.537
pw_gather_s 2722 13.2 3.862 4.175 3.862 4.175
array2fm_buffer_send 1 8.0 2.946 3.181 2.946 3.181
pw_poisson_solve 545 10.5 1.118 1.157 2.189 2.410
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=112.677018, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2738.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 344 10.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.029 0.040 397.012 397.013
farming_run 1 2.0 396.190 396.195 396.960 396.962
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 16777216 0.0% 0.0% 100.0%
flops 14 x 32 x 32 565182464 0.0% 0.0% 100.0%
flops 29 x 32 x 32 585367552 0.0% 0.0% 100.0%
flops 14 x 14 x 32 626196480 0.0% 0.0% 100.0%
flops 29 x 14 x 32 638582784 0.0% 0.0% 100.0%
flops 14 x 29 x 32 638582784 0.0% 0.0% 100.0%
flops 29 x 29 x 32 682057728 0.0% 0.0% 100.0%
flops 14 x 32 x 14 897827128576 0.0% 0.0% 100.0%
flops 29 x 32 x 14 929989394432 0.0% 0.0% 100.0%
flops 14 x 32 x 29 929989394432 0.0% 0.0% 100.0%
flops 29 x 32 x 29 963203301376 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693481172992 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753962643456 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.172206E+12 0.0% 0.0% 100.0%
flops max/rank 150.696064E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249788821 0.0% 0.0% 100.0%
number of processed stacks 98736 0.0% 0.0% 100.0%
average stack size 0.0 0.0 2529.9
marketing flops 7.174951E+12
-------------------------------------------------------------------------------
# multiplications 1140
max memory usage/rank 1.226863E+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 744 385880.
MP_Allreduce 1941 22272.
MP_Sync 37
MP_Alltoall 77 8656213.
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 0.015 0.047 209.715 209.716
qs_energies 1 2.0 0.000 0.000 209.474 209.481
scf_env_do_scf 1 3.0 0.000 0.000 106.779 106.779
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.768 105.776
rebuild_ks_matrix 4 6.0 0.000 0.000 105.766 105.774
qs_ks_build_kohn_sham_matrix 4 7.0 0.058 0.066 105.766 105.774
hfx_ks_matrix 4 8.0 0.001 0.001 105.358 105.361
integrate_four_center 4 9.0 0.144 0.458 105.357 105.360
mp2_main 1 3.0 0.000 0.000 102.394 102.401
mp2_gpw_main 1 4.0 0.032 0.047 101.232 101.241
integrate_four_center_main 4 10.0 0.099 0.477 96.766 99.934
integrate_four_center_bin 266 11.0 96.668 99.932 96.668 99.932
init_scf_loop 1 4.0 0.000 0.000 92.422 92.423
mp2_ri_gpw_compute_in 1 5.0 0.064 0.064 74.805 75.766
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.003 54.420 55.380
mp2_eri_3c_integrate_gpw 91 7.0 0.143 0.162 42.011 47.187
integrate_v_rspace 95 8.0 0.401 0.576 28.426 33.440
pw_transfer 2240 10.6 0.145 0.165 29.884 30.409
fft_wrap_pw1pw2 1868 11.4 0.017 0.020 28.906 29.442
grid_integrate_task_list 95 9.0 23.698 28.919 23.698 28.919
ao_to_mo_and_store_B_mult_1 91 7.0 10.721 28.256 10.721 28.256
mp2_ri_gpw_compute_en 1 5.0 0.055 0.062 26.276 27.914
fft_wrap_pw1pw2_100 730 12.4 1.260 1.440 26.600 27.144
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.842 1.916 24.676 24.686
get_2c_integrals 1 6.0 0.000 0.000 20.306 20.321
compute_2c_integrals 1 7.0 0.002 0.003 19.281 19.284
compute_2c_integrals_loop_lm 1 8.0 0.001 0.002 18.815 19.151
mp2_eri_2c_integrate_gpw 1 9.0 1.737 1.849 18.814 19.151
fft3d_s 1823 13.4 18.396 18.912 18.409 18.925
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.353 14.353
calculate_wavefunction 91 8.0 2.013 2.044 9.728 9.964
mp2_ri_gpw_compute_en_expansio 172 7.0 0.557 0.582 8.750 9.478
potential_pw2rs 186 10.0 0.033 0.035 8.629 9.313
local_gemm 172 8.0 8.194 8.906 8.194 8.906
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.215 8.556
calc_potential_gpw 182 9.5 0.002 0.002 7.906 8.239
mp2_ri_gpw_compute_en_comm 22 7.0 0.500 0.520 7.694 8.232
collocate_single_gaussian 91 10.0 0.017 0.023 7.858 8.080
mp2_ri_gpw_compute_en_ener 172 7.0 6.344 6.455 6.344 6.455
mp_sendrecv_dm3 2068 8.0 5.728 6.286 5.728 6.286
mp_sync 37 10.5 3.282 6.031 3.282 6.031
pw_gather_s 912 13.2 4.908 5.386 4.908 5.386
pw_scatter_s 910 13.7 3.936 4.428 3.936 4.428
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.232471, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1514.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 452.763648E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62385.
MP_Allreduce 10249 272.
MP_Sync 530
MP_Alltoall 2083 579164.
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 0.034 0.060 54.670 54.673
qs_mol_dyn_low 1 2.0 0.004 0.008 54.178 54.185
qs_forces 11 3.9 0.003 0.015 54.104 54.106
qs_energies 11 4.9 0.002 0.005 52.571 52.583
scf_env_do_scf 11 5.9 0.000 0.001 46.282 46.283
scf_env_do_scf_inner_loop 108 6.5 0.003 0.009 44.026 44.027
qs_scf_new_mos 108 7.5 0.000 0.001 33.299 33.604
qs_scf_loop_do_ot 108 8.5 0.000 0.001 33.298 33.603
dbcsr_multiply_generic 2286 12.5 0.091 0.096 32.866 33.308
ot_scf_mini 108 9.5 0.002 0.003 31.669 31.850
multiply_cannon 2286 13.5 0.185 0.195 25.661 26.970
multiply_cannon_loop 2286 14.5 1.460 1.549 24.880 26.254
velocity_verlet 10 3.0 0.001 0.001 25.951 25.951
ot_mini 108 10.5 0.001 0.001 18.627 18.889
qs_ot_get_derivative 108 11.5 0.001 0.001 15.784 15.978
mp_waitall_1 245248 16.5 8.074 13.905 8.074 13.905
multiply_cannon_metrocomm3 54864 15.5 0.067 0.075 5.621 12.374
multiply_cannon_multrec 54864 15.5 4.279 6.637 7.763 11.241
qs_ot_get_p 119 10.4 0.001 0.002 8.396 8.724
rebuild_ks_matrix 119 8.3 0.000 0.000 8.302 8.445
qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.019 8.302 8.444
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.338 7.472
multiply_cannon_sync_h2d 54864 15.5 5.984 7.227 5.984 7.227
mp_sum_l 7207 12.9 4.945 6.612 4.945 6.612
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.314 5.762
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.621 5.664
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.301 5.414
init_scf_run 11 5.9 0.000 0.001 4.979 4.979
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.978 4.979
sum_up_and_integrate 119 10.3 0.012 0.014 4.804 4.811
integrate_v_rspace 119 11.3 0.003 0.003 4.792 4.801
cp_dbcsr_syevd 50 12.0 0.002 0.003 4.748 4.748
dbcsr_mm_accdrv_process 76910 16.1 1.142 1.822 3.406 4.745
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.542 4.725
calculate_rho_elec 119 8.7 0.012 0.017 4.541 4.724
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.460 4.461
cp_fm_redistribute_end 50 14.0 2.271 4.421 2.280 4.425
cp_fm_diag_elpa_base 50 14.0 2.138 4.298 2.142 4.304
rs_pw_transfer 974 11.9 0.012 0.013 3.030 3.180
calculate_dm_sparse 119 9.5 0.000 0.001 3.024 3.174
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.895 3.093
apply_single 119 13.6 0.000 0.000 2.895 3.092
jit_kernel_multiply 13 15.8 2.203 2.835 2.203 2.835
calculate_first_density_matrix 1 7.0 0.000 0.002 2.654 2.657
density_rs2pw 119 9.7 0.004 0.005 2.516 2.642
ot_diis_step 108 11.5 0.006 0.006 2.615 2.616
multiply_cannon_metrocomm1 54864 15.5 0.053 0.058 1.620 2.576
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.454 2.511
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 2.372 2.374
wfi_extrapolate 11 7.9 0.001 0.001 2.258 2.258
init_scf_loop 11 6.9 0.001 0.006 2.233 2.236
acc_transpose_blocks 54864 15.5 0.225 0.251 1.728 2.200
pw_transfer 1439 11.6 0.053 0.059 2.103 2.189
potential_pw2rs 119 12.3 0.004 0.004 2.119 2.138
grid_integrate_task_list 119 12.3 2.029 2.118 2.029 2.118
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 2.026 2.116
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.016 2.095
make_m2s 4572 13.5 0.053 0.056 1.961 2.016
make_images 4572 14.5 0.132 0.137 1.879 1.933
mp_sum_d 4131 12.0 1.286 1.894 1.286 1.894
fft3d_ps 1201 14.6 0.373 0.482 1.798 1.883
mp_alltoall_d11v 2130 13.8 1.444 1.714 1.444 1.714
fft_wrap_pw1pw2_140 487 13.2 0.081 0.095 1.544 1.637
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.563 1.591
mp_waitany 12084 13.8 1.260 1.385 1.260 1.385
grid_collocate_task_list 119 9.7 1.294 1.376 1.294 1.376
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=54.673000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=431.363636, yerr=0.979121
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 488.390656E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62664.
MP_Allreduce 10226 305.
MP_Sync 54
MP_Alltoall 2060 1487792.
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 0.018 0.037 39.656 39.658
qs_mol_dyn_low 1 2.0 0.003 0.004 39.401 39.409
qs_forces 11 3.9 0.002 0.002 39.320 39.321
qs_energies 11 4.9 0.002 0.006 37.609 37.612
scf_env_do_scf 11 5.9 0.001 0.002 32.448 32.449
scf_env_do_scf_inner_loop 108 6.5 0.002 0.008 29.875 29.875
dbcsr_multiply_generic 2286 12.5 0.099 0.102 21.292 21.633
qs_scf_new_mos 108 7.5 0.001 0.001 20.356 20.601
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.355 20.601
ot_scf_mini 108 9.5 0.002 0.003 19.472 19.641
velocity_verlet 10 3.0 0.001 0.001 18.543 18.545
multiply_cannon 2286 13.5 0.207 0.217 16.247 17.931
multiply_cannon_loop 2286 14.5 0.896 0.977 15.130 16.605
ot_mini 108 10.5 0.001 0.001 11.811 12.042
mp_waitall_1 200699 16.5 5.718 10.931 5.718 10.931
qs_ot_get_derivative 108 11.5 0.001 0.001 9.347 9.519
multiply_cannon_metrocomm3 27432 15.5 0.067 0.070 4.129 9.319
multiply_cannon_multrec 27432 15.5 1.972 4.380 5.855 8.626
rebuild_ks_matrix 119 8.3 0.000 0.000 7.398 7.534
qs_ks_build_kohn_sham_matrix 119 9.3 0.014 0.021 7.398 7.533
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.552 6.676
dbcsr_mm_accdrv_process 47894 16.0 3.056 5.160 3.813 5.645
qs_ot_get_p 119 10.4 0.001 0.001 4.764 4.994
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.595 4.444
sum_up_and_integrate 119 10.3 0.024 0.027 4.344 4.349
integrate_v_rspace 119 11.3 0.002 0.003 4.319 4.327
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.052 4.126
apply_single 119 13.6 0.000 0.000 3.052 4.126
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.006 4.036
calculate_rho_elec 119 8.7 0.021 0.024 4.005 4.035
mp_sum_l 7207 12.9 2.073 4.015 2.073 4.015
init_scf_run 11 5.9 0.000 0.001 3.895 3.896
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.895 3.895
rs_pw_transfer 974 11.9 0.010 0.011 2.798 3.283
qs_ot_p2m_diag 50 11.0 0.009 0.013 3.253 3.271
multiply_cannon_sync_h2d 27432 15.5 2.192 2.915 2.192 2.915
density_rs2pw 119 9.7 0.004 0.005 2.327 2.838
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.811 2.812
make_m2s 4572 13.5 0.052 0.054 2.528 2.750
make_images 4572 14.5 0.200 0.236 2.441 2.661
init_scf_loop 11 6.9 0.001 0.006 2.551 2.551
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.436 2.437
ot_diis_step 108 11.5 0.011 0.011 2.411 2.412
cp_fm_redistribute_end 50 14.0 1.232 2.398 1.237 2.402
cp_fm_diag_elpa_base 50 14.0 1.130 2.293 1.160 2.339
calculate_first_density_matrix 1 7.0 0.000 0.003 2.330 2.332
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.171 2.262
pw_transfer 1439 11.6 0.065 0.072 2.137 2.173
calculate_dm_sparse 119 9.5 0.000 0.001 2.021 2.097
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.045 2.084
potential_pw2rs 119 12.3 0.006 0.006 2.057 2.073
grid_integrate_task_list 119 12.3 1.834 1.941 1.834 1.941
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.891 1.893
fft3d_ps 1201 14.6 0.520 0.572 1.747 1.786
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.677 1.718
prepare_preconditioner 11 7.9 0.000 0.000 1.610 1.635
make_preconditioner 11 8.9 0.000 0.001 1.610 1.635
fft_wrap_pw1pw2_140 487 13.2 0.079 0.086 1.566 1.606
jit_kernel_multiply 8 16.1 0.705 1.577 0.705 1.577
make_images_data 4572 15.5 0.044 0.051 1.154 1.569
make_full_inverse_cholesky 11 9.9 0.000 0.001 1.511 1.567
acc_transpose_blocks 27432 15.5 0.108 0.113 1.189 1.496
wfi_extrapolate 11 7.9 0.001 0.001 1.493 1.493
hybrid_alltoall_any 4725 16.4 0.051 0.112 1.012 1.482
mp_alltoall_d11v 2130 13.8 1.237 1.415 1.237 1.415
grid_collocate_task_list 119 9.7 1.226 1.365 1.226 1.365
mp_allgather_i34 2286 14.5 0.561 1.311 0.561 1.311
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.252 1.259
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.164 1.210
mp_sum_d 4131 12.0 0.622 1.065 0.622 1.065
mp_waitany 5720 13.7 0.548 1.055 0.548 1.055
rs_pw_transfer_RS2PW_140 130 11.5 0.139 0.148 0.573 1.054
qs_energies_init_hamiltonians 11 5.9 0.000 0.003 0.983 0.985
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.908 0.921
mp_alltoall_z22v 1201 16.6 0.790 0.895 0.790 0.895
acc_transpose_blocks_kernels 27432 16.5 0.181 0.272 0.655 0.866
rs_pw_transfer_PW2RS_50 119 14.3 0.590 0.608 0.753 0.825
make_images_sizes 4572 15.5 0.005 0.005 0.561 0.805
mp_alltoall_i44 4572 16.5 0.557 0.801 0.557 0.801
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=39.658000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=463.909091, yerr=1.083307
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 522.219520E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63497.
MP_Allreduce 10075 307.
MP_Sync 54
MP_Alltoall 1821 1955501.
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 0.010 0.026 32.802 32.803
qs_mol_dyn_low 1 2.0 0.003 0.003 32.584 32.592
qs_forces 11 3.9 0.002 0.002 32.520 32.522
qs_energies 11 4.9 0.001 0.001 30.928 30.931
scf_env_do_scf 11 5.9 0.000 0.001 26.311 26.311
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 23.680 23.680
dbcsr_multiply_generic 2286 12.5 0.093 0.097 16.375 16.483
velocity_verlet 10 3.0 0.001 0.001 15.613 15.614
qs_scf_new_mos 108 7.5 0.001 0.001 15.218 15.233
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.217 15.233
ot_scf_mini 108 9.5 0.002 0.002 14.489 14.503
multiply_cannon 2286 13.5 0.195 0.201 12.979 13.784
multiply_cannon_loop 2286 14.5 0.633 0.667 12.179 13.003
ot_mini 108 10.5 0.001 0.001 8.825 8.846
qs_ot_get_derivative 108 11.5 0.001 0.001 7.317 7.330
multiply_cannon_multrec 18288 15.5 1.943 2.918 6.761 7.002
rebuild_ks_matrix 119 8.3 0.000 0.000 6.534 6.550
qs_ks_build_kohn_sham_matrix 119 9.3 0.016 0.019 6.533 6.549
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.781 5.796
dbcsr_mm_accdrv_process 38222 16.0 4.077 5.266 4.733 5.562
sum_up_and_integrate 119 10.3 0.030 0.031 4.094 4.100
integrate_v_rspace 119 11.3 0.003 0.003 4.063 4.072
mp_waitall_1 158411 16.6 2.690 3.998 2.690 3.998
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.597 3.605
calculate_rho_elec 119 8.7 0.031 0.032 3.596 3.605
qs_ot_get_p 119 10.4 0.001 0.001 3.491 3.518
init_scf_run 11 5.9 0.000 0.001 3.459 3.459
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.458 3.458
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.728 3.348
rs_pw_transfer 974 11.9 0.009 0.010 2.390 2.622
init_scf_loop 11 6.9 0.000 0.000 2.612 2.613
multiply_cannon_metrocomm3 18288 15.5 0.044 0.045 1.464 2.587
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.378 2.386
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.044 2.362
apply_single 119 13.6 0.000 0.000 2.044 2.362
density_rs2pw 119 9.7 0.004 0.004 2.110 2.353
calculate_first_density_matrix 1 7.0 0.000 0.000 2.201 2.203
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.104 2.104
pw_transfer 1439 11.6 0.066 0.072 2.064 2.075
make_m2s 4572 13.5 0.044 0.045 1.842 1.996
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.970 1.985
make_images 4572 14.5 0.190 0.203 1.757 1.910
grid_integrate_task_list 119 12.3 1.793 1.908 1.793 1.908
potential_pw2rs 119 12.3 0.007 0.008 1.853 1.865
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.826 1.836
calculate_dm_sparse 119 9.5 0.000 0.000 1.823 1.836
cp_fm_diag_elpa_base 50 14.0 1.798 1.811 1.824 1.834
prepare_preconditioner 11 7.9 0.000 0.000 1.797 1.805
make_preconditioner 11 8.9 0.000 0.000 1.797 1.805
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.646 1.731
jit_kernel_multiply 9 16.3 0.605 1.677 0.605 1.677
fft3d_ps 1201 14.6 0.530 0.549 1.650 1.665
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.660 1.662
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.601 1.609
fft_wrap_pw1pw2_140 487 13.2 0.091 0.097 1.550 1.562
mp_sum_l 7207 12.9 1.166 1.544 1.166 1.544
multiply_cannon_sync_h2d 18288 15.5 1.348 1.517 1.348 1.517
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.506 1.513
ot_diis_step 108 11.5 0.011 0.011 1.484 1.485
grid_collocate_task_list 119 9.7 1.215 1.372 1.215 1.372
acc_transpose_blocks 18288 15.5 0.075 0.076 1.230 1.258
wfi_extrapolate 11 7.9 0.001 0.001 1.208 1.208
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.188 1.195
make_images_data 4572 15.5 0.044 0.048 0.827 0.988
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.936 0.938
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.894 0.918
hybrid_alltoall_any 4725 16.4 0.055 0.112 0.714 0.891
mp_alltoall_d11v 2130 13.8 0.721 0.854 0.721 0.854
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.824 0.826
mp_alltoall_z22v 1201 16.6 0.713 0.813 0.713 0.813
acc_transpose_blocks_kernels 18288 16.5 0.211 0.219 0.785 0.801
cp_fm_cholesky_invert 11 10.9 0.760 0.764 0.760 0.764
mp_waitany 9880 13.7 0.505 0.764 0.505 0.764
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.663 0.744
rs_pw_transfer_RS2PW_140 130 11.5 0.118 0.122 0.494 0.727
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=32.803000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.818182, yerr=1.991718
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 558.129152E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63496.
MP_Allreduce 10074 349.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.032 35.515 35.516
qs_mol_dyn_low 1 2.0 0.003 0.003 35.198 35.205
qs_forces 11 3.9 0.002 0.002 35.079 35.080
qs_energies 11 4.9 0.002 0.006 33.345 33.350
scf_env_do_scf 11 5.9 0.001 0.001 28.344 28.346
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 24.837 24.837
velocity_verlet 10 3.0 0.001 0.002 18.016 18.020
dbcsr_multiply_generic 2286 12.5 0.099 0.102 17.889 18.008
qs_scf_new_mos 108 7.5 0.001 0.001 16.217 16.270
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.217 16.269
ot_scf_mini 108 9.5 0.002 0.003 15.278 15.333
multiply_cannon 2286 13.5 0.227 0.278 14.347 14.701
multiply_cannon_loop 2286 14.5 0.936 0.968 13.406 13.745
ot_mini 108 10.5 0.001 0.001 9.136 9.205
multiply_cannon_multrec 27432 15.5 2.330 3.028 8.575 8.949
qs_ot_get_derivative 108 11.5 0.001 0.001 7.331 7.385
dbcsr_mm_accdrv_process 47916 15.9 5.423 7.093 6.149 7.360
rebuild_ks_matrix 119 8.3 0.000 0.000 6.717 6.769
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.015 6.716 6.769
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.972 6.019
sum_up_and_integrate 119 10.3 0.035 0.038 3.896 3.906
integrate_v_rspace 119 11.3 0.003 0.003 3.861 3.871
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.639 3.672
calculate_rho_elec 119 8.7 0.040 0.046 3.638 3.671
init_scf_run 11 5.9 0.000 0.002 3.637 3.637
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.636 3.637
qs_ot_get_p 119 10.4 0.001 0.001 3.542 3.617
init_scf_loop 11 6.9 0.001 0.007 3.482 3.482
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.760 3.147
prepare_preconditioner 11 7.9 0.000 0.000 2.571 2.578
make_preconditioner 11 8.9 0.000 0.001 2.571 2.578
make_full_inverse_cholesky 11 9.9 0.000 0.001 2.180 2.505
mp_waitall_1 137007 16.6 1.784 2.328 1.784 2.328
make_m2s 4572 13.5 0.054 0.056 2.176 2.285
rs_pw_transfer 974 11.9 0.009 0.009 2.088 2.283
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.237 2.245
density_rs2pw 119 9.7 0.004 0.004 2.058 2.243
calculate_first_density_matrix 1 7.0 0.000 0.002 2.239 2.240
pw_transfer 1439 11.6 0.066 0.071 2.145 2.184
make_images 4572 14.5 0.270 0.331 2.070 2.177
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.030 2.155
apply_single 119 13.6 0.000 0.000 2.030 2.155
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.052 2.096
calculate_dm_sparse 119 9.5 0.000 0.000 2.025 2.088
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.906 1.933
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.918 1.919
grid_integrate_task_list 119 12.3 1.830 1.906 1.830 1.906
jit_kernel_multiply 10 15.9 0.665 1.824 0.665 1.824
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.812 1.813
ot_diis_step 108 11.5 0.012 0.012 1.765 1.766
fft3d_ps 1201 14.6 0.562 0.609 1.726 1.763
fft_wrap_pw1pw2_140 487 13.2 0.088 0.095 1.687 1.736
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.655 1.663
cp_fm_diag_elpa_base 50 14.0 1.620 1.635 1.653 1.661
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.626 1.644
potential_pw2rs 119 12.3 0.008 0.010 1.631 1.637
acc_transpose_blocks 27432 15.5 0.112 0.114 1.457 1.478
grid_collocate_task_list 119 9.7 1.222 1.364 1.222 1.364
wfi_extrapolate 11 7.9 0.001 0.001 1.348 1.348
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.761 1.333
mp_sum_l 7207 12.9 0.953 1.268 0.953 1.268
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.212 1.221
cp_fm_upper_to_full 72 14.2 0.809 1.155 0.809 1.155
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.102 1.121
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.098 1.099
dbcsr_complete_redistribute 329 12.2 0.119 0.141 0.792 1.071
multiply_cannon_sync_h2d 27432 15.5 0.999 1.057 0.999 1.057
make_images_data 4572 15.5 0.045 0.049 0.848 0.964
mp_alltoall_d11v 2130 13.8 0.743 0.962 0.743 0.962
hybrid_alltoall_any 4725 16.4 0.061 0.150 0.739 0.883
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.792 0.871
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.581 0.852
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.837 0.842
acc_transpose_blocks_kernels 27432 16.5 0.269 0.276 0.826 0.834
mp_alltoall_z22v 1201 16.6 0.747 0.779 0.747 0.779
cp_fm_cholesky_invert 11 10.9 0.772 0.775 0.772 0.775
mp_alltoall_i22 627 13.8 0.432 0.730 0.432 0.730
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=35.516000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=530.727273, yerr=3.305393
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 610.250752E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63495.
MP_Allreduce 10074 348.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.027 28.488 28.489
qs_mol_dyn_low 1 2.0 0.003 0.003 28.317 28.327
qs_forces 11 3.9 0.002 0.002 28.251 28.251
qs_energies 11 4.9 0.001 0.002 26.507 26.510
scf_env_do_scf 11 5.9 0.000 0.001 21.655 21.656
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 19.138 19.138
velocity_verlet 10 3.0 0.001 0.001 14.610 14.613
dbcsr_multiply_generic 2286 12.5 0.092 0.095 12.120 12.194
qs_scf_new_mos 108 7.5 0.001 0.001 10.971 11.000
qs_scf_loop_do_ot 108 8.5 0.001 0.001 10.970 10.999
ot_scf_mini 108 9.5 0.002 0.002 10.294 10.323
multiply_cannon 2286 13.5 0.228 0.235 9.581 10.048
multiply_cannon_loop 2286 14.5 0.329 0.341 8.681 8.869
multiply_cannon_multrec 9144 15.5 1.581 1.824 5.881 6.137
rebuild_ks_matrix 119 8.3 0.000 0.000 6.104 6.124
qs_ks_build_kohn_sham_matrix 119 9.3 0.015 0.016 6.103 6.123
ot_mini 108 10.5 0.001 0.001 5.623 5.660
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.427 5.446
qs_ot_get_derivative 108 11.5 0.001 0.001 4.320 4.349
dbcsr_mm_accdrv_process 12550 15.8 3.249 3.976 4.198 4.270
sum_up_and_integrate 119 10.3 0.038 0.042 3.756 3.760
integrate_v_rspace 119 11.3 0.003 0.003 3.718 3.722
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.629 3.643
calculate_rho_elec 119 8.7 0.060 0.061 3.629 3.643
init_scf_run 11 5.9 0.000 0.001 3.412 3.412
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.411 3.411
qs_ot_get_p 119 10.4 0.001 0.001 2.902 2.941
init_scf_loop 11 6.9 0.000 0.000 2.496 2.498
calculate_first_density_matrix 1 7.0 0.000 0.000 2.244 2.248
pw_transfer 1439 11.6 0.066 0.069 2.220 2.230
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.125 2.138
density_rs2pw 119 9.7 0.004 0.004 1.962 2.104
grid_integrate_task_list 119 12.3 1.847 1.932 1.847 1.932
mp_waitall_1 115863 16.7 1.446 1.932 1.446 1.932
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.917 1.920
make_m2s 4572 13.5 0.034 0.035 1.750 1.912
rs_pw_transfer 974 11.9 0.008 0.008 1.754 1.879
make_images 4572 14.5 0.268 0.303 1.661 1.823
fft3d_ps 1201 14.6 0.570 0.582 1.785 1.795
calculate_dm_sparse 119 9.5 0.000 0.000 1.751 1.771
fft_wrap_pw1pw2_140 487 13.2 0.087 0.090 1.747 1.760
prepare_preconditioner 11 7.9 0.000 0.000 1.737 1.741
make_preconditioner 11 8.9 0.000 0.000 1.737 1.740
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.718 1.719
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.628 1.653
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.620 1.621
jit_kernel_multiply 8 15.4 0.911 1.565 0.911 1.565
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.530 1.552
potential_pw2rs 119 12.3 0.010 0.010 1.485 1.488
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.454 1.466
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.423 1.431
cp_fm_diag_elpa_base 50 14.0 1.396 1.414 1.422 1.430
grid_collocate_task_list 119 9.7 1.272 1.393 1.272 1.393
ot_diis_step 108 11.5 0.013 0.013 1.290 1.290
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.224 1.225
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.162 1.187
apply_single 119 13.6 0.000 0.000 1.162 1.187
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.169 1.176
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.127 1.141
wfi_extrapolate 11 7.9 0.001 0.001 1.115 1.115
make_images_data 4572 15.5 0.039 0.042 0.814 1.014
hybrid_alltoall_any 4725 16.4 0.062 0.176 0.787 1.012
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.868 0.919
acc_transpose_blocks 9144 15.5 0.038 0.038 0.888 0.895
cp_fm_cholesky_invert 11 10.9 0.888 0.891 0.888 0.891
mp_alltoall_d11v 2130 13.8 0.760 0.888 0.760 0.888
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.800 0.802
multiply_cannon_sync_h2d 9144 15.5 0.712 0.785 0.712 0.785
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.726 0.734
qs_env_update_s_mstruct 11 6.9 0.000 0.001 0.678 0.729
multiply_cannon_metrocomm3 9144 15.5 0.018 0.019 0.335 0.720
mp_alltoall_z22v 1201 16.6 0.662 0.718 0.662 0.718
acc_transpose_blocks_kernels 9144 16.5 0.118 0.121 0.646 0.650
mp_allgather_i34 2286 14.5 0.232 0.646 0.232 0.646
yz_to_x 606 15.1 0.266 0.276 0.606 0.625
x_to_yz 595 16.2 0.276 0.288 0.597 0.605
mp_waitany 5200 13.7 0.450 0.581 0.450 0.581
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=28.489000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=576.181818, yerr=7.145478
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 762.523648E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63729.
MP_Allreduce 10074 433.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.037 41.602 41.603
qs_mol_dyn_low 1 2.0 0.003 0.003 41.398 41.407
qs_forces 11 3.9 0.002 0.002 41.331 41.333
qs_energies 11 4.9 0.001 0.001 39.342 39.346
scf_env_do_scf 11 5.9 0.001 0.001 33.612 33.612
scf_env_do_scf_inner_loop 108 6.5 0.004 0.012 25.644 25.644
velocity_verlet 10 3.0 0.001 0.001 23.456 23.461
dbcsr_multiply_generic 2286 12.5 0.101 0.103 17.311 17.456
qs_scf_new_mos 108 7.5 0.001 0.001 15.799 15.895
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.798 15.894
ot_scf_mini 108 9.5 0.002 0.002 14.717 14.818
multiply_cannon 2286 13.5 0.302 0.310 13.484 14.389
multiply_cannon_loop 2286 14.5 0.344 0.349 12.236 13.128
ot_mini 108 10.5 0.001 0.001 8.726 8.846
multiply_cannon_multrec 9144 15.5 3.379 4.671 8.553 8.646
init_scf_loop 11 6.9 0.000 0.000 7.941 7.941
rebuild_ks_matrix 119 8.3 0.000 0.000 7.361 7.506
qs_ks_build_kohn_sham_matrix 119 9.3 0.015 0.015 7.361 7.505
prepare_preconditioner 11 7.9 0.000 0.000 6.966 6.980
make_preconditioner 11 8.9 0.000 0.000 6.966 6.980
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.474 6.857
qs_ot_get_derivative 108 11.5 0.001 0.001 6.688 6.788
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.657 6.787
dbcsr_mm_accdrv_process 12550 15.8 4.080 5.598 5.052 6.321
cp_fm_upper_to_full 72 14.2 3.254 4.692 3.254 4.692
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.264 4.269
calculate_rho_elec 119 8.7 0.118 0.121 4.264 4.269
sum_up_and_integrate 119 10.3 0.065 0.067 4.044 4.049
integrate_v_rspace 119 11.3 0.004 0.004 3.978 3.985
init_scf_run 11 5.9 0.000 0.001 3.707 3.707
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.707 3.707
qs_ot_get_p 119 10.4 0.001 0.001 3.285 3.427
mp_waitall_1 94719 16.7 2.320 3.234 2.320 3.234
dbcsr_complete_redistribute 329 12.2 0.287 0.293 2.024 2.887
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.445 2.862
pw_transfer 1439 11.6 0.069 0.070 2.823 2.827
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.724 2.728
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.718 2.561
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.188 2.475
apply_single 119 13.6 0.000 0.000 2.188 2.475
mp_alltoall_i22 627 13.8 1.562 2.468 1.562 2.468
make_m2s 4572 13.5 0.038 0.039 2.281 2.446
fft3d_ps 1201 14.6 0.601 0.609 2.348 2.350
make_images 4572 14.5 0.353 0.384 2.160 2.325
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.488 2.321
fft_wrap_pw1pw2_140 487 13.2 0.095 0.098 2.300 2.307
calculate_first_density_matrix 1 7.0 0.000 0.000 2.255 2.259
density_rs2pw 119 9.7 0.004 0.004 2.189 2.210
calculate_dm_sparse 119 9.5 0.000 0.000 2.157 2.178
multiply_cannon_metrocomm3 9144 15.5 0.020 0.020 1.256 2.115
grid_integrate_task_list 119 12.3 2.058 2.072 2.058 2.072
ot_diis_step 108 11.5 0.014 0.015 2.009 2.009
qs_ot_p2m_diag 50 11.0 0.043 0.044 1.995 1.996
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.965 1.967
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.761 1.764
mp_sum_l 7207 12.9 1.005 1.748 1.005 1.748
jit_kernel_multiply 10 15.5 0.945 1.729 0.945 1.729
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.669 1.723
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.718 1.718
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.615 1.676
grid_collocate_task_list 119 9.7 1.490 1.501 1.490 1.501
cp_fm_cholesky_invert 11 10.9 1.467 1.470 1.467 1.470
potential_pw2rs 119 12.3 0.014 0.014 1.445 1.447
rs_pw_transfer 974 11.9 0.009 0.010 1.421 1.443
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.421 1.421
cp_fm_diag_elpa_base 50 14.0 1.273 1.327 1.419 1.419
hybrid_alltoall_any 4725 16.4 0.087 0.146 1.158 1.403
wfi_extrapolate 11 7.9 0.001 0.001 1.382 1.382
make_images_data 4572 15.5 0.043 0.046 1.099 1.312
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.221 1.231
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.126 1.146
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.095 1.109
mp_alltoall_d11v 2130 13.8 1.058 1.083 1.058 1.083
multiply_cannon_sync_h2d 9144 15.5 1.042 1.049 1.042 1.049
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.988 1.031
yz_to_x 606 15.1 0.459 0.471 0.938 0.947
qs_create_task_list 11 7.9 0.000 0.000 0.934 0.946
generate_qs_task_list 11 8.9 0.370 0.388 0.934 0.946
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.916 0.930
acc_transpose_blocks 9144 15.5 0.037 0.038 0.892 0.898
copy_dbcsr_to_fm 153 11.3 0.002 0.002 0.794 0.877
mp_alltoall_z22v 1201 16.6 0.797 0.841 0.797 0.841
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=41.603000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=712.545455, yerr=15.393999
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 504.995840E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66218.
MP_Allreduce 9696 492.
MP_Sync 52
MP_Alltoall 1938 1379060.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.038 0.062 81.944 81.946
qs_mol_dyn_low 1 2.0 0.003 0.005 81.577 81.588
qs_forces 11 3.9 0.004 0.011 81.501 81.503
qs_energies 11 4.9 0.002 0.004 78.589 78.602
scf_env_do_scf 11 5.9 0.000 0.001 69.716 69.718
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 64.069 64.070
dbcsr_multiply_generic 2055 12.4 0.102 0.106 49.842 50.033
qs_scf_new_mos 99 7.5 0.000 0.001 46.351 46.468
qs_scf_loop_do_ot 99 8.5 0.001 0.001 46.351 46.468
ot_scf_mini 99 9.5 0.002 0.002 44.006 44.111
velocity_verlet 10 3.0 0.001 0.002 42.832 42.833
multiply_cannon 2055 13.4 0.180 0.185 41.752 42.442
multiply_cannon_loop 2055 14.4 1.493 1.520 40.777 41.437
ot_mini 99 10.5 0.001 0.001 25.734 25.832
qs_ot_get_derivative 99 11.5 0.001 0.002 19.078 19.164
multiply_cannon_multrec 49320 15.4 12.515 13.221 17.356 18.243
rebuild_ks_matrix 110 8.3 0.000 0.000 14.811 14.954
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.016 14.810 14.954
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.997 13.132
mp_waitall_1 220248 16.4 10.343 11.045 10.343 11.045
multiply_cannon_sync_h2d 49320 15.4 10.390 10.862 10.390 10.862
qs_ot_get_p 110 10.4 0.001 0.001 9.969 10.106
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.253 7.802
apply_preconditioner_dbcsr 110 12.6 0.000 0.001 7.139 7.624
apply_single 110 13.6 0.000 0.000 7.139 7.624
sum_up_and_integrate 110 10.3 0.037 0.046 7.300 7.315
integrate_v_rspace 110 11.3 0.003 0.005 7.263 7.287
multiply_cannon_metrocomm3 49320 15.4 0.078 0.080 6.164 7.109
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.784 6.900
calculate_rho_elec 110 8.6 0.021 0.026 6.784 6.899
qs_ot_p2m_diag 48 11.0 0.013 0.019 6.842 6.865
init_scf_run 11 5.9 0.000 0.001 6.797 6.797
scf_env_initial_rho_setup 11 6.9 0.001 0.003 6.796 6.797
ot_diis_step 99 11.5 0.006 0.009 6.494 6.494
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.972 5.972
init_scf_loop 11 6.9 0.000 0.000 5.616 5.616
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.388 5.412
cp_fm_diag_elpa_base 48 14.0 5.374 5.398 5.386 5.410
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.247 5.305
dbcsr_mm_accdrv_process 87628 16.1 1.903 1.970 4.717 4.995
rs_pw_transfer 902 11.9 0.012 0.013 3.900 4.457
mp_sum_l 6514 12.8 3.735 4.334 3.735 4.334
density_rs2pw 110 9.6 0.004 0.005 3.602 4.123
wfi_extrapolate 11 7.9 0.001 0.003 4.122 4.122
make_m2s 4110 13.4 0.060 0.066 3.906 4.007
make_images 4110 14.4 0.176 0.189 3.811 3.916
calculate_dm_sparse 110 9.5 0.000 0.001 3.718 3.799
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.538 3.543
pw_transfer 1331 11.6 0.055 0.067 3.388 3.474
prepare_preconditioner 11 7.9 0.000 0.000 3.429 3.452
make_preconditioner 11 8.9 0.000 0.000 3.429 3.452
grid_integrate_task_list 110 12.3 3.251 3.394 3.251 3.394
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.300 3.387
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.213 3.264
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.182 3.237
multiply_cannon_metrocomm1 49320 15.4 0.060 0.062 2.223 3.178
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.055 3.093
fft3d_ps 1111 14.6 0.778 0.880 2.828 2.899
fft_wrap_pw1pw2_140 451 13.1 0.170 0.190 2.776 2.869
potential_pw2rs 110 12.3 0.006 0.007 2.812 2.849
mp_alltoall_d11v 2046 13.8 2.081 2.608 2.081 2.608
calculate_first_density_matrix 1 7.0 0.000 0.002 2.588 2.591
jit_kernel_multiply 13 15.9 2.538 2.551 2.538 2.551
mp_waitany 14300 13.8 1.840 2.485 1.840 2.485
grid_collocate_task_list 110 9.6 2.084 2.294 2.084 2.294
acc_transpose_blocks 49320 15.4 0.206 0.214 2.054 2.101
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.956 1.977
cp_fm_cholesky_invert 11 10.9 1.892 1.896 1.892 1.896
make_images_data 4110 15.4 0.042 0.045 1.766 1.881
hybrid_alltoall_any 4261 16.3 0.082 0.479 1.537 1.816
mp_sum_d 3885 11.9 1.277 1.722 1.277 1.722
mp_alltoall_z22v 1111 16.6 1.396 1.675 1.396 1.675
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.652 1.672
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=81.946000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=477.909091, yerr=3.175318
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/10/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 390.715586E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 5019072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 196.1
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 588.058624E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66437.
MP_Allreduce 9695 570.
MP_Sync 52
MP_Alltoall 1717 1325455.
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 0.014 0.027 71.178 71.179
qs_mol_dyn_low 1 2.0 0.003 0.003 70.961 70.971
qs_forces 11 3.9 0.003 0.003 70.864 70.864
qs_energies 11 4.9 0.001 0.002 67.482 67.485
scf_env_do_scf 11 5.9 0.000 0.001 58.802 58.805
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 50.796 50.796
dbcsr_multiply_generic 2055 12.4 0.113 0.118 37.974 38.169
velocity_verlet 10 3.0 0.001 0.001 37.298 37.300
qs_scf_new_mos 99 7.5 0.001 0.001 34.035 34.137
qs_scf_loop_do_ot 99 8.5 0.001 0.001 34.035 34.137
ot_scf_mini 99 9.5 0.003 0.003 32.361 32.453
multiply_cannon 2055 13.4 0.222 0.242 31.124 32.206
multiply_cannon_loop 2055 14.4 0.925 0.958 29.684 30.523
ot_mini 99 10.5 0.001 0.001 18.560 18.669
multiply_cannon_multrec 24660 15.4 7.560 9.013 13.794 15.377
rebuild_ks_matrix 110 8.3 0.000 0.001 13.982 14.077
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.017 13.982 14.076
qs_ot_get_derivative 99 11.5 0.001 0.001 12.708 12.805
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.313 12.397
mp_waitall_1 176588 16.5 7.740 10.706 7.740 10.706
multiply_cannon_sync_h2d 24660 15.4 6.960 8.066 6.960 8.066
init_scf_loop 11 6.9 0.000 0.000 7.958 7.958
multiply_cannon_metrocomm3 24660 15.4 0.068 0.072 5.216 7.858
qs_ot_get_p 110 10.4 0.001 0.001 7.136 7.268
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.542 7.257
apply_single 110 13.6 0.000 0.001 6.542 7.257
sum_up_and_integrate 110 10.3 0.053 0.059 6.799 6.811
integrate_v_rspace 110 11.3 0.003 0.003 6.746 6.758
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.497 6.507
calculate_rho_elec 110 8.6 0.040 0.048 6.496 6.507
dbcsr_mm_accdrv_process 52282 16.1 4.679 5.756 6.075 6.383
init_scf_run 11 5.9 0.000 0.002 6.288 6.289
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.288 6.289
prepare_preconditioner 11 7.9 0.000 0.000 5.890 5.905
make_preconditioner 11 8.9 0.000 0.000 5.890 5.905
ot_diis_step 99 11.5 0.010 0.010 5.791 5.792
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.459 5.616
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.780 5.470
qs_ot_p2m_diag 48 11.0 0.029 0.044 5.103 5.124
make_m2s 4110 13.4 0.057 0.060 4.320 4.862
make_images 4110 14.4 0.398 0.445 4.212 4.750
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.654 4.654
pw_transfer 1331 11.6 0.066 0.074 3.994 4.162
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.887 4.058
density_rs2pw 110 9.6 0.004 0.005 3.602 4.046
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.996 4.008
cp_fm_diag_elpa_base 48 14.0 3.945 3.969 3.993 4.004
wfi_extrapolate 11 7.9 0.001 0.001 3.653 3.653
rs_pw_transfer 902 11.9 0.012 0.014 3.142 3.613
fft3d_ps 1111 14.6 1.107 1.331 3.239 3.390
fft_wrap_pw1pw2_140 451 13.1 0.202 0.220 3.197 3.370
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.311 3.369
grid_integrate_task_list 110 12.3 3.154 3.346 3.154 3.346
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.236 3.238
calculate_dm_sparse 110 9.5 0.001 0.001 3.083 3.117
make_images_data 4110 15.4 0.046 0.050 2.345 2.911
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.874 2.903
hybrid_alltoall_any 4261 16.3 0.102 0.441 2.055 2.815
cp_fm_cholesky_invert 11 10.9 2.788 2.795 2.788 2.795
potential_pw2rs 110 12.3 0.008 0.008 2.597 2.615
mp_sum_l 6514 12.8 1.905 2.543 1.905 2.543
calculate_first_density_matrix 1 7.0 0.000 0.000 2.537 2.540
grid_collocate_task_list 110 9.6 2.051 2.470 2.051 2.470
mp_alltoall_d11v 2046 13.8 1.874 2.259 1.874 2.259
jit_kernel_multiply 11 16.2 1.041 2.167 1.041 2.167
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.991 2.007
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.868 1.869
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.836 1.843
mp_allgather_i34 2055 14.4 0.769 1.790 0.769 1.790
mp_waitany 10164 13.8 1.212 1.668 1.212 1.668
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.647 1.661
multiply_cannon_metrocomm4 22605 15.4 0.074 0.077 0.810 1.654
acc_transpose_blocks 24660 15.4 0.110 0.114 1.538 1.570
mp_alltoall_z22v 1111 16.6 1.431 1.548 1.431 1.548
dbcsr_complete_redistribute 325 12.2 0.235 0.304 1.244 1.511
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.370 1.477
cp_fm_cholesky_decompose 22 10.9 1.457 1.463 1.457 1.463
mp_irecv_dv 57340 16.2 0.684 1.439 0.684 1.439
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=71.179000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=556.181818, yerr=7.107207
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 661.954560E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66428.
MP_Allreduce 9694 567.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.040 60.799 60.801
qs_mol_dyn_low 1 2.0 0.003 0.003 60.425 60.436
qs_forces 11 3.9 0.003 0.005 60.352 60.353
qs_energies 11 4.9 0.002 0.006 57.097 57.101
scf_env_do_scf 11 5.9 0.001 0.004 49.064 49.065
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 40.452 40.453
velocity_verlet 10 3.0 0.001 0.002 32.948 32.950
dbcsr_multiply_generic 2055 12.4 0.107 0.110 28.385 28.655
qs_scf_new_mos 99 7.5 0.001 0.001 25.136 25.275
qs_scf_loop_do_ot 99 8.5 0.001 0.001 25.136 25.274
ot_scf_mini 99 9.5 0.002 0.003 23.914 24.059
multiply_cannon 2055 13.4 0.209 0.221 22.076 23.474
multiply_cannon_loop 2055 14.4 0.613 0.635 20.941 22.072
ot_mini 99 10.5 0.001 0.001 13.548 13.695
rebuild_ks_matrix 110 8.3 0.000 0.000 12.554 12.695
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.021 12.554 12.695
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.064 11.185
multiply_cannon_multrec 16440 15.4 3.811 4.572 9.571 10.242
mp_waitall_1 139946 16.5 7.162 9.868 7.162 9.868
qs_ot_get_derivative 99 11.5 0.001 0.001 9.082 9.229
init_scf_loop 11 6.9 0.003 0.015 8.574 8.575
multiply_cannon_metrocomm3 16440 15.4 0.044 0.045 4.484 7.114
prepare_preconditioner 11 7.9 0.001 0.004 6.759 6.781
make_preconditioner 11 8.9 0.001 0.003 6.759 6.781
sum_up_and_integrate 110 10.3 0.061 0.062 6.672 6.685
integrate_v_rspace 110 11.3 0.003 0.003 6.611 6.625
make_full_inverse_cholesky 11 9.9 0.000 0.001 6.101 6.454
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.147 6.156
calculate_rho_elec 110 8.6 0.059 0.059 6.146 6.155
qs_ot_get_p 110 10.4 0.001 0.001 5.570 5.755
dbcsr_mm_accdrv_process 34862 16.1 4.536 5.307 5.615 5.746
init_scf_run 11 5.9 0.000 0.001 5.609 5.610
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.609 5.609
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.001 5.420
apply_single 110 13.6 0.000 0.000 5.001 5.420
make_m2s 4110 13.4 0.050 0.051 4.154 4.476
density_rs2pw 110 9.6 0.004 0.005 3.230 4.456
ot_diis_step 99 11.5 0.011 0.011 4.435 4.435
make_images 4110 14.4 0.392 0.512 4.040 4.361
multiply_cannon_sync_h2d 16440 15.4 3.632 4.301 3.632 4.301
rs_pw_transfer 902 11.9 0.011 0.012 2.773 3.985
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.934 3.939
pw_transfer 1331 11.6 0.066 0.074 3.820 3.830
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.098 3.755
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.713 3.725
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.591 3.591
grid_integrate_task_list 110 12.3 3.155 3.414 3.155 3.414
fft_wrap_pw1pw2_140 451 13.1 0.214 0.219 3.192 3.207
wfi_extrapolate 11 7.9 0.001 0.001 3.049 3.049
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.038 3.048
cp_fm_diag_elpa_base 48 14.0 2.971 3.001 3.037 3.046
fft3d_ps 1111 14.6 1.092 1.110 3.012 3.024
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.810 2.812
make_images_data 4110 15.4 0.043 0.048 2.414 2.800
cp_fm_cholesky_invert 11 10.9 2.702 2.708 2.702 2.708
hybrid_alltoall_any 4261 16.3 0.106 0.376 2.146 2.675
calculate_dm_sparse 110 9.5 0.001 0.001 2.543 2.577
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.447 2.509
multiply_cannon_metrocomm4 14385 15.4 0.045 0.049 0.904 2.501
calculate_first_density_matrix 1 7.0 0.001 0.008 2.463 2.464
mp_waitany 17072 13.8 1.182 2.443 1.182 2.443
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.350 2.431
grid_collocate_task_list 110 9.6 2.083 2.424 2.083 2.424
potential_pw2rs 110 12.3 0.010 0.010 2.395 2.402
mp_irecv_dv 48980 15.7 0.832 2.369 0.832 2.369
mp_sum_l 6514 12.8 1.546 2.131 1.546 2.131
rs_pw_transfer_RS2PW_140 121 11.5 0.177 0.183 0.893 2.098
mp_alltoall_d11v 2046 13.8 1.795 2.079 1.795 2.079
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 1.975 1.977
dbcsr_complete_redistribute 325 12.2 0.324 0.354 1.449 1.909
cp_fm_upper_to_full 70 14.2 1.403 1.784 1.403 1.784
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.693 1.714
cp_fm_cholesky_decompose 22 10.9 1.589 1.611 1.589 1.611
mp_allgather_i34 2055 14.4 0.462 1.531 0.462 1.531
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.505 1.521
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.365 1.456
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.986 1.439
acc_transpose_blocks 16440 15.4 0.070 0.073 1.258 1.280
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.258 1.269
rs_gather_matrices 110 12.3 0.234 0.261 0.981 1.246
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=60.801000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=627.181818, yerr=9.193062
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 723.865600E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66426.
MP_Allreduce 9694 608.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.041 65.539 65.540
qs_mol_dyn_low 1 2.0 0.003 0.004 65.246 65.256
qs_forces 11 3.9 0.003 0.005 65.151 65.153
qs_energies 11 4.9 0.003 0.009 61.697 61.702
scf_env_do_scf 11 5.9 0.001 0.002 53.430 53.433
scf_env_do_scf_inner_loop 99 6.5 0.003 0.010 41.761 41.761
velocity_verlet 10 3.0 0.001 0.001 36.969 36.972
dbcsr_multiply_generic 2055 12.4 0.114 0.118 29.335 29.538
qs_scf_new_mos 99 7.5 0.001 0.001 26.555 26.657
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.554 26.656
ot_scf_mini 99 9.5 0.003 0.003 24.942 25.050
multiply_cannon 2055 13.4 0.244 0.264 22.498 23.613
multiply_cannon_loop 2055 14.4 0.883 0.904 21.095 21.602
ot_mini 99 10.5 0.001 0.001 14.116 14.237
multiply_cannon_multrec 24660 15.4 4.238 6.798 12.664 13.909
rebuild_ks_matrix 110 8.3 0.000 0.000 12.323 12.432
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.015 12.323 12.431
init_scf_loop 11 6.9 0.002 0.008 11.628 11.629
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.875 10.967
qs_ot_get_derivative 99 11.5 0.001 0.001 9.987 10.096
prepare_preconditioner 11 7.9 0.000 0.000 9.873 9.886
make_preconditioner 11 8.9 0.001 0.003 9.873 9.886
make_full_inverse_cholesky 11 9.9 0.000 0.001 8.089 9.547
dbcsr_mm_accdrv_process 52304 16.0 6.811 8.372 8.278 9.192
sum_up_and_integrate 110 10.3 0.068 0.071 6.586 6.596
integrate_v_rspace 110 11.3 0.003 0.003 6.518 6.531
mp_waitall_1 121746 16.5 4.295 6.236 4.295 6.236
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.193 6.202
calculate_rho_elec 110 8.6 0.078 0.082 6.192 6.201
qs_ot_get_p 110 10.4 0.001 0.001 5.867 6.004
make_m2s 4110 13.4 0.060 0.063 5.268 5.596
init_scf_run 11 5.9 0.000 0.002 5.476 5.477
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.476 5.476
make_images 4110 14.4 0.575 0.695 5.128 5.452
cp_fm_upper_to_full 70 14.2 3.303 4.806 3.303 4.806
qs_ot_p2m_diag 48 11.0 0.055 0.064 4.092 4.107
ot_diis_step 99 11.5 0.011 0.011 4.094 4.094
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.007 4.083
apply_single 110 13.6 0.000 0.000 4.006 4.083
pw_transfer 1331 11.6 0.065 0.075 3.955 3.985
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.849 3.883
dbcsr_complete_redistribute 325 12.2 0.424 0.464 2.701 3.842
density_rs2pw 110 9.6 0.004 0.005 3.131 3.637
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.612 3.612
grid_integrate_task_list 110 12.3 3.259 3.490 3.259 3.490
fft_wrap_pw1pw2_140 451 13.1 0.202 0.215 3.307 3.342
multiply_cannon_sync_h2d 24660 15.4 3.186 3.342 3.186 3.342
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.233 3.283
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.129 3.259
fft3d_ps 1111 14.6 1.097 1.136 3.142 3.168
multiply_cannon_metrocomm3 24660 15.4 0.035 0.036 1.531 3.166
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.057 3.069
cp_fm_diag_elpa_base 48 14.0 2.907 2.966 3.055 3.067
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.004 3.066
calculate_dm_sparse 110 9.5 0.001 0.001 2.927 2.961
wfi_extrapolate 11 7.9 0.001 0.001 2.960 2.960
make_images_data 4110 15.4 0.047 0.051 2.627 2.959
hybrid_alltoall_any 4261 16.3 0.119 0.455 2.277 2.949
rs_pw_transfer 902 11.9 0.010 0.011 2.409 2.931
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.774 2.886
mp_alltoall_i22 605 13.7 1.634 2.796 1.634 2.796
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.756 2.757
cp_fm_cholesky_invert 11 10.9 2.704 2.712 2.704 2.712
grid_collocate_task_list 110 9.6 2.182 2.455 2.182 2.455
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.411 2.442
calculate_first_density_matrix 1 7.0 0.001 0.003 2.411 2.413
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.293 2.295
potential_pw2rs 110 12.3 0.012 0.013 2.234 2.244
jit_kernel_multiply 12 15.7 1.131 2.227 1.131 2.227
mp_alltoall_d11v 2046 13.8 1.744 2.116 1.744 2.116
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.687 1.718
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.699 1.712
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.600 1.711
cp_fm_cholesky_decompose 22 10.9 1.636 1.684 1.636 1.684
mp_waitany 13376 13.8 1.064 1.579 1.064 1.579
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.551 1.565
multiply_cannon_metrocomm4 20550 15.4 0.058 0.061 0.838 1.528
acc_transpose_blocks 24660 15.4 0.104 0.106 1.502 1.519
mp_allgather_i34 2055 14.4 0.499 1.457 0.499 1.457
mp_irecv_dv 62702 16.1 0.737 1.449 0.737 1.449
mp_sum_l 6514 12.8 0.887 1.407 0.887 1.407
rs_pw_transfer_RS2PW_140 121 11.5 0.168 0.176 0.813 1.341
qs_env_update_s_mstruct 11 6.9 0.001 0.004 1.237 1.334
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=65.540000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=687.181818, yerr=7.517473
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 840.175616E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66424.
MP_Allreduce 9694 649.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.033 55.917 55.919
qs_mol_dyn_low 1 2.0 0.003 0.003 55.594 55.602
qs_forces 11 3.9 0.002 0.003 55.524 55.524
qs_energies 11 4.9 0.001 0.001 51.811 51.816
scf_env_do_scf 11 5.9 0.001 0.001 43.570 43.571
scf_env_do_scf_inner_loop 99 6.5 0.003 0.008 35.800 35.800
velocity_verlet 10 3.0 0.001 0.001 31.288 31.291
dbcsr_multiply_generic 2055 12.4 0.106 0.108 22.655 22.803
qs_scf_new_mos 99 7.5 0.001 0.001 20.589 20.650
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.588 20.650
ot_scf_mini 99 9.5 0.002 0.002 19.346 19.372
multiply_cannon 2055 13.4 0.244 0.256 17.291 18.632
multiply_cannon_loop 2055 14.4 0.320 0.332 15.936 16.282
rebuild_ks_matrix 110 8.3 0.000 0.000 12.013 12.054
qs_ks_build_kohn_sham_matrix 110 9.3 0.015 0.016 12.012 12.053
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.668 10.703
ot_mini 99 10.5 0.001 0.001 10.390 10.416
multiply_cannon_multrec 8220 15.4 3.242 4.482 7.480 8.615
init_scf_loop 11 6.9 0.000 0.000 7.722 7.723
mp_waitall_1 103326 16.6 5.920 7.497 5.920 7.497
qs_ot_get_derivative 99 11.5 0.001 0.001 6.579 6.610
sum_up_and_integrate 110 10.3 0.080 0.081 6.596 6.609
integrate_v_rspace 110 11.3 0.003 0.003 6.516 6.529
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.381 6.395
calculate_rho_elec 110 8.6 0.116 0.116 6.380 6.394
prepare_preconditioner 11 7.9 0.000 0.000 6.055 6.058
make_preconditioner 11 8.9 0.000 0.000 6.055 6.058
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.634 5.712
qs_ot_get_p 110 10.4 0.001 0.001 5.077 5.117
init_scf_run 11 5.9 0.000 0.002 5.103 5.104
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.103 5.103
dbcsr_mm_accdrv_process 17442 15.9 2.745 3.809 4.108 5.008
multiply_cannon_metrocomm3 8220 15.4 0.017 0.018 3.042 4.373
make_m2s 4110 13.4 0.038 0.039 4.095 4.338
make_images 4110 14.4 0.638 0.704 3.966 4.207
pw_transfer 1331 11.6 0.066 0.072 4.132 4.141
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 4.024 4.037
ot_diis_step 99 11.5 0.012 0.012 3.786 3.786
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.702 3.747
apply_single 110 13.6 0.000 0.000 3.702 3.747
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.693 3.697
grid_integrate_task_list 110 12.3 3.343 3.553 3.343 3.553
density_rs2pw 110 9.6 0.004 0.004 3.138 3.511
fft_wrap_pw1pw2_140 451 13.1 0.215 0.219 3.476 3.494
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.386 3.387
fft3d_ps 1111 14.6 1.146 1.168 3.267 3.277
multiply_cannon_sync_h2d 8220 15.4 2.914 3.017 2.914 3.017
cp_fm_cholesky_invert 11 10.9 2.960 2.964 2.960 2.964
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.823 2.831
cp_fm_diag_elpa_base 48 14.0 2.766 2.795 2.821 2.829
wfi_extrapolate 11 7.9 0.001 0.001 2.758 2.758
qs_energies_init_hamiltonians 11 5.9 0.002 0.004 2.681 2.682
make_images_data 4110 15.4 0.038 0.043 2.290 2.665
rs_pw_transfer 902 11.9 0.010 0.010 2.216 2.615
hybrid_alltoall_any 4261 16.3 0.201 0.864 2.211 2.556
grid_collocate_task_list 110 9.6 2.271 2.536 2.271 2.536
calculate_dm_sparse 110 9.5 0.001 0.001 2.485 2.531
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.491 2.492
calculate_first_density_matrix 1 7.0 0.000 0.000 2.248 2.249
potential_pw2rs 110 12.3 0.015 0.016 2.154 2.162
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.129 2.149
mp_alltoall_d11v 2046 13.8 1.610 2.017 1.610 2.017
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.773 1.992
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 1.884 1.900
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.782 1.798
jit_kernel_multiply 10 15.9 1.049 1.768 1.049 1.768
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.680 1.687
cp_fm_cholesky_decompose 22 10.9 1.669 1.684 1.669 1.684
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.512 1.640
dbcsr_complete_redistribute 325 12.2 0.561 0.590 1.471 1.556
mp_allgather_i34 2055 14.4 0.540 1.545 0.540 1.545
mp_waitany 9240 13.8 1.056 1.462 1.056 1.462
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.445 1.459
qs_create_task_list 11 7.9 0.000 0.000 1.223 1.328
generate_qs_task_list 11 8.9 0.379 0.446 1.223 1.328
rs_gather_matrices 110 12.3 0.327 0.371 0.931 1.312
multiply_cannon_metrocomm4 6165 15.4 0.018 0.020 0.477 1.284
mp_irecv_dv 24056 15.7 0.452 1.241 0.452 1.241
rs_pw_transfer_RS2PW_140 121 11.5 0.162 0.166 0.785 1.182
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.143 1.168
multiply_cannon_metrocomm1 8220 15.4 0.020 0.021 0.738 1.143
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=55.919000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=788.000000, yerr=12.067989
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 1.364169E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67104.
MP_Allreduce 9672 819.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.020 0.045 89.885 89.886
qs_mol_dyn_low 1 2.0 0.003 0.003 89.561 89.567
qs_forces 11 3.9 0.003 0.003 89.485 89.487
qs_energies 11 4.9 0.002 0.002 85.245 85.247
scf_env_do_scf 11 5.9 0.001 0.001 75.148 75.149
velocity_verlet 10 3.0 0.001 0.001 57.489 57.495
scf_env_do_scf_inner_loop 99 6.5 0.003 0.009 45.779 45.780
init_scf_loop 11 6.9 0.000 0.000 29.293 29.296
dbcsr_multiply_generic 2055 12.4 0.119 0.122 28.837 28.857
prepare_preconditioner 11 7.9 0.000 0.000 27.218 27.226
make_preconditioner 11 8.9 0.000 0.000 27.218 27.226
qs_scf_new_mos 99 7.5 0.001 0.001 27.032 27.080
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.032 27.080
make_full_inverse_cholesky 11 9.9 0.000 0.000 21.243 26.679
ot_scf_mini 99 9.5 0.002 0.002 25.249 25.284
multiply_cannon 2055 13.4 0.350 0.385 21.693 22.691
multiply_cannon_loop 2055 14.4 0.341 0.345 19.857 20.296
cp_fm_upper_to_full 70 14.2 12.992 18.680 12.992 18.680
rebuild_ks_matrix 110 8.3 0.001 0.001 14.472 14.484
qs_ks_build_kohn_sham_matrix 110 9.3 0.017 0.019 14.471 14.483
ot_mini 99 10.5 0.001 0.001 13.778 13.808
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.098 13.109
dbcsr_complete_redistribute 325 12.2 1.017 1.040 7.612 10.992
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.561 9.938
multiply_cannon_multrec 8220 15.4 4.369 4.601 9.559 9.638
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.959 9.295
qs_ot_get_derivative 99 11.5 0.001 0.001 9.172 9.204
mp_alltoall_i22 605 13.7 5.572 8.942 5.572 8.942
mp_waitall_1 84994 16.7 7.754 8.481 7.754 8.481
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.946 7.983
calculate_rho_elec 110 8.6 0.227 0.228 7.945 7.982
sum_up_and_integrate 110 10.3 0.150 0.151 7.531 7.544
integrate_v_rspace 110 11.3 0.004 0.004 7.381 7.394
qs_ot_get_p 110 10.4 0.001 0.001 6.212 6.253
make_m2s 4110 13.4 0.043 0.043 5.313 5.850
init_scf_run 11 5.9 0.000 0.001 5.781 5.781
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.781 5.781
make_images 4110 14.4 0.883 0.931 5.125 5.661
cp_fm_cholesky_invert 11 10.9 5.631 5.638 5.631 5.638
pw_transfer 1331 11.6 0.075 0.076 5.446 5.457
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 4.949 5.397
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 5.329 5.340
dbcsr_mm_accdrv_process 11614 15.7 3.274 3.633 5.048 5.272
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.724 5.185
apply_single 110 13.6 0.000 0.000 4.724 5.185
fft_wrap_pw1pw2_140 451 13.1 0.222 0.224 4.597 4.608
ot_diis_step 99 11.5 0.015 0.017 4.581 4.581
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.574 4.580
fft3d_ps 1111 14.6 1.294 1.302 4.523 4.533
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.116 4.117
density_rs2pw 110 9.6 0.004 0.004 3.956 3.971
multiply_cannon_sync_h2d 8220 15.4 3.948 3.953 3.948 3.953
grid_integrate_task_list 110 12.3 3.672 3.729 3.672 3.729
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.695 3.696
hybrid_alltoall_any 4261 16.3 0.256 0.553 2.911 3.600
make_images_data 4110 15.4 0.042 0.045 2.875 3.541
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.476 3.476
cp_fm_diag_elpa_base 48 14.0 2.918 3.134 3.474 3.474
wfi_extrapolate 11 7.9 0.001 0.001 3.424 3.424
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.925 3.351
calculate_dm_sparse 110 9.5 0.001 0.001 3.156 3.177
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.877 2.882
grid_collocate_task_list 110 9.6 2.640 2.658 2.640 2.658
potential_pw2rs 110 12.3 0.021 0.022 2.603 2.609
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.453 2.476
rs_pw_transfer 902 11.9 0.011 0.011 2.371 2.411
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.366 2.376
qs_env_update_s_mstruct 11 6.9 0.001 0.001 2.209 2.260
calculate_first_density_matrix 1 7.0 0.000 0.000 2.230 2.230
mp_alltoall_d11v 2046 13.8 2.063 2.167 2.063 2.167
cp_fm_cholesky_decompose 22 10.9 2.134 2.159 2.134 2.159
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.087 2.158
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.981 1.983
qs_create_task_list 11 7.9 0.000 0.000 1.900 1.947
generate_qs_task_list 11 8.9 0.735 0.790 1.900 1.947
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.788 1.803
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=89.886000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1212.090909, yerr=66.090096
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 631.484416E+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 57910.
MP_Allreduce 10979 803.
MP_Sync 87
MP_Alltoall 2226 2520958.
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 0.021 0.050 208.662 208.673
qs_mol_dyn_low 1 2.0 0.003 0.003 208.281 208.297
qs_forces 11 3.9 0.004 0.005 208.183 208.186
qs_energies 11 4.9 0.002 0.003 202.556 202.575
scf_env_do_scf 11 5.9 0.001 0.001 185.987 185.991
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 165.125 165.128
velocity_verlet 10 3.0 0.001 0.001 125.128 125.130
qs_scf_new_mos 117 7.6 0.001 0.001 124.227 124.525
qs_scf_loop_do_ot 117 8.6 0.001 0.001 124.226 124.524
dbcsr_multiply_generic 2507 12.6 0.175 0.183 123.804 124.393
ot_scf_mini 117 9.6 0.003 0.003 117.586 117.826
multiply_cannon 2507 13.6 0.236 0.246 100.620 102.199
multiply_cannon_loop 2507 14.6 2.053 2.093 98.391 99.785
ot_mini 117 10.6 0.001 0.001 65.400 65.702
multiply_cannon_multrec 60168 15.6 33.176 35.091 41.405 43.461
qs_ot_get_derivative 117 11.6 0.001 0.001 40.873 41.137
rebuild_ks_matrix 128 8.3 0.001 0.001 34.143 34.456
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.020 34.143 34.455
mp_waitall_1 267128 16.5 28.218 31.680 28.218 31.680
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.688 30.980
qs_ot_get_p 128 10.4 0.001 0.001 30.313 30.559
multiply_cannon_sync_h2d 60168 15.6 27.517 28.757 27.517 28.757
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.137 24.841
apply_single 128 13.6 0.001 0.001 24.137 24.841
ot_diis_step 117 11.6 0.008 0.008 24.271 24.273
qs_ot_p2m_diag 83 11.4 0.079 0.091 23.823 23.882
cp_dbcsr_syevd 83 12.4 0.004 0.005 21.251 21.252
init_scf_loop 11 6.9 0.000 0.000 20.782 20.783
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 19.572 19.750
cp_fm_diag_elpa 83 13.4 0.000 0.000 18.186 18.226
cp_fm_diag_elpa_base 83 14.4 18.111 18.175 18.182 18.223
multiply_cannon_metrocomm3 60168 15.6 0.112 0.119 15.483 17.835
prepare_preconditioner 11 7.9 0.000 0.000 16.145 16.191
make_preconditioner 11 8.9 0.000 0.000 16.145 16.191
make_full_inverse_cholesky 11 9.9 0.000 0.000 15.350 15.544
sum_up_and_integrate 128 10.3 0.089 0.107 14.650 14.663
integrate_v_rspace 128 11.3 0.004 0.005 14.560 14.579
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.218 14.365
calculate_rho_elec 128 8.7 0.046 0.065 14.217 14.364
make_m2s 5014 13.6 0.104 0.113 13.842 14.254
make_images 5014 14.6 0.401 0.417 13.661 14.083
init_scf_run 11 5.9 0.000 0.005 12.382 12.383
scf_env_initial_rho_setup 11 6.9 0.003 0.004 12.382 12.383
density_rs2pw 128 9.7 0.006 0.007 7.517 10.601
mp_sum_l 7870 13.0 8.073 9.350 8.073 9.350
cp_fm_cholesky_invert 11 10.9 9.252 9.259 9.252 9.259
wfi_extrapolate 11 7.9 0.001 0.001 9.200 9.201
rs_pw_transfer 1046 11.9 0.017 0.019 6.064 9.119
pw_transfer 1547 11.6 0.077 0.100 8.290 8.551
calculate_dm_sparse 128 9.5 0.001 0.001 8.448 8.545
fft_wrap_pw1pw2 1291 12.7 0.011 0.013 8.084 8.349
dbcsr_mm_accdrv_process 124484 16.2 3.154 3.257 7.789 8.267
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 7.923 8.068
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.851 7.970
multiply_cannon_metrocomm1 60168 15.6 0.087 0.092 6.058 7.919
make_images_data 5014 15.6 0.066 0.072 6.796 7.812
grid_integrate_task_list 128 12.3 7.108 7.540 7.108 7.540
fft_wrap_pw1pw2_140 523 13.2 0.444 0.512 6.973 7.299
hybrid_alltoall_any 5200 16.5 0.293 2.259 5.942 7.190
fft3d_ps 1291 14.7 2.104 2.889 6.870 7.061
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.641 6.649
mp_alltoall_d11v 2415 14.1 4.386 5.905 4.386 5.905
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.785 5.903
mp_waitany 16020 13.9 2.705 5.884 2.705 5.884
grid_collocate_task_list 128 9.7 4.561 5.801 4.561 5.801
rs_pw_transfer_RS2PW_140 139 11.5 0.280 0.299 2.118 5.214
potential_pw2rs 128 12.3 0.009 0.011 4.947 4.992
cp_fm_cholesky_decompose 22 10.9 4.667 4.680 4.667 4.680
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=208.673000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=598.909091, yerr=7.051171
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 828.563456E+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 58363.
MP_Allreduce 10978 967.
MP_Sync 87
MP_Alltoall 1969 5618984.
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 0.019 0.041 192.614 192.616
qs_mol_dyn_low 1 2.0 0.003 0.003 192.189 192.202
qs_forces 11 3.9 0.004 0.004 192.099 192.102
qs_energies 11 4.9 0.002 0.002 185.224 185.233
scf_env_do_scf 11 5.9 0.001 0.001 168.914 168.924
scf_env_do_scf_inner_loop 117 6.6 0.003 0.010 135.503 135.506
velocity_verlet 10 3.0 0.001 0.001 121.430 121.432
dbcsr_multiply_generic 2507 12.6 0.185 0.191 97.167 98.444
qs_scf_new_mos 117 7.6 0.001 0.001 94.915 95.412
qs_scf_loop_do_ot 117 8.6 0.001 0.001 94.914 95.411
ot_scf_mini 117 9.6 0.004 0.004 90.102 90.630
multiply_cannon 2507 13.6 0.478 0.533 76.966 81.311
multiply_cannon_loop 2507 14.6 1.251 1.295 73.981 76.739
ot_mini 117 10.6 0.001 0.001 49.845 50.461
mp_waitall_1 214728 16.6 24.738 37.608 24.738 37.608
multiply_cannon_multrec 30084 15.6 21.888 26.115 31.514 36.137
rebuild_ks_matrix 128 8.3 0.001 0.001 33.618 34.258
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.021 33.617 34.258
init_scf_loop 11 6.9 0.000 0.000 33.315 33.316
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.246 30.831
prepare_preconditioner 11 7.9 0.000 0.000 28.876 28.959
make_preconditioner 11 8.9 0.000 0.000 28.876 28.959
multiply_cannon_metrocomm3 30084 15.6 0.092 0.099 15.802 28.487
qs_ot_get_derivative 117 11.6 0.001 0.002 27.889 28.430
make_full_inverse_cholesky 11 9.9 0.000 0.000 27.572 28.134
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.010 23.164
apply_single 128 13.6 0.001 0.001 22.010 23.163
qs_ot_get_p 128 10.4 0.001 0.001 22.097 22.890
multiply_cannon_sync_h2d 30084 15.6 19.199 21.893 19.199 21.893
ot_diis_step 117 11.6 0.014 0.015 21.774 21.776
qs_ot_p2m_diag 83 11.4 0.187 0.215 17.256 17.292
cp_fm_cholesky_invert 11 10.9 16.920 16.932 16.920 16.932
cp_dbcsr_syevd 83 12.4 0.005 0.006 16.102 16.102
make_m2s 5014 13.6 0.089 0.094 14.400 15.659
make_images 5014 14.6 1.157 1.340 14.193 15.454
sum_up_and_integrate 128 10.3 0.116 0.134 15.030 15.060
integrate_v_rspace 128 11.3 0.004 0.004 14.914 14.947
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.366 14.398
calculate_rho_elec 128 8.7 0.088 0.106 14.366 14.397
cp_fm_diag_elpa 83 13.4 0.000 0.000 12.867 12.895
cp_fm_diag_elpa_base 83 14.4 12.611 12.698 12.860 12.891
init_scf_run 11 5.9 0.000 0.002 11.507 11.508
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.506 11.508
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 10.962 11.400
multiply_cannon_metrocomm4 27577 15.6 0.098 0.114 3.786 10.814
mp_irecv_dv 69486 16.3 3.587 10.420 3.587 10.420
make_images_data 5014 15.6 0.064 0.071 8.773 10.312
density_rs2pw 128 9.7 0.006 0.007 7.722 10.094
dbcsr_mm_accdrv_process 62242 16.2 4.507 5.278 9.084 9.723
hybrid_alltoall_any 5200 16.5 0.343 1.505 7.449 9.507
pw_transfer 1547 11.6 0.087 0.113 9.250 9.322
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 9.024 9.102
wfi_extrapolate 11 7.9 0.001 0.001 8.424 8.424
rs_pw_transfer 1046 11.9 0.014 0.017 5.892 8.279
fft_wrap_pw1pw2_140 523 13.2 0.473 0.529 7.935 8.020
grid_integrate_task_list 128 12.3 7.173 7.611 7.173 7.611
fft3d_ps 1291 14.7 2.756 2.924 7.390 7.425
cp_fm_cholesky_decompose 22 10.9 7.026 7.106 7.026 7.106
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.279 7.057
calculate_dm_sparse 128 9.5 0.001 0.001 6.478 6.641
mp_sum_l 7870 13.0 4.026 6.336 4.026 6.336
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.072 6.081
grid_collocate_task_list 128 9.7 4.704 5.877 4.704 5.877
mp_alltoall_d11v 2415 14.1 4.199 5.521 4.199 5.521
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.358 5.428
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.231 5.381
potential_pw2rs 128 12.3 0.014 0.016 5.259 5.281
mp_waitany 11748 13.9 2.485 4.929 2.485 4.929
mp_allgather_i34 2507 14.6 1.555 4.683 1.555 4.683
rs_pw_transfer_RS2PW_140 139 11.5 0.352 0.392 2.087 4.469
dbcsr_complete_redistribute 395 12.7 0.771 0.857 3.247 4.110
mp_sum_d 4461 12.1 2.652 4.030 2.652 4.030
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=192.616000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=789.818182, yerr=3.737237
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528903135232 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 2.928533E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 3984192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1695.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 942.882816E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1042912
MPI messages size (bytes):
total size 2.716210E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.604448E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 281856 36943429632
131072 < size <= 4194304 660064 996105256960
4194304 < size <= 16777216 65632 931531265168
16777216 < size 28672 751619276800
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58214.
MP_Allreduce 11002 1006.
MP_Sync 87
MP_Alltoall 1712 9388896.
MP_SendRecv 7936 75008.
MP_ISendRecv 7936 75008.
MP_Wait 21820
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.071 177.436 177.437
qs_mol_dyn_low 1 2.0 0.003 0.005 176.954 176.966
qs_forces 11 3.9 0.004 0.004 176.776 176.784
qs_energies 11 4.9 0.002 0.006 170.129 170.137
scf_env_do_scf 11 5.9 0.001 0.003 154.628 154.629
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 119.401 119.401
velocity_verlet 10 3.0 0.001 0.001 114.207 114.211
dbcsr_multiply_generic 2507 12.6 0.180 0.183 81.474 82.641
qs_scf_new_mos 117 7.6 0.001 0.001 80.734 81.082
qs_scf_loop_do_ot 117 8.6 0.001 0.001 80.733 81.081
ot_scf_mini 117 9.6 0.004 0.004 76.599 76.990
multiply_cannon 2507 13.6 0.496 0.517 61.614 65.951
multiply_cannon_loop 2507 14.6 0.854 0.890 58.732 61.430
ot_mini 117 10.6 0.001 0.001 42.549 42.933
init_scf_loop 11 6.9 0.001 0.004 35.127 35.129
mp_waitall_1 170520 16.6 25.422 34.916 25.422 34.916
rebuild_ks_matrix 128 8.3 0.001 0.001 31.089 31.641
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.021 31.088 31.640
prepare_preconditioner 11 7.9 0.000 0.001 31.042 31.093
make_preconditioner 11 8.9 0.001 0.007 31.042 31.093
make_full_inverse_cholesky 11 9.9 0.000 0.001 28.694 30.081
qs_ks_update_qs_env 128 7.6 0.001 0.001 28.001 28.512
multiply_cannon_multrec 20056 15.6 13.400 17.206 22.078 25.809
multiply_cannon_metrocomm3 20056 15.6 0.059 0.065 15.580 24.903
qs_ot_get_derivative 117 11.6 0.001 0.002 22.656 23.049
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 20.024 21.014
apply_single 128 13.6 0.001 0.001 20.024 21.013
qs_ot_get_p 128 10.4 0.001 0.001 19.430 19.963
ot_diis_step 117 11.6 0.018 0.018 19.789 19.789
multiply_cannon_sync_h2d 20056 15.6 14.240 16.317 14.240 16.317
make_m2s 5014 13.6 0.080 0.085 14.872 15.704
make_images 5014 14.6 1.172 1.267 14.641 15.468
qs_ot_p2m_diag 83 11.4 0.265 0.272 15.224 15.237
sum_up_and_integrate 128 10.3 0.133 0.145 14.881 14.911
integrate_v_rspace 128 11.3 0.004 0.004 14.748 14.776
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.526 14.553
calculate_rho_elec 128 8.7 0.132 0.147 14.525 14.553
cp_fm_cholesky_invert 11 10.9 14.541 14.550 14.541 14.550
cp_dbcsr_syevd 83 12.4 0.005 0.005 14.208 14.210
cp_fm_diag_elpa 83 13.4 0.000 0.000 10.983 11.003
cp_fm_diag_elpa_base 83 14.4 10.573 10.733 10.980 10.999
make_images_data 5014 15.6 0.061 0.069 9.285 10.528
init_scf_run 11 5.9 0.000 0.001 10.490 10.491
scf_env_initial_rho_setup 11 6.9 0.001 0.002 10.490 10.490
hybrid_alltoall_any 5200 16.5 0.434 1.996 7.976 9.829
pw_transfer 1547 11.6 0.087 0.107 9.279 9.394
density_rs2pw 128 9.7 0.006 0.006 7.401 9.354
multiply_cannon_metrocomm4 17549 15.6 0.063 0.071 3.486 9.350
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 9.053 9.173
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 8.884 9.170
mp_irecv_dv 50230 16.2 3.362 9.095 3.362 9.095
dbcsr_mm_accdrv_process 41502 16.2 4.284 4.950 8.132 8.242
fft_wrap_pw1pw2_140 523 13.2 0.478 0.521 8.006 8.135
grid_integrate_task_list 128 12.3 7.287 7.763 7.287 7.763
wfi_extrapolate 11 7.9 0.001 0.001 7.474 7.474
cp_fm_cholesky_decompose 22 10.9 7.378 7.409 7.378 7.409
fft3d_ps 1291 14.7 2.673 2.890 7.319 7.396
cp_fm_upper_to_full 105 14.8 5.724 7.234 5.724 7.234
rs_pw_transfer 1046 11.9 0.014 0.015 5.250 7.201
dbcsr_complete_redistribute 395 12.7 1.170 1.221 4.631 6.409
calculate_dm_sparse 128 9.5 0.001 0.001 5.729 5.805
grid_collocate_task_list 128 9.7 4.903 5.780 4.903 5.780
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.445 5.449
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.546 5.236
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.433 5.204
mp_alltoall_d11v 2415 14.1 4.381 5.103 4.381 5.103
potential_pw2rs 128 12.3 0.020 0.023 4.942 4.964
mp_sum_l 7870 13.0 3.189 4.727 3.189 4.727
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.524 4.680
mp_allgather_i34 2507 14.6 1.404 4.555 1.404 4.555
mp_waitany 11748 13.9 2.340 4.408 2.340 4.408
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.050 4.084
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.327 4.057
mp_alltoall_i22 716 14.1 1.918 3.863 1.918 3.863
rs_pw_transfer_RS2PW_140 139 11.5 0.330 0.348 1.909 3.852
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 3.772 3.773
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.599 3.641
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=177.437000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=889.181818, yerr=10.070824
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/18/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 4.353788E+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 5977344 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.131545E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1143192
MPI messages size (bytes):
total size 2.023815E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.770319E+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 319024 36042702848
131072 < size <= 4194304 715736 785529176064
4194304 < size <= 16777216 70320 665379241840
16777216 < size 30720 536870912000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58354.
MP_Allreduce 10977 1091.
MP_Sync 87
MP_Alltoall 1712 12503084.
MP_SendRecv 5888 75008.
MP_ISendRecv 5888 75008.
MP_Wait 22442
MP_ISend 14952 244818.
MP_IRecv 14952 244818.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.025 0.044 189.421 189.423
qs_mol_dyn_low 1 2.0 0.003 0.003 188.999 189.008
qs_forces 11 3.9 0.004 0.004 188.886 188.894
qs_energies 11 4.9 0.003 0.007 181.662 181.674
scf_env_do_scf 11 5.9 0.001 0.002 165.027 165.034
velocity_verlet 10 3.0 0.001 0.001 125.101 125.104
scf_env_do_scf_inner_loop 117 6.6 0.004 0.010 118.638 118.639
qs_scf_new_mos 117 7.6 0.001 0.001 80.906 81.221
qs_scf_loop_do_ot 117 8.6 0.001 0.001 80.905 81.220
dbcsr_multiply_generic 2507 12.6 0.187 0.192 78.770 79.425
ot_scf_mini 117 9.6 0.003 0.004 76.419 76.686
multiply_cannon 2507 13.6 0.556 0.596 54.414 56.977
multiply_cannon_loop 2507 14.6 1.182 1.208 50.798 52.514
init_scf_loop 11 6.9 0.001 0.007 46.265 46.267
ot_mini 117 10.6 0.001 0.001 42.127 42.394
prepare_preconditioner 11 7.9 0.000 0.000 42.136 42.156
make_preconditioner 11 8.9 0.000 0.003 42.136 42.156
make_full_inverse_cholesky 11 9.9 0.000 0.001 35.895 40.816
multiply_cannon_multrec 30084 15.6 14.147 19.325 26.052 30.948
rebuild_ks_matrix 128 8.3 0.001 0.001 30.039 30.320
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.021 30.038 30.320
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.064 27.306
mp_waitall_1 147882 16.7 16.955 26.717 16.955 26.717
qs_ot_get_derivative 117 11.6 0.001 0.002 22.517 22.788
make_m2s 5014 13.6 0.096 0.101 20.019 21.026
qs_ot_get_p 128 10.4 0.001 0.001 20.610 20.929
make_images 5014 14.6 1.939 2.228 19.713 20.720
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 19.007 19.497
apply_single 128 13.6 0.001 0.001 19.006 19.497
ot_diis_step 117 11.6 0.017 0.019 19.486 19.488
cp_fm_upper_to_full 105 14.8 11.339 16.670 11.339 16.670
qs_ot_p2m_diag 83 11.4 0.342 0.390 16.485 16.537
cp_fm_cholesky_invert 11 10.9 16.199 16.207 16.199 16.207
cp_dbcsr_syevd 83 12.4 0.005 0.005 15.196 15.197
sum_up_and_integrate 128 10.3 0.140 0.152 14.899 14.928
multiply_cannon_metrocomm3 30084 15.6 0.046 0.049 6.335 14.920
integrate_v_rspace 128 11.3 0.004 0.004 14.759 14.791
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.705 14.743
calculate_rho_elec 128 8.7 0.175 0.190 14.705 14.742
multiply_cannon_sync_h2d 30084 15.6 11.734 12.932 11.734 12.932
dbcsr_complete_redistribute 395 12.7 1.504 1.623 8.817 12.564
make_images_data 5014 15.6 0.065 0.069 10.719 12.354
cp_fm_diag_elpa 83 13.4 0.000 0.001 12.012 12.022
cp_fm_diag_elpa_base 83 14.4 11.043 11.370 12.004 12.015
dbcsr_mm_accdrv_process 62264 16.2 7.302 8.327 11.481 11.967
hybrid_alltoall_any 5200 16.5 0.524 2.205 9.735 11.393
copy_fm_to_dbcsr 209 11.7 0.001 0.002 7.432 11.194
init_scf_run 11 5.9 0.000 0.001 10.722 10.723
scf_env_initial_rho_setup 11 6.9 0.001 0.002 10.721 10.723
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.219 9.857
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.341 9.534
pw_transfer 1547 11.6 0.089 0.107 9.482 9.522
fft_wrap_pw1pw2 1291 12.7 0.010 0.012 9.254 9.300
mp_alltoall_i22 716 14.1 5.487 9.198 5.487 9.198
density_rs2pw 128 9.7 0.006 0.006 7.167 8.413
fft_wrap_pw1pw2_140 523 13.2 0.482 0.492 8.180 8.251
grid_integrate_task_list 128 12.3 7.536 7.976 7.536 7.976
cp_fm_cholesky_decompose 22 10.9 7.617 7.714 7.617 7.714
wfi_extrapolate 11 7.9 0.001 0.001 7.656 7.656
fft3d_ps 1291 14.7 2.769 2.829 7.478 7.522
multiply_cannon_metrocomm4 25070 15.6 0.077 0.087 2.784 7.211
mp_irecv_dv 76098 16.2 2.636 6.944 2.636 6.944
calculate_dm_sparse 128 9.5 0.001 0.001 6.141 6.207
rs_pw_transfer 1046 11.9 0.013 0.014 4.768 6.045
grid_collocate_task_list 128 9.7 5.078 5.754 5.078 5.754
mp_alltoall_d11v 2415 14.1 4.738 5.645 4.738 5.645
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.408 5.452
potential_pw2rs 128 12.3 0.022 0.023 4.720 4.744
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 4.487 4.488
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.339 4.419
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.345 4.418
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.177 4.244
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=189.423000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1068.454545, yerr=18.724625
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/19/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 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 11444706349056 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019187724288 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019187724288 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 92.796577E+12 0.0% 0.0% 100.0%
flops max/rank 5.820059E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500544 0.0% 0.0% 100.0%
number of processed stacks 1944496 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3448.5
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.510691E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 238560
MPI messages size (bytes):
total size 1.321104E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.537828E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 112800 59139686400
4194304 < size <= 16777216 104112 545846722560
16777216 < size 20064 716108613552
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8852 52.
MP_Alltoall 9584 804353.
MP_ISend 39716 2104723.
MP_IRecv 39716 2103824.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3949 58975.
MP_Allreduce 10870 1179.
MP_Sync 86
MP_Alltoall 1700 18828186.
MP_SendRecv 3810 122880.
MP_ISendRecv 3810 122880.
MP_Wait 16000
MP_ISend 10600 423612.
MP_IRecv 10600 423612.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.035 177.502 177.504
qs_mol_dyn_low 1 2.0 0.003 0.003 177.092 177.105
qs_forces 11 3.9 0.004 0.004 176.871 176.875
qs_energies 11 4.9 0.002 0.002 169.213 169.220
scf_env_do_scf 11 5.9 0.001 0.001 151.873 151.882
scf_env_do_scf_inner_loop 116 6.6 0.003 0.009 115.475 115.475
velocity_verlet 10 3.0 0.001 0.001 115.187 115.198
qs_scf_new_mos 116 7.6 0.001 0.001 76.645 76.747
qs_scf_loop_do_ot 116 8.6 0.001 0.001 76.644 76.746
dbcsr_multiply_generic 2485 12.5 0.179 0.185 73.824 74.311
ot_scf_mini 116 9.6 0.003 0.004 72.274 72.326
multiply_cannon 2485 13.5 0.575 0.613 54.460 58.425
multiply_cannon_loop 2485 14.5 0.440 0.450 49.710 50.649
ot_mini 116 10.6 0.001 0.001 39.548 39.610
init_scf_loop 11 6.9 0.000 0.000 36.247 36.248
mp_waitall_1 124680 16.7 25.626 32.962 25.626 32.962
prepare_preconditioner 11 7.9 0.000 0.000 32.260 32.289
make_preconditioner 11 8.9 0.000 0.000 32.260 32.289
make_full_inverse_cholesky 11 9.9 0.000 0.000 30.047 30.320
rebuild_ks_matrix 127 8.3 0.001 0.001 30.004 30.074
qs_ks_build_kohn_sham_matrix 127 9.3 0.019 0.021 30.003 30.073
qs_ks_update_qs_env 127 7.6 0.001 0.001 27.226 27.296
multiply_cannon_multrec 9940 15.5 10.378 14.368 17.763 20.558
qs_ot_get_p 127 10.4 0.001 0.001 19.899 19.994
qs_ot_get_derivative 116 11.6 0.001 0.002 19.754 19.802
ot_diis_step 116 11.6 0.019 0.020 19.720 19.720
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.330 19.607
apply_single 127 13.6 0.001 0.001 19.330 19.606
multiply_cannon_metrocomm3 9940 15.5 0.022 0.023 12.551 19.223
cp_fm_cholesky_invert 11 10.9 18.436 18.442 18.436 18.442
make_m2s 4970 13.5 0.065 0.070 15.699 17.929
make_images 4970 14.5 2.257 2.811 15.399 17.622
qs_ot_p2m_diag 82 11.4 0.489 0.495 16.044 16.057
qs_rho_update_rho_low 127 7.7 0.001 0.001 15.556 15.586
calculate_rho_elec 127 8.7 0.257 0.267 15.555 15.585
sum_up_and_integrate 127 10.3 0.180 0.191 15.202 15.245
integrate_v_rspace 127 11.3 0.004 0.004 15.021 15.073
cp_dbcsr_syevd 82 12.4 0.005 0.005 14.874 14.876
make_images_data 4970 15.5 0.052 0.059 9.454 12.054
multiply_cannon_sync_h2d 9940 15.5 11.512 11.992 11.512 11.992
cp_fm_diag_elpa 82 13.4 0.000 0.000 11.752 11.763
cp_fm_diag_elpa_base 82 14.4 11.509 11.593 11.747 11.758
hybrid_alltoall_any 5155 16.4 0.809 3.556 9.282 11.515
init_scf_run 11 5.9 0.000 0.001 10.517 10.517
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.517 10.517
pw_transfer 1535 11.6 0.085 0.093 10.003 10.044
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 9.781 9.830
fft_wrap_pw1pw2_140 519 13.2 0.492 0.512 8.553 8.606
grid_integrate_task_list 127 12.3 7.680 8.237 7.680 8.237
cp_fm_cholesky_decompose 22 10.9 8.079 8.171 8.079 8.171
density_rs2pw 127 9.7 0.005 0.006 7.175 8.042
fft3d_ps 1281 14.7 2.694 2.772 7.955 7.980
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 7.925 7.965
dbcsr_mm_accdrv_process 20590 16.1 2.529 3.390 7.015 7.653
wfi_extrapolate 11 7.9 0.001 0.001 7.478 7.478
mp_allgather_i34 2485 14.5 2.965 6.903 2.965 6.903
multiply_cannon_metrocomm1 9940 15.5 0.028 0.028 4.295 6.867
mp_alltoall_d11v 2401 14.1 5.204 6.340 5.204 6.340
calculate_dm_sparse 127 9.5 0.001 0.001 6.130 6.191
grid_collocate_task_list 127 9.7 5.352 6.034 5.352 6.034
dbcsr_complete_redistribute 393 12.7 2.119 2.212 5.357 5.704
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.343 5.344
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.138 5.153
rs_pw_transfer 1038 11.9 0.012 0.013 4.264 5.142
potential_pw2rs 127 12.3 0.026 0.028 4.695 4.712
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.064 4.096
multiply_cannon_metrocomm4 7455 15.5 0.025 0.027 1.838 3.944
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.563 3.912
mp_irecv_dv 28618 15.9 1.801 3.877 1.801 3.877
copy_fm_to_dbcsr 208 11.6 0.002 0.002 3.504 3.819
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.733 3.770
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.678 3.683
copy_dbcsr_to_fm 185 11.7 0.004 0.004 3.594 3.672
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=177.504000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1406.363636, yerr=43.220939
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410020794368 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 1957875093504 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963540389888 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714619674624 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 11444697722880 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019185291264 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019185291264 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624869085184 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.815908E+12 0.0% 0.0% 100.0%
flops max/rank 11.608822E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705793568 0.0% 0.0% 100.0%
number of processed stacks 1951120 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3436.9
marketing flops 143.527070E+12
-------------------------------------------------------------------------------
# multiplications 2494
max memory usage/rank 3.021652E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 99760
MPI messages size (bytes):
total size 1.127724E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.304370E+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 45128 35047604224
4194304 < size <= 16777216 43984 376564613120
16777216 < size 10032 716108641296
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3956 59805.
MP_Allreduce 10887 1517.
MP_Sync 87
MP_Alltoall 1700 36954490.
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 0.052 0.119 300.686 300.715
qs_mol_dyn_low 1 2.0 0.003 0.003 300.078 300.085
qs_forces 11 3.9 0.004 0.004 299.974 299.977
qs_energies 11 4.9 0.002 0.002 290.443 290.455
scf_env_do_scf 11 5.9 0.001 0.001 267.736 267.747
velocity_verlet 10 3.0 0.001 0.002 216.024 216.032
scf_env_do_scf_inner_loop 116 6.6 0.004 0.009 141.867 141.868
init_scf_loop 11 6.9 0.000 0.000 125.607 125.610
prepare_preconditioner 11 7.9 0.000 0.000 120.452 120.475
make_preconditioner 11 8.9 0.000 0.000 120.452 120.475
make_full_inverse_cholesky 11 9.9 0.000 0.000 96.394 117.627
qs_scf_new_mos 116 7.6 0.001 0.001 90.791 90.877
qs_scf_loop_do_ot 116 8.6 0.001 0.001 90.790 90.876
ot_scf_mini 116 9.6 0.003 0.004 86.080 86.159
dbcsr_multiply_generic 2494 12.5 0.211 0.218 82.966 83.441
cp_fm_upper_to_full 105 14.8 53.575 76.642 53.575 76.642
multiply_cannon 2494 13.5 0.696 0.743 58.558 58.953
multiply_cannon_loop 2494 14.5 0.468 0.480 55.042 56.246
ot_mini 116 10.6 0.001 0.001 44.015 44.097
dbcsr_complete_redistribute 395 12.7 3.963 4.003 29.711 42.623
copy_fm_to_dbcsr 209 11.7 0.002 0.002 26.321 39.204
rebuild_ks_matrix 127 8.3 0.001 0.001 38.644 38.669
qs_ks_build_kohn_sham_matrix 127 9.3 0.048 0.048 38.644 38.669
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 24.013 36.735
qs_ks_update_qs_env 127 7.6 0.001 0.001 35.542 35.568
mp_alltoall_i22 714 14.1 21.832 34.916 21.832 34.916
cp_fm_cholesky_invert 11 10.9 33.665 33.671 33.665 33.671
mp_waitall_1 103128 16.8 27.909 31.519 27.909 31.519
qs_ot_get_p 127 10.4 0.001 0.001 26.640 26.712
qs_ot_get_derivative 116 11.6 0.002 0.002 24.134 24.217
qs_ot_p2m_diag 83 11.4 0.878 0.883 22.576 22.602
cp_dbcsr_syevd 83 12.4 0.006 0.006 20.863 20.864
qs_rho_update_rho_low 127 7.7 0.001 0.001 20.768 20.793
calculate_rho_elec 127 8.7 0.479 0.480 20.767 20.793
make_m2s 4988 13.5 0.076 0.078 19.746 20.554
make_images 4988 14.5 3.750 3.920 19.272 20.084
sum_up_and_integrate 127 10.3 0.320 0.325 19.812 19.876
ot_diis_step 116 11.6 0.022 0.022 19.840 19.841
multiply_cannon_metrocomm3 9976 15.5 0.023 0.024 18.629 19.675
integrate_v_rspace 127 11.3 0.004 0.004 19.491 19.557
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.070 19.168
apply_single 127 13.6 0.001 0.001 19.069 19.167
multiply_cannon_multrec 9976 15.5 10.547 12.422 18.255 18.347
cp_fm_diag_elpa 83 13.4 0.000 0.000 17.623 17.624
cp_fm_diag_elpa_base 83 14.4 13.184 14.848 17.618 17.618
multiply_cannon_sync_h2d 9976 15.5 15.538 15.550 15.538 15.550
init_scf_run 11 5.9 0.000 0.001 12.607 12.607
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.606 12.607
hybrid_alltoall_any 5174 16.4 1.293 3.023 10.744 12.512
pw_transfer 1535 11.6 0.092 0.092 12.361 12.367
fft_wrap_pw1pw2 1281 12.7 0.011 0.011 12.127 12.132
make_images_data 4988 15.5 0.060 0.065 10.505 12.078
fft_wrap_pw1pw2_140 519 13.2 0.539 0.543 10.755 10.767
fft3d_ps 1281 14.7 2.728 2.739 10.174 10.183
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.401 9.467
dbcsr_mm_accdrv_process 20626 16.0 3.789 5.687 7.471 9.348
wfi_extrapolate 11 7.9 0.001 0.001 9.233 9.233
mp_alltoall_d11v 2407 14.1 8.219 9.123 8.219 9.123
cp_fm_cholesky_decompose 22 10.9 8.985 8.994 8.985 8.994
grid_integrate_task_list 127 12.3 8.512 8.705 8.512 8.705
density_rs2pw 127 9.7 0.005 0.005 8.415 8.572
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 8.100 8.101
calculate_dm_sparse 127 9.5 0.001 0.001 6.662 6.724
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.586 6.656
grid_collocate_task_list 127 9.7 6.313 6.349 6.313 6.349
rs_scatter_matrices 138 9.7 3.684 4.547 6.016 6.275
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.071 6.125
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=300.715000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2735.363636, yerr=160.937748
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 1.262072E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255669.
MP_Allreduce 3059 6274.
MP_Sync 4
MP_Alltoall 54 6805335.
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.048 84.993 84.994
qs_energies 1 2.0 0.000 0.000 84.512 84.526
ls_scf 1 3.0 0.000 0.000 83.599 83.612
dbcsr_multiply_generic 111 6.7 0.014 0.016 72.571 72.756
multiply_cannon 111 7.7 0.017 0.020 55.849 57.100
multiply_cannon_loop 111 8.7 0.209 0.223 52.419 53.846
ls_scf_main 1 4.0 0.000 0.000 52.360 52.371
density_matrix_trs4 2 5.0 0.002 0.003 46.837 46.910
ls_scf_init_scf 1 4.0 0.000 0.000 28.221 28.239
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.053 27.116
mp_waitall_1 11031 10.9 22.395 25.859 22.395 25.859
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 24.944 24.966
multiply_cannon_multrec 2664 9.7 8.166 8.949 15.610 17.329
multiply_cannon_sync_h2d 2664 9.7 13.573 15.619 13.573 15.619
make_m2s 222 7.7 0.009 0.011 13.056 13.598
make_images 222 8.7 0.099 0.109 13.033 13.577
multiply_cannon_metrocomm1 2664 9.7 0.009 0.010 9.654 11.845
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.443 8.929
make_images_data 222 9.7 0.004 0.005 7.624 8.118
dbcsr_mm_accdrv_process 4760 10.4 0.518 0.611 7.064 8.029
hybrid_alltoall_any 227 10.6 0.214 1.823 6.496 7.711
dbcsr_mm_accdrv_process_sort 4760 11.4 6.345 7.260 6.345 7.260
calculate_norms 4752 9.8 5.506 6.150 5.506 6.150
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.041 5.128
mp_sum_l 807 5.4 3.131 4.574 3.131 4.574
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.052 3.729
mp_irecv_dv 6231 10.9 2.034 3.703 2.034 3.703
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.376 3.515
arnoldi_extremal 4 6.8 0.000 0.000 3.266 3.293
arnoldi_normal_ev 4 7.8 0.001 0.003 3.266 3.293
make_images_sizes 222 9.7 0.000 0.000 0.682 3.286
mp_alltoall_i44 222 10.7 0.682 3.286 0.682 3.286
build_subspace 16 8.4 0.009 0.012 3.169 3.171
ls_scf_post 1 4.0 0.000 0.000 3.017 3.031
ls_scf_store_result 1 5.0 0.000 0.000 2.836 2.880
dbcsr_special_finalize 555 9.7 0.005 0.006 2.394 2.791
dbcsr_merge_single_wm 555 10.7 0.463 0.616 2.386 2.783
make_images_pack 222 9.7 2.208 2.618 2.210 2.620
dbcsr_matrix_vector_mult 304 9.0 0.006 0.013 2.355 2.619
dbcsr_sort_data 658 11.4 2.180 2.487 2.180 2.487
dbcsr_matrix_vector_mult_local 304 10.0 2.071 2.457 2.073 2.459
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.273 2.330
buffer_matrices_ensure_size 222 8.7 1.753 2.092 1.753 2.092
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.834 1.852
rebuild_ks_matrix 3 7.3 0.000 0.000 1.824 1.842
qs_ks_build_kohn_sham_matrix 3 8.3 0.017 0.043 1.824 1.842
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.994000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1142.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 2.098172E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 10339.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.041 90.874 90.875
qs_energies 1 2.0 0.000 0.000 90.405 90.409
ls_scf 1 3.0 0.000 0.000 89.090 89.094
dbcsr_multiply_generic 111 6.7 0.015 0.016 75.031 75.329
multiply_cannon 111 7.7 0.027 0.038 53.369 56.604
ls_scf_main 1 4.0 0.000 0.000 54.668 54.673
multiply_cannon_loop 111 8.7 0.116 0.123 50.110 53.257
density_matrix_trs4 2 5.0 0.002 0.003 48.923 49.103
ls_scf_init_scf 1 4.0 0.000 0.000 30.905 30.906
mp_waitall_1 9105 10.9 21.063 30.149 21.063 30.149
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.693 29.808
multiply_cannon_multrec 1332 9.7 13.099 16.823 22.446 27.363
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.256 27.270
multiply_cannon_metrocomm3 1332 9.7 0.006 0.007 11.817 21.000
make_m2s 222 7.7 0.006 0.008 15.174 15.803
make_images 222 8.7 1.579 1.953 15.144 15.774
dbcsr_mm_accdrv_process 4041 10.4 0.271 0.457 8.949 10.608
dbcsr_mm_accdrv_process_sort 4041 11.4 8.543 10.150 8.543 10.150
make_images_data 222 9.7 0.004 0.004 8.624 9.609
hybrid_alltoall_any 227 10.6 0.522 2.475 8.046 8.981
mp_sum_l 807 5.4 5.187 8.089 5.187 8.089
multiply_cannon_metrocomm4 1221 9.7 0.007 0.008 3.245 7.721
mp_irecv_dv 3311 11.0 3.225 7.666 3.225 7.666
calculate_norms 2376 9.8 5.998 6.864 5.998 6.864
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.971 6.338
multiply_cannon_sync_h2d 1332 9.7 4.783 6.290 4.783 6.290
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.036 5.241
arnoldi_extremal 4 6.8 0.000 0.000 4.725 4.745
arnoldi_normal_ev 4 7.8 0.001 0.005 4.725 4.745
build_subspace 16 8.4 0.014 0.021 4.457 4.465
ls_scf_post 1 4.0 0.000 0.000 3.517 3.521
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.152 3.387
ls_scf_store_result 1 5.0 0.000 0.000 3.230 3.343
dbcsr_matrix_vector_mult_local 304 10.0 2.738 3.221 2.741 3.223
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.625 2.735
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 1.253 2.593
make_images_pack 222 9.7 2.022 2.442 2.024 2.444
mp_allgather_i34 111 8.7 0.967 2.294 0.967 2.294
dbcsr_sort_data 436 11.2 1.773 2.030 1.773 2.030
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.889 1.891
rebuild_ks_matrix 3 7.3 0.000 0.000 1.876 1.878
qs_ks_build_kohn_sham_matrix 3 8.3 0.002 0.004 1.876 1.878
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=90.875000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1733.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 2.704884E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265470.
MP_Allreduce 3058 11181.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.038 0.053 93.678 93.679
qs_energies 1 2.0 0.000 0.000 93.095 93.099
ls_scf 1 3.0 0.000 0.000 91.696 91.702
dbcsr_multiply_generic 111 6.7 0.015 0.017 76.277 76.550
ls_scf_main 1 4.0 0.000 0.000 57.478 57.483
multiply_cannon 111 7.7 0.042 0.078 52.624 57.090
multiply_cannon_loop 111 8.7 0.100 0.106 49.079 52.369
density_matrix_trs4 2 5.0 0.002 0.003 51.512 51.672
mp_waitall_1 7281 11.0 23.743 32.573 23.743 32.573
ls_scf_init_scf 1 4.0 0.000 0.000 30.683 30.685
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.447 29.514
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.042 27.052
multiply_cannon_multrec 888 9.7 12.569 15.254 21.246 24.489
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 10.963 21.355
make_m2s 222 7.7 0.006 0.007 17.180 18.354
make_images 222 8.7 1.968 2.271 17.142 18.314
make_images_data 222 9.7 0.003 0.004 9.896 10.832
hybrid_alltoall_any 227 10.6 0.619 2.848 9.556 10.734
dbcsr_mm_accdrv_process 3754 10.4 0.261 0.416 8.189 9.429
dbcsr_mm_accdrv_process_sort 3754 11.4 7.817 9.013 7.817 9.013
mp_sum_l 807 5.4 4.916 8.472 4.916 8.472
multiply_cannon_sync_h2d 888 9.7 6.047 7.309 6.047 7.309
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.441 6.985
mp_irecv_dv 2335 11.1 2.426 6.935 2.426 6.935
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.792 6.732
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 3.506 6.505
arnoldi_extremal 4 6.8 0.000 0.000 5.135 5.150
arnoldi_normal_ev 4 7.8 0.001 0.005 5.135 5.150
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.888 5.063
build_subspace 16 8.4 0.014 0.020 4.828 4.834
calculate_norms 1584 9.8 4.268 4.656 4.268 4.656
mp_allgather_i34 111 8.7 1.381 3.834 1.381 3.834
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.454 3.793
dbcsr_matrix_vector_mult_local 304 10.0 3.015 3.587 3.017 3.589
ls_scf_post 1 4.0 0.000 0.000 3.535 3.541
ls_scf_store_result 1 5.0 0.000 0.000 3.274 3.364
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.870 2.972
dbcsr_sort_data 325 11.1 1.917 2.193 1.917 2.193
make_images_pack 222 9.7 1.810 2.136 1.813 2.139
make_images_sizes 222 9.7 0.000 0.000 0.933 2.053
mp_alltoall_i44 222 10.7 0.932 2.052 0.932 2.052
dbcsr_finalize 304 7.8 0.026 0.031 1.610 1.957
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.932 1.934
rebuild_ks_matrix 3 7.3 0.000 0.000 1.914 1.916
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.002 1.914 1.916
dbcsr_data_release 9322 10.9 1.317 1.914 1.317 1.914
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=93.679000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2246.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 3.320758E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 13371.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.074 0.093 97.729 97.729
qs_energies 1 2.0 0.000 0.000 97.066 97.074
ls_scf 1 3.0 0.000 0.000 95.421 95.428
dbcsr_multiply_generic 111 6.7 0.016 0.020 79.023 79.306
ls_scf_main 1 4.0 0.000 0.000 59.317 59.319
multiply_cannon 111 7.7 0.043 0.104 51.992 56.404
density_matrix_trs4 2 5.0 0.002 0.003 53.152 53.270
multiply_cannon_loop 111 8.7 0.115 0.125 46.941 50.461
ls_scf_init_scf 1 4.0 0.000 0.000 32.849 32.851
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.629 31.685
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.168 29.182
mp_waitall_1 6369 11.0 23.005 28.824 23.005 28.824
multiply_cannon_multrec 1332 9.7 14.238 17.347 22.207 25.364
make_m2s 222 7.7 0.006 0.008 21.179 22.613
make_images 222 8.7 3.141 3.608 21.128 22.565
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.276 17.403
make_images_data 222 9.7 0.004 0.004 11.859 13.425
hybrid_alltoall_any 227 10.6 0.797 3.785 11.201 12.975
dbcsr_mm_accdrv_process 3641 10.4 0.221 0.385 7.613 9.184
dbcsr_mm_accdrv_process_sort 3641 11.4 7.239 8.762 7.239 8.762
mp_sum_l 807 5.4 4.277 8.383 4.277 8.383
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.310 6.694
multiply_cannon_sync_h2d 1332 9.7 5.543 6.081 5.543 6.081
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.097 6.074
mp_irecv_dv 3229 10.9 2.074 5.992 2.074 5.992
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.700 5.688
arnoldi_extremal 4 6.8 0.000 0.000 5.324 5.337
arnoldi_normal_ev 4 7.8 0.001 0.005 5.324 5.337
build_subspace 16 8.4 0.014 0.021 4.986 4.993
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.549 4.723
calculate_norms 2376 9.8 4.181 4.542 4.181 4.542
mp_allgather_i34 111 8.7 2.175 4.458 2.175 4.458
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.585 3.892
dbcsr_matrix_vector_mult_local 304 10.0 3.182 3.684 3.184 3.685
dbcsr_sort_data 658 11.4 3.070 3.435 3.070 3.435
ls_scf_post 1 4.0 0.000 0.000 3.255 3.262
dbcsr_special_finalize 555 9.7 0.006 0.007 2.820 3.239
dbcsr_merge_single_wm 555 10.7 0.536 0.664 2.811 3.230
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.062 3.112
ls_scf_store_result 1 5.0 0.000 0.000 2.981 3.058
dbcsr_data_release 10477 10.7 1.576 2.402 1.576 2.402
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.983 1.985
dbcsr_finalize 304 7.8 0.049 0.061 1.800 1.981
rebuild_ks_matrix 3 7.3 0.000 0.000 1.960 1.961
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.003 1.959 1.961
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.729000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2681.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 4.637090E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265558.
MP_Allreduce 3049 15663.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.057 0.072 100.191 100.192
qs_energies 1 2.0 0.000 0.000 99.310 99.315
ls_scf 1 3.0 0.000 0.000 97.365 97.376
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.520 78.817
ls_scf_main 1 4.0 0.000 0.000 62.652 62.654
multiply_cannon 111 7.7 0.082 0.153 55.689 60.477
density_matrix_trs4 2 5.0 0.002 0.003 55.300 55.379
multiply_cannon_loop 111 8.7 0.070 0.087 51.231 53.576
mp_waitall_1 5436 11.0 26.985 32.002 26.985 32.002
ls_scf_init_scf 1 4.0 0.000 0.000 31.063 31.068
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.697 29.732
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.493 27.502
multiply_cannon_multrec 444 9.7 14.138 16.102 21.286 24.007
make_m2s 222 7.7 0.004 0.005 17.821 20.271
make_images 222 8.7 3.723 4.419 17.759 20.212
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.558 16.698
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.042 14.758
make_images_data 222 9.7 0.003 0.004 10.185 12.463
hybrid_alltoall_any 227 10.6 0.788 3.759 9.957 12.268
multiply_cannon_sync_h2d 444 9.7 6.540 8.052 6.540 8.052
dbcsr_mm_accdrv_process 3003 10.4 0.179 0.357 6.860 7.987
dbcsr_mm_accdrv_process_sort 3003 11.4 6.541 7.642 6.541 7.642
mp_allgather_i34 111 8.7 2.690 7.031 2.690 7.031
arnoldi_extremal 4 6.8 0.000 0.000 5.847 5.855
arnoldi_normal_ev 4 7.8 0.002 0.005 5.846 5.855
build_subspace 16 8.4 0.015 0.020 5.447 5.456
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.656 4.807
mp_sum_l 807 5.4 3.056 4.636 3.056 4.636
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.181 4.378
dbcsr_matrix_vector_mult_local 304 10.0 3.682 4.156 3.684 4.159
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.725 3.826
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.575 3.764
mp_irecv_dv 1241 11.2 1.555 3.718 1.555 3.718
calculate_norms 792 9.8 3.562 3.701 3.562 3.701
ls_scf_post 1 4.0 0.000 0.000 3.650 3.655
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.147 3.576
make_images_sizes 222 9.7 0.000 0.000 0.982 3.525
mp_alltoall_i44 222 10.7 0.982 3.525 0.982 3.525
ls_scf_store_result 1 5.0 0.000 0.000 3.423 3.476
dbcsr_finalize 304 7.8 0.062 0.078 2.197 2.264
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.237 2.239
rebuild_ks_matrix 3 7.3 0.000 0.000 2.205 2.207
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.002 2.205 2.207
dbcsr_merge_all 275 8.9 0.473 0.526 2.048 2.100
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=100.192000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3688.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/4e8b7c4a5ed53c7d43bc840588464933da05e6c1_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 8.716464E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284111.
MP_Allreduce 3043 21950.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.133 0.139 108.865 108.865
qs_energies 1 2.0 0.000 0.000 107.403 107.417
ls_scf 1 3.0 0.000 0.000 104.475 104.488
dbcsr_multiply_generic 111 6.7 0.023 0.025 77.783 77.891
ls_scf_main 1 4.0 0.000 0.000 66.543 66.545
density_matrix_trs4 2 5.0 0.002 0.003 57.157 57.205
multiply_cannon 111 7.7 0.122 0.221 50.258 52.325
multiply_cannon_loop 111 8.7 0.068 0.070 46.762 47.654
ls_scf_init_scf 1 4.0 0.000 0.000 34.232 34.233
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.567 32.578
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.740 29.751
mp_waitall_1 4527 11.1 22.536 26.112 22.536 26.112
make_m2s 222 7.7 0.005 0.005 23.934 24.956
make_images 222 8.7 4.592 4.955 23.828 24.849
multiply_cannon_multrec 444 9.7 17.863 18.645 22.535 23.226
hybrid_alltoall_any 227 10.6 1.661 3.629 12.932 15.639
make_images_data 222 9.7 0.003 0.003 13.167 15.611
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.790 11.213
multiply_cannon_sync_h2d 444 9.7 8.823 8.869 8.823 8.869
arnoldi_extremal 4 6.8 0.000 0.000 7.428 7.442
arnoldi_normal_ev 4 7.8 0.003 0.009 7.428 7.442
build_subspace 16 8.4 0.026 0.036 6.872 6.884
dbcsr_matrix_vector_mult 304 9.0 0.017 0.033 5.468 5.637
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.325 5.417
dbcsr_matrix_vector_mult_local 304 10.0 5.007 5.323 5.010 5.325
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.041 5.290
dbcsr_mm_accdrv_process 1814 10.4 0.254 0.320 4.505 4.629
dbcsr_mm_accdrv_process_sort 1814 11.4 4.205 4.336 4.205 4.336
ls_scf_post 1 4.0 0.000 0.000 3.700 3.713
mp_allgather_i34 111 8.7 1.150 3.601 1.150 3.601
make_images_sizes 222 9.7 0.000 0.000 1.427 3.523
mp_alltoall_i44 222 10.7 1.427 3.523 1.427 3.523
ls_scf_store_result 1 5.0 0.000 0.000 3.418 3.430
calculate_norms 792 9.8 3.264 3.322 3.264 3.322
dbcsr_finalize 304 7.8 0.082 0.089 3.092 3.163
dbcsr_merge_all 275 8.9 0.892 0.916 2.876 2.943
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.897 2.897
dbcsr_complete_redistribute 5 7.6 1.433 1.469 2.772 2.888
dbcsr_data_release 12724 10.6 2.330 2.850 2.330 2.850
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.623 2.625
rebuild_ks_matrix 3 7.3 0.000 0.000 2.557 2.558
qs_ks_build_kohn_sham_matrix 3 8.3 0.002 0.003 2.557 2.558
matrix_ls_to_qs 2 6.0 0.000 0.000 2.429 2.551
dbcsr_sort_data 325 11.1 2.446 2.499 2.446 2.499
dbcsr_new_transposed 4 7.5 0.243 0.250 2.270 2.284
dbcsr_frobenius_norm 74 6.6 2.053 2.131 2.189 2.229
dbcsr_add_d 103 6.2 0.000 0.000 2.133 2.202
dbcsr_add_anytype 103 7.2 0.860 0.891 2.133 2.202
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=108.865000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6814.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 4e8b7c4a5ed53c7d43bc840588464933da05e6c1
Summary: empty
Status: OK