=== 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: d4020ab16046caa29083b8d6b10b5b5cd0a3b986
################# 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/01
job id: 45599043
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/02
job id: 45599044
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/03
job id: 45599045
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/04
job id: 45599046
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/05
job id: 45599047
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/06
job id: 45599048
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/07
job id: 45599049
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/08
job id: 45599050
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/09
job id: 45599051
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/10
job id: 45599057
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/11
job id: 45599061
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/12
job id: 45599080
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/13
job id: 45599084
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/14
job id: 45599086
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/15
job id: 45599087
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/16
job id: 45599088
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/17
job id: 45599089
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/18
job id: 45599090
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/19
job id: 45599091
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/20
job id: 45599094
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/21
job id: 45599098
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/22
job id: 45599108
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/23
job id: 45599110
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/24
job id: 45599111
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/25
job id: 45599112
--- 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/26
job id: 45599113
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: 512 H2O (4 NVE MD steps on 64 nodes)
input file: benchmarks/QS/00512_H2O/H2O-512_md.inp
required files: []
output file: result.log
# nodes = 64
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/27
job id: 45599115
--- Point ---
name: 601
plot: h2o_512_md
regex: CP2K
label: (64n/12r/1t)
--- Point ---
name: 602
plot: h2o_512_md_mem
regex: Estimated peak process memory
label: (64n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 344 9.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.022 0.035 140.101 140.102
farming_run 1 2.0 138.939 138.941 140.068 140.071
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.448116E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1103589.
MP_Allreduce 491 2254389.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.030 117.874 117.875
qs_energies 1 2.0 0.000 0.000 117.572 117.573
mp2_main 1 3.0 0.001 0.001 114.880 114.882
mp2_gpw_main 1 4.0 0.019 0.026 113.574 113.575
mp2_ri_gpw_compute_in 1 5.0 0.172 0.174 93.920 94.394
mp2_ri_gpw_compute_in_loop 1 6.0 0.005 0.005 55.428 55.903
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.168 41.747 47.145
get_2c_integrals 1 6.0 0.000 0.000 37.578 38.319
integrate_v_rspace 273 8.0 0.438 0.451 25.128 30.303
pw_transfer 6555 10.6 0.375 0.387 27.392 27.846
grid_integrate_task_list 273 9.0 20.939 26.598 20.939 26.598
fft_wrap_pw1pw2 5465 11.4 0.045 0.047 26.048 26.486
fft_wrap_pw1pw2_100 2178 12.4 1.197 1.449 23.564 24.004
compute_2c_integrals 1 7.0 0.003 0.003 19.765 19.766
rpa_ri_compute_en 1 5.0 0.015 0.018 19.539 19.734
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.859 19.402
mp2_eri_2c_integrate_gpw 1 9.0 2.374 2.428 18.856 19.401
cp_fm_cholesky_decompose 12 8.2 18.061 18.837 18.061 18.837
cholesky_decomp 1 7.0 0.000 0.000 16.653 17.410
fft3d_s 5443 13.4 16.150 16.653 16.173 16.675
ao_to_mo_and_store_B_mult_1 272 7.0 10.846 15.554 10.846 15.554
calculate_wavefunction 272 8.0 5.427 5.569 12.552 13.172
rpa_num_int 1 6.0 0.000 0.001 10.859 10.859
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.779 10.825
calc_mat_Q 8 8.0 0.000 0.000 9.343 9.454
contract_S_to_Q 8 9.0 0.000 0.000 8.765 8.872
calc_potential_gpw 544 9.5 0.005 0.005 8.213 8.655
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.202 8.549
parallel_gemm_fm 14 9.1 0.000 0.000 8.359 8.441
parallel_gemm_fm_cosma 14 10.1 8.359 8.441 8.359 8.441
potential_pw2rs 545 10.0 0.108 0.110 7.708 8.308
create_integ_mat 1 6.0 0.022 0.027 8.094 8.094
collocate_single_gaussian 272 10.0 0.040 0.042 7.463 7.770
array2fm 1 7.0 0.000 0.000 6.657 7.140
pw_scatter_s 2720 13.7 4.417 4.614 4.417 4.614
pw_gather_s 2722 13.2 3.869 4.329 3.869 4.329
array2fm_buffer_send 1 8.0 2.983 3.157 2.983 3.157
scf_env_do_scf 1 3.0 0.000 0.000 2.392 2.392
pw_poisson_solve 545 10.5 1.101 1.140 2.150 2.364
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=113.574073, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2720.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 344 10.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.031 0.039 401.015 401.017
farming_run 1 2.0 398.945 398.967 400.961 400.964
-------------------------------------------------------------------------------
@@@@@@@@@@ Run number: 2 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 16777216 0.0% 0.0% 100.0%
flops 14 x 32 x 32 565182464 0.0% 0.0% 100.0%
flops 29 x 32 x 32 585367552 0.0% 0.0% 100.0%
flops 14 x 14 x 32 626196480 0.0% 0.0% 100.0%
flops 29 x 14 x 32 638582784 0.0% 0.0% 100.0%
flops 14 x 29 x 32 638582784 0.0% 0.0% 100.0%
flops 29 x 29 x 32 682057728 0.0% 0.0% 100.0%
flops 14 x 32 x 14 897827141120 0.0% 0.0% 100.0%
flops 29 x 32 x 14 929989394432 0.0% 0.0% 100.0%
flops 14 x 32 x 29 929989394432 0.0% 0.0% 100.0%
flops 29 x 32 x 29 963203301376 0.0% 0.0% 100.0%
flops 32 x 32 x 14 1693481172992 0.0% 0.0% 100.0%
flops 32 x 32 x 29 1753962643456 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 7.172206E+12 0.0% 0.0% 100.0%
flops max/rank 150.696064E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 249788822 0.0% 0.0% 100.0%
number of processed stacks 98736 0.0% 0.0% 100.0%
average stack size 0.0 0.0 2529.9
marketing flops 7.174951E+12
-------------------------------------------------------------------------------
# multiplications 1140
max memory usage/rank 1.226318E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 743 386399.
MP_Allreduce 1941 22272.
MP_Sync 37
MP_Alltoall 77 25922478.
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 0.010 0.027 210.085 210.086
qs_energies 1 2.0 0.000 0.000 209.838 209.851
scf_env_do_scf 1 3.0 0.000 0.000 107.009 107.009
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.978 105.986
rebuild_ks_matrix 4 6.0 0.000 0.000 105.977 105.985
qs_ks_build_kohn_sham_matrix 4 7.0 0.055 0.062 105.977 105.985
hfx_ks_matrix 4 8.0 0.001 0.001 105.572 105.576
integrate_four_center 4 9.0 0.143 0.459 105.571 105.576
mp2_main 1 3.0 0.003 0.011 102.508 102.520
mp2_gpw_main 1 4.0 0.042 0.070 101.449 101.456
integrate_four_center_main 4 10.0 0.089 0.643 96.543 99.622
integrate_four_center_bin 264 11.0 96.455 99.558 96.455 99.558
init_scf_loop 1 4.0 0.000 0.000 92.703 92.703
mp2_ri_gpw_compute_in 1 5.0 0.067 0.072 74.958 76.003
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 54.475 55.535
mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.159 42.104 47.367
integrate_v_rspace 95 8.0 0.398 0.566 28.508 33.740
pw_transfer 2240 10.6 0.143 0.168 29.965 30.494
fft_wrap_pw1pw2 1868 11.4 0.018 0.021 28.962 29.440
grid_integrate_task_list 95 9.0 23.772 29.148 23.772 29.148
mp2_ri_gpw_compute_en 1 5.0 0.059 0.072 26.308 28.045
ao_to_mo_and_store_B_mult_1 91 7.0 10.690 27.558 10.690 27.558
fft_wrap_pw1pw2_100 730 12.4 1.287 1.442 26.662 27.095
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.849 1.898 24.592 24.603
get_2c_integrals 1 6.0 0.000 0.000 20.348 20.427
compute_2c_integrals 1 7.0 0.006 0.014 19.331 19.347
compute_2c_integrals_loop_lm 1 8.0 0.010 0.024 18.933 19.194
mp2_eri_2c_integrate_gpw 1 9.0 1.727 1.941 18.923 19.193
fft3d_s 1823 13.4 18.449 18.764 18.463 18.778
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.303 14.303
calculate_wavefunction 91 8.0 2.012 2.047 9.740 9.983
potential_pw2rs 186 10.0 0.034 0.035 8.690 9.214
mp2_ri_gpw_compute_en_expansio 172 7.0 0.555 0.582 8.804 9.118
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.285 8.584
local_gemm 172 8.0 8.249 8.575 8.249 8.575
collocate_single_gaussian 91 10.0 0.017 0.024 7.912 8.154
calc_potential_gpw 182 9.5 0.002 0.002 7.923 8.148
mp2_ri_gpw_compute_en_comm 22 7.0 0.540 0.560 7.555 7.974
mp_sync 37 10.5 3.697 6.534 3.697 6.534
mp2_ri_gpw_compute_en_ener 172 7.0 6.340 6.434 6.340 6.434
mp_sendrecv_dm3 2068 8.0 5.479 5.882 5.479 5.882
pw_gather_s 912 13.2 4.911 5.321 4.911 5.321
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.447930, 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/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 452.096000E+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
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 0.012 0.037 60.765 60.775
qs_mol_dyn_low 1 2.0 0.003 0.003 60.077 60.084
qs_forces 11 3.9 0.003 0.003 60.009 60.011
qs_energies 11 4.9 0.001 0.001 58.367 58.387
scf_env_do_scf 11 5.9 0.000 0.001 50.765 50.766
scf_env_do_scf_inner_loop 108 6.5 0.002 0.011 48.142 48.143
qs_scf_new_mos 108 7.5 0.000 0.001 36.199 36.496
qs_scf_loop_do_ot 108 8.5 0.000 0.001 36.198 36.495
dbcsr_multiply_generic 2286 12.5 0.093 0.097 35.979 36.487
ot_scf_mini 108 9.5 0.002 0.002 34.459 34.637
multiply_cannon 2286 13.5 0.190 0.202 27.744 29.546
multiply_cannon_loop 2286 14.5 1.464 1.554 26.701 28.527
velocity_verlet 10 3.0 0.001 0.001 28.430 28.430
ot_mini 108 10.5 0.001 0.001 20.277 20.515
qs_ot_get_derivative 108 11.5 0.001 0.001 17.225 17.406
mp_waitall_1 245248 16.5 9.706 16.593 9.706 16.593
multiply_cannon_metrocomm3 54864 15.5 0.068 0.074 6.441 14.525
multiply_cannon_multrec 54864 15.5 4.226 6.534 7.740 11.153
qs_ot_get_p 119 10.4 0.001 0.001 9.359 9.645
rebuild_ks_matrix 119 8.3 0.000 0.000 8.902 9.059
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.902 9.059
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.840 7.978
multiply_cannon_sync_h2d 54864 15.5 5.853 7.004 5.853 7.004
mp_sum_l 7207 12.9 5.312 6.977 5.312 6.977
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 6.010 6.470
init_scf_run 11 5.9 0.000 0.001 6.300 6.300
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.300 6.300
qs_ot_p2m_diag 50 11.0 0.004 0.006 6.131 6.172
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.529 5.636
qs_rho_update_rho_low 119 7.7 0.000 0.001 5.286 5.422
calculate_rho_elec 119 8.7 0.012 0.017 5.286 5.422
sum_up_and_integrate 119 10.3 0.012 0.014 5.321 5.327
integrate_v_rspace 119 11.3 0.002 0.003 5.309 5.316
cp_dbcsr_syevd 50 12.0 0.002 0.003 5.298 5.298
dbcsr_mm_accdrv_process 76910 16.1 1.125 1.830 3.435 4.869
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.846 4.847
cp_fm_redistribute_end 50 14.0 2.452 4.775 2.472 4.781
cp_fm_diag_elpa_base 50 14.0 2.293 4.612 2.304 4.632
calculate_first_density_matrix 1 7.0 0.000 0.000 3.908 3.913
rs_pw_transfer 974 11.9 0.011 0.012 3.660 3.802
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.555 3.558
acc_transpose_blocks 54864 15.5 0.225 0.253 2.514 3.389
multiply_cannon_metrocomm1 54864 15.5 0.053 0.057 1.975 3.331
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.234 3.276
density_rs2pw 119 9.7 0.004 0.005 3.149 3.265
calculate_dm_sparse 119 9.5 0.000 0.001 3.069 3.220
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.972 3.152
apply_single 119 13.6 0.000 0.000 2.972 3.152
jit_kernel_multiply 13 15.8 2.250 2.968 2.250 2.968
pw_transfer 1439 11.6 0.052 0.056 2.645 2.739
ot_diis_step 108 11.5 0.006 0.006 2.738 2.739
make_m2s 4572 13.5 0.053 0.056 2.616 2.694
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 2.569 2.666
potential_pw2rs 119 12.3 0.004 0.004 2.610 2.648
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.588 2.634
make_images 4572 14.5 0.133 0.138 2.534 2.612
init_scf_loop 11 6.9 0.000 0.000 2.593 2.593
fft3d_ps 1201 14.6 0.378 0.491 2.217 2.304
wfi_extrapolate 11 7.9 0.001 0.001 2.304 2.304
acc_transpose_blocks_kernels 54864 16.5 0.247 0.388 1.559 2.234
grid_integrate_task_list 119 12.3 2.011 2.120 2.011 2.120
fft_wrap_pw1pw2_140 487 13.2 0.179 0.199 1.964 2.059
mp_alltoall_d11v 2130 13.8 1.768 1.978 1.768 1.978
mp_sum_d 4135 12.0 1.369 1.946 1.369 1.946
jit_kernel_transpose 5 15.5 1.312 1.861 1.312 1.861
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.523 1.550
make_images_sizes 4572 15.5 0.004 0.005 1.247 1.538
mp_alltoall_i44 4572 16.5 1.243 1.534 1.243 1.534
mp_waitany 12084 13.8 1.312 1.444 1.312 1.444
prepare_preconditioner 11 7.9 0.000 0.000 1.369 1.398
make_preconditioner 11 8.9 0.000 0.000 1.368 1.398
mp_alltoall_z22v 1201 16.6 1.310 1.380 1.310 1.380
grid_collocate_task_list 119 9.7 1.288 1.378 1.288 1.378
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.305 1.340
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=60.775000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=430.909091, yerr=1.504813
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 487.645184E+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 1223828.
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 0.020 0.036 45.615 45.616
qs_mol_dyn_low 1 2.0 0.003 0.004 45.304 45.333
qs_forces 11 3.9 0.003 0.005 45.242 45.243
qs_energies 11 4.9 0.001 0.002 43.416 43.420
scf_env_do_scf 11 5.9 0.000 0.001 37.813 37.814
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 34.609 34.610
dbcsr_multiply_generic 2286 12.5 0.102 0.104 23.420 24.075
qs_scf_new_mos 108 7.5 0.001 0.001 23.288 23.513
qs_scf_loop_do_ot 108 8.5 0.001 0.001 23.288 23.512
ot_scf_mini 108 9.5 0.002 0.003 22.262 22.433
velocity_verlet 10 3.0 0.001 0.001 21.354 21.355
multiply_cannon 2286 13.5 0.208 0.215 17.055 19.225
multiply_cannon_loop 2286 14.5 0.895 0.970 15.579 18.097
ot_mini 108 10.5 0.001 0.001 13.383 13.641
mp_waitall_1 200699 16.5 7.341 12.934 7.341 12.934
qs_ot_get_derivative 108 11.5 0.001 0.001 10.703 10.880
multiply_cannon_metrocomm3 27432 15.5 0.067 0.071 4.669 10.167
rebuild_ks_matrix 119 8.3 0.000 0.000 8.416 8.635
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 8.415 8.635
multiply_cannon_multrec 27432 15.5 1.918 4.290 5.799 8.546
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.469 7.669
qs_ot_get_p 119 10.4 0.001 0.001 5.716 5.960
dbcsr_mm_accdrv_process 47894 16.0 3.040 4.934 3.812 5.650
sum_up_and_integrate 119 10.3 0.024 0.027 5.279 5.284
integrate_v_rspace 119 11.3 0.002 0.003 5.255 5.264
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.990 5.028
calculate_rho_elec 119 8.7 0.021 0.024 4.990 5.027
rs_pw_transfer 974 11.9 0.010 0.011 4.375 4.887
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.931 4.760
mp_sum_l 7207 12.9 2.679 4.755 2.679 4.755
init_scf_run 11 5.9 0.000 0.002 4.273 4.274
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.273 4.274
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.176 4.227
apply_single 119 13.6 0.000 0.000 3.176 4.227
qs_ot_p2m_diag 50 11.0 0.009 0.013 3.802 3.829
density_rs2pw 119 9.7 0.004 0.004 3.274 3.789
make_m2s 4572 13.5 0.052 0.054 3.240 3.617
make_images 4572 14.5 0.200 0.236 3.152 3.526
cp_dbcsr_syevd 50 12.0 0.003 0.003 3.291 3.293
init_scf_loop 11 6.9 0.000 0.000 3.164 3.164
potential_pw2rs 119 12.3 0.006 0.006 2.952 2.974
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.760 2.846
multiply_cannon_sync_h2d 27432 15.5 2.080 2.785 2.080 2.785
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.720 2.721
cp_fm_redistribute_end 50 14.0 1.373 2.665 1.385 2.671
ot_diis_step 108 11.5 0.011 0.011 2.577 2.578
cp_fm_diag_elpa_base 50 14.0 1.244 2.521 1.277 2.578
pw_transfer 1439 11.6 0.065 0.069 2.405 2.456
calculate_first_density_matrix 1 7.0 0.000 0.000 2.379 2.380
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.313 2.365
jit_kernel_multiply 12 16.2 0.719 2.280 0.719 2.280
calculate_dm_sparse 119 9.5 0.000 0.001 2.183 2.272
prepare_preconditioner 11 7.9 0.000 0.000 2.153 2.179
make_preconditioner 11 8.9 0.000 0.000 2.153 2.179
make_images_data 4572 15.5 0.044 0.051 1.443 2.095
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.031 2.094
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.058 2.060
hybrid_alltoall_any 4725 16.4 0.051 0.112 1.241 1.964
grid_integrate_task_list 119 12.3 1.835 1.930 1.835 1.930
fft3d_ps 1201 14.6 0.522 0.577 1.864 1.912
wfi_extrapolate 11 7.9 0.001 0.001 1.822 1.823
mp_alltoall_d11v 2130 13.8 1.547 1.799 1.547 1.799
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.730 1.790
mp_allgather_i34 2286 14.5 0.918 1.780 0.918 1.780
fft_wrap_pw1pw2_140 487 13.2 0.202 0.211 1.586 1.635
acc_transpose_blocks 27432 15.5 0.108 0.112 1.195 1.495
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.368 1.427
make_images_sizes 4572 15.5 0.005 0.005 0.984 1.371
mp_alltoall_i44 4572 16.5 0.979 1.366 0.979 1.366
grid_collocate_task_list 119 9.7 1.231 1.358 1.231 1.358
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.285 1.295
rs_pw_transfer_PW2RS_50 119 14.3 0.588 0.608 1.130 1.258
mp_sum_d 4135 12.0 0.738 1.248 0.738 1.248
mp_sendrecv_dv 16779 12.7 1.208 1.246 1.208 1.246
mp_waitany 5720 13.7 0.631 1.153 0.631 1.153
rs_pw_transfer_RS2PW_140 130 11.5 0.138 0.146 0.631 1.128
mp_alltoall_z22v 1201 16.6 0.988 1.100 0.988 1.100
cp_fm_cholesky_invert 11 10.9 1.021 1.027 1.021 1.027
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.006 1.025
rs_pw_transfer_RS2PW_50 119 11.7 0.272 0.288 0.928 1.018
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.000 1.001
dbcsr_complete_redistribute 329 12.2 0.081 0.108 0.894 0.976
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=45.616000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=464.727273, yerr=1.762793
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 520.450048E+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 744231.
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 0.051 0.102 40.114 40.116
qs_mol_dyn_low 1 2.0 0.003 0.004 39.710 39.717
qs_forces 11 3.9 0.002 0.003 39.623 39.625
qs_energies 11 4.9 0.007 0.023 37.940 37.946
scf_env_do_scf 11 5.9 0.003 0.016 32.821 32.821
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 29.320 29.320
dbcsr_multiply_generic 2286 12.5 0.097 0.098 18.993 19.282
qs_scf_new_mos 108 7.5 0.001 0.001 19.051 19.067
qs_scf_loop_do_ot 108 8.5 0.001 0.001 19.050 19.067
velocity_verlet 10 3.0 0.001 0.001 18.946 18.948
ot_scf_mini 108 9.5 0.003 0.003 18.143 18.160
multiply_cannon 2286 13.5 0.196 0.203 14.084 15.317
multiply_cannon_loop 2286 14.5 0.634 0.666 12.947 14.384
ot_mini 108 10.5 0.001 0.001 10.654 10.702
qs_ot_get_derivative 108 11.5 0.001 0.001 8.842 8.863
rebuild_ks_matrix 119 8.3 0.000 0.000 7.569 7.634
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 7.569 7.634
multiply_cannon_multrec 18288 15.5 1.880 2.757 6.693 6.918
mp_waitall_1 158411 16.6 4.697 6.813 4.697 6.813
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.748 6.810
dbcsr_mm_accdrv_process 38222 16.0 4.018 5.221 4.732 5.546
sum_up_and_integrate 119 10.3 0.031 0.032 5.007 5.013
integrate_v_rspace 119 11.3 0.003 0.003 4.976 4.986
qs_ot_get_p 119 10.4 0.001 0.001 4.919 4.981
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.595 4.614
calculate_rho_elec 119 8.7 0.031 0.031 4.594 4.613
rs_pw_transfer 974 11.9 0.009 0.010 3.871 4.121
multiply_cannon_metrocomm3 18288 15.5 0.044 0.045 2.184 4.079
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.179 3.802
init_scf_run 11 5.9 0.000 0.001 3.789 3.789
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.789 3.789
init_scf_loop 11 6.9 0.001 0.004 3.456 3.457
density_rs2pw 119 9.7 0.004 0.006 3.051 3.293
qs_ot_p2m_diag 50 11.0 0.012 0.013 3.265 3.282
make_m2s 4572 13.5 0.044 0.045 2.698 2.951
cp_dbcsr_syevd 50 12.0 0.003 0.005 2.906 2.907
make_images 4572 14.5 0.191 0.205 2.613 2.867
potential_pw2rs 119 12.3 0.007 0.008 2.713 2.727
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.194 2.555
apply_single 119 13.6 0.000 0.000 2.194 2.555
prepare_preconditioner 11 7.9 0.000 0.000 2.508 2.515
make_preconditioner 11 8.9 0.000 0.001 2.507 2.515
cp_fm_diag_elpa 50 13.0 0.000 0.001 2.481 2.492
cp_fm_diag_elpa_base 50 14.0 2.446 2.462 2.476 2.488
pw_transfer 1439 11.6 0.066 0.070 2.436 2.458
make_full_inverse_cholesky 11 9.9 0.000 0.001 2.343 2.435
mp_sum_l 7207 12.9 1.825 2.418 1.825 2.418
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.343 2.364
calculate_first_density_matrix 1 7.0 0.001 0.002 2.292 2.292
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.245 2.260
jit_kernel_multiply 11 16.0 0.662 2.057 0.662 2.057
calculate_dm_sparse 119 9.5 0.000 0.001 2.013 2.028
grid_integrate_task_list 119 12.3 1.790 1.906 1.790 1.906
fft3d_ps 1201 14.6 0.534 0.551 1.829 1.851
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.795 1.797
ot_diis_step 108 11.5 0.011 0.012 1.737 1.738
fft_wrap_pw1pw2_140 487 13.2 0.254 0.263 1.567 1.587
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.562 1.581
make_images_data 4572 15.5 0.044 0.048 1.264 1.573
multiply_cannon_sync_h2d 18288 15.5 1.239 1.479 1.239 1.479
wfi_extrapolate 11 7.9 0.001 0.001 1.427 1.427
hybrid_alltoall_any 4725 16.4 0.055 0.113 1.072 1.407
grid_collocate_task_list 119 9.7 1.213 1.349 1.213 1.349
acc_transpose_blocks 18288 15.5 0.075 0.077 1.245 1.263
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.224 1.233
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.171 1.222
cp_fm_cholesky_invert 11 10.9 1.177 1.185 1.177 1.185
mp_alltoall_z22v 1201 16.6 1.077 1.147 1.077 1.147
mp_alltoall_d11v 2130 13.8 1.002 1.129 1.002 1.129
mp_allgather_i34 2286 14.5 0.597 1.097 0.597 1.097
make_images_sizes 4572 15.5 0.005 0.005 0.743 1.073
mp_alltoall_i44 4572 16.5 0.738 1.068 0.738 1.068
mp_sendrecv_dv 11067 12.7 1.008 1.023 1.008 1.023
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.012 1.014
mp_waitany 9880 13.7 0.640 0.926 0.640 0.926
rs_pw_transfer_PW2RS_50 119 14.3 0.401 0.411 0.875 0.922
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.895 0.900
rs_pw_transfer_RS2PW_140 130 11.5 0.122 0.136 0.611 0.876
dbcsr_complete_redistribute 329 12.2 0.096 0.118 0.728 0.838
acc_transpose_blocks_kernels 18288 16.5 0.211 0.218 0.805 0.815
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=40.116000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.181818, yerr=1.192262
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 551.579648E+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 0.040 0.091 41.933 41.959
qs_mol_dyn_low 1 2.0 0.003 0.004 41.591 41.601
qs_forces 11 3.9 0.002 0.003 41.507 41.509
qs_energies 11 4.9 0.003 0.022 39.646 39.653
scf_env_do_scf 11 5.9 0.000 0.001 33.718 33.719
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 29.504 29.505
velocity_verlet 10 3.0 0.001 0.001 20.787 20.789
dbcsr_multiply_generic 2286 12.5 0.098 0.101 20.336 20.480
qs_scf_new_mos 108 7.5 0.001 0.001 19.328 19.382
qs_scf_loop_do_ot 108 8.5 0.001 0.001 19.328 19.381
ot_scf_mini 108 9.5 0.002 0.003 18.295 18.348
multiply_cannon 2286 13.5 0.235 0.270 15.479 16.015
multiply_cannon_loop 2286 14.5 0.937 0.969 14.240 14.593
ot_mini 108 10.5 0.001 0.001 10.679 10.747
multiply_cannon_multrec 27432 15.5 2.367 3.068 9.010 9.487
qs_ot_get_derivative 108 11.5 0.001 0.001 8.736 8.797
dbcsr_mm_accdrv_process 47916 15.9 5.478 7.159 6.547 7.849
rebuild_ks_matrix 119 8.3 0.000 0.000 7.630 7.677
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.014 7.629 7.677
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.788 6.830
qs_ot_get_p 119 10.4 0.001 0.001 4.771 4.849
sum_up_and_integrate 119 10.3 0.035 0.038 4.717 4.725
integrate_v_rspace 119 11.3 0.003 0.003 4.682 4.691
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.449 4.489
calculate_rho_elec 119 8.7 0.040 0.046 4.449 4.488
init_scf_run 11 5.9 0.000 0.001 4.392 4.392
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.391 4.392
mp_waitall_1 137007 16.6 3.276 4.281 3.276 4.281
init_scf_loop 11 6.9 0.000 0.002 4.181 4.182
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.197 3.667
rs_pw_transfer 974 11.9 0.009 0.010 3.388 3.581
prepare_preconditioner 11 7.9 0.000 0.001 3.248 3.255
make_preconditioner 11 8.9 0.000 0.001 3.248 3.254
make_m2s 4572 13.5 0.055 0.056 3.108 3.243
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.826 3.173
make_images 4572 14.5 0.271 0.333 3.000 3.137
qs_ot_p2m_diag 50 11.0 0.015 0.023 3.080 3.090
density_rs2pw 119 9.7 0.004 0.004 2.821 3.007
calculate_first_density_matrix 1 7.0 0.000 0.000 2.760 2.761
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.700 2.700
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.183 2.657
apply_single 119 13.6 0.000 0.000 2.183 2.657
pw_transfer 1439 11.6 0.065 0.070 2.449 2.490
potential_pw2rs 119 12.3 0.009 0.010 2.415 2.420
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.382 2.411
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.356 2.400
jit_kernel_multiply 10 15.9 1.007 2.378 1.007 2.378
calculate_dm_sparse 119 9.5 0.000 0.000 2.280 2.341
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.273 2.281
cp_fm_diag_elpa_base 50 14.0 2.229 2.249 2.270 2.279
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.224 2.226
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.997 2.011
mp_sum_l 7207 12.9 1.354 1.965 1.354 1.965
grid_integrate_task_list 119 12.3 1.822 1.909 1.822 1.909
ot_diis_step 108 11.5 0.012 0.013 1.889 1.889
fft_wrap_pw1pw2_140 487 13.2 0.289 0.301 1.791 1.839
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.902 1.829
fft3d_ps 1201 14.6 0.557 0.613 1.793 1.820
make_images_data 4572 15.5 0.045 0.049 1.335 1.587
wfi_extrapolate 11 7.9 0.001 0.001 1.565 1.565
acc_transpose_blocks 27432 15.5 0.113 0.114 1.504 1.543
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.383 1.392
grid_collocate_task_list 119 9.7 1.221 1.333 1.221 1.333
hybrid_alltoall_any 4725 16.4 0.062 0.150 1.114 1.332
dbcsr_complete_redistribute 329 12.2 0.117 0.141 1.005 1.294
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.253 1.274
cp_fm_upper_to_full 72 14.2 0.857 1.232 0.857 1.232
mp_alltoall_d11v 2130 13.8 1.028 1.219 1.028 1.219
mp_alltoall_z22v 1201 16.6 1.120 1.159 1.120 1.159
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 1.148 1.150
cp_fm_cholesky_invert 11 10.9 1.143 1.148 1.143 1.148
make_images_sizes 4572 15.5 0.005 0.005 0.754 1.039
mp_alltoall_i44 4572 16.5 0.749 1.034 0.749 1.034
multiply_cannon_sync_h2d 27432 15.5 0.957 1.012 0.957 1.012
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.682 0.964
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.935 0.940
acc_transpose_blocks_kernels 27432 16.5 0.268 0.276 0.878 0.910
mp_sendrecv_dv 8211 12.7 0.866 0.887 0.866 0.887
multiply_cannon_metrocomm1 27432 15.5 0.032 0.033 0.406 0.878
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.792 0.873
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=41.959000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=523.727273, yerr=2.299838
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/07/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 117.977176E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1384136 0.0% 0.0% 100.0%
average stack size 0.0 0.0 106.2
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 614.912000E+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 0.015 0.052 33.641 33.651
qs_mol_dyn_low 1 2.0 0.003 0.003 33.430 33.435
qs_forces 11 3.9 0.002 0.002 33.359 33.361
qs_energies 11 4.9 0.001 0.001 31.480 31.484
scf_env_do_scf 11 5.9 0.000 0.001 26.225 26.225
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 22.655 22.656
velocity_verlet 10 3.0 0.001 0.001 17.888 17.890
dbcsr_multiply_generic 2286 12.5 0.092 0.096 13.930 14.067
qs_scf_new_mos 108 7.5 0.001 0.001 13.344 13.374
qs_scf_loop_do_ot 108 8.5 0.001 0.001 13.343 13.373
ot_scf_mini 108 9.5 0.002 0.002 12.563 12.588
multiply_cannon 2286 13.5 0.230 0.237 10.474 11.231
multiply_cannon_loop 2286 14.5 0.330 0.342 9.365 9.678
ot_mini 108 10.5 0.001 0.001 6.955 6.988
rebuild_ks_matrix 119 8.3 0.000 0.000 6.873 6.906
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.873 6.906
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.103 6.131
multiply_cannon_multrec 9144 15.5 1.619 1.814 5.777 5.966
qs_ot_get_derivative 108 11.5 0.001 0.001 5.547 5.573
sum_up_and_integrate 119 10.3 0.038 0.041 4.399 4.405
integrate_v_rspace 119 11.3 0.003 0.003 4.362 4.368
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.291 4.312
calculate_rho_elec 119 8.7 0.060 0.061 4.291 4.312
dbcsr_mm_accdrv_process 12550 15.8 3.186 4.064 4.056 4.142
init_scf_run 11 5.9 0.000 0.001 3.746 3.746
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.745 3.746
qs_ot_get_p 119 10.4 0.001 0.001 3.566 3.609
mp_waitall_1 115863 16.7 2.962 3.560 2.962 3.560
init_scf_loop 11 6.9 0.000 0.000 3.532 3.533
rs_pw_transfer 974 11.9 0.008 0.008 2.726 2.896
density_rs2pw 119 9.7 0.004 0.004 2.529 2.685
prepare_preconditioner 11 7.9 0.000 0.000 2.657 2.665
make_preconditioner 11 8.9 0.000 0.000 2.657 2.665
make_m2s 4572 13.5 0.034 0.035 2.377 2.600
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.524 2.564
make_images 4572 14.5 0.268 0.301 2.287 2.512
pw_transfer 1439 11.6 0.066 0.070 2.365 2.381
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.338 2.341
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.270 2.288
calculate_first_density_matrix 1 7.0 0.000 0.000 2.123 2.159
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.101 2.102
potential_pw2rs 119 12.3 0.010 0.011 2.017 2.025
grid_integrate_task_list 119 12.3 1.875 1.971 1.875 1.971
calculate_dm_sparse 119 9.5 0.000 0.000 1.895 1.928
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.910 1.926
fft_wrap_pw1pw2_140 487 13.2 0.367 0.376 1.747 1.766
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.743 1.752
cp_fm_diag_elpa_base 50 14.0 1.712 1.733 1.741 1.750
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.640 1.642
fft3d_ps 1201 14.6 0.561 0.572 1.599 1.613
wfi_extrapolate 11 7.9 0.001 0.001 1.528 1.528
make_images_data 4572 15.5 0.039 0.043 1.164 1.521
cp_fm_cholesky_invert 11 10.9 1.461 1.465 1.461 1.465
grid_collocate_task_list 119 9.7 1.297 1.428 1.297 1.428
jit_kernel_multiply 8 15.7 0.833 1.419 0.833 1.419
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.403 1.414
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.400 1.413
ot_diis_step 108 11.5 0.013 0.013 1.377 1.377
hybrid_alltoall_any 4725 16.4 0.063 0.176 1.071 1.340
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.228 1.265
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.228 1.257
apply_single 119 13.6 0.000 0.000 1.228 1.257
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.232 1.233
mp_alltoall_d11v 2130 13.8 1.058 1.155 1.058 1.155
multiply_cannon_metrocomm1 9144 15.5 0.022 0.022 0.729 1.008
mp_allgather_i34 2286 14.5 0.439 1.005 0.439 1.005
multiply_cannon_metrocomm3 9144 15.5 0.019 0.019 0.607 0.984
mp_alltoall_z22v 1201 16.6 0.913 0.953 0.913 0.953
acc_transpose_blocks 9144 15.5 0.037 0.038 0.935 0.943
mp_sum_l 7207 12.9 0.664 0.929 0.664 0.929
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.868 0.923
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.899 0.915
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.898 0.904
make_images_sizes 4572 15.5 0.005 0.005 0.553 0.856
mp_alltoall_i44 4572 16.5 0.547 0.851 0.547 0.851
multiply_cannon_sync_h2d 9144 15.5 0.704 0.777 0.704 0.777
dbcsr_complete_redistribute 329 12.2 0.160 0.167 0.719 0.761
cp_fm_cholesky_decompose 22 10.9 0.746 0.750 0.746 0.750
qs_env_update_s_mstruct 11 6.9 0.001 0.001 0.685 0.735
mp_waitany 5200 13.7 0.534 0.705 0.534 0.705
acc_transpose_blocks_kernels 9144 16.5 0.117 0.119 0.695 0.700
mp_sendrecv_dv 5355 12.7 0.680 0.699 0.680 0.699
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=33.651000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=581.727273, yerr=6.510005
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 748.679168E+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 0.016 0.039 47.972 47.973
qs_mol_dyn_low 1 2.0 0.003 0.003 47.692 47.699
qs_forces 11 3.9 0.002 0.002 47.614 47.615
qs_energies 11 4.9 0.001 0.001 45.482 45.489
scf_env_do_scf 11 5.9 0.001 0.001 39.067 39.067
scf_env_do_scf_inner_loop 108 6.5 0.004 0.008 29.814 29.817
velocity_verlet 10 3.0 0.001 0.001 27.363 27.370
dbcsr_multiply_generic 2286 12.5 0.100 0.103 19.737 20.032
qs_scf_new_mos 108 7.5 0.001 0.001 18.622 18.723
qs_scf_loop_do_ot 108 8.5 0.001 0.001 18.621 18.722
ot_scf_mini 108 9.5 0.002 0.002 17.426 17.526
multiply_cannon 2286 13.5 0.306 0.315 14.665 15.946
multiply_cannon_loop 2286 14.5 0.345 0.350 13.178 14.395
ot_mini 108 10.5 0.001 0.001 10.272 10.416
init_scf_loop 11 6.9 0.000 0.000 9.211 9.215
multiply_cannon_multrec 9144 15.5 3.501 4.923 8.819 8.889
rebuild_ks_matrix 119 8.3 0.000 0.000 8.199 8.368
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 8.199 8.368
qs_ot_get_derivative 108 11.5 0.001 0.001 8.050 8.147
prepare_preconditioner 11 7.9 0.000 0.000 8.121 8.135
make_preconditioner 11 8.9 0.000 0.000 8.121 8.135
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.623 8.001
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.370 7.524
dbcsr_mm_accdrv_process 12550 15.8 4.165 5.758 5.193 6.522
mp_waitall_1 94719 16.7 3.676 5.055 3.676 5.055
qs_rho_update_rho_low 119 7.7 0.001 0.001 5.042 5.049
calculate_rho_elec 119 8.7 0.118 0.121 5.041 5.049
sum_up_and_integrate 119 10.3 0.064 0.066 4.676 4.682
cp_fm_upper_to_full 72 14.2 3.220 4.650 3.220 4.650
integrate_v_rspace 119 11.3 0.003 0.003 4.611 4.618
init_scf_run 11 5.9 0.000 0.001 4.259 4.259
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.259 4.259
qs_ot_get_p 119 10.4 0.001 0.001 4.070 4.226
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.969 3.457
make_m2s 4572 13.5 0.038 0.038 3.080 3.422
make_images 4572 14.5 0.352 0.380 2.959 3.302
pw_transfer 1439 11.6 0.069 0.069 3.235 3.253
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 3.137 3.154
dbcsr_complete_redistribute 329 12.2 0.281 0.285 2.181 3.080
multiply_cannon_metrocomm3 9144 15.5 0.020 0.020 1.794 2.982
density_rs2pw 119 9.7 0.004 0.004 2.817 2.837
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.783 2.668
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.391 2.630
apply_single 119 13.6 0.000 0.000 2.390 2.630
fft_wrap_pw1pw2_140 487 13.2 0.619 0.625 2.540 2.562
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.542 2.544
calculate_first_density_matrix 1 7.0 0.000 0.000 2.465 2.465
mp_alltoall_i22 627 13.8 1.551 2.458 1.551 2.458
calculate_dm_sparse 119 9.5 0.000 0.000 2.386 2.410
mp_sum_l 7207 12.9 1.453 2.405 1.453 2.405
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.493 2.372
cp_fm_cholesky_invert 11 10.9 2.337 2.345 2.337 2.345
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.232 2.233
rs_pw_transfer 974 11.9 0.009 0.009 2.176 2.209
grid_integrate_task_list 119 12.3 2.151 2.174 2.151 2.174
ot_diis_step 108 11.5 0.014 0.015 2.168 2.169
fft3d_ps 1201 14.6 0.599 0.609 2.151 2.161
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.136 2.137
make_images_data 4572 15.5 0.043 0.045 1.617 2.132
hybrid_alltoall_any 4725 16.4 0.087 0.149 1.595 2.092
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.008 2.059
potential_pw2rs 119 12.3 0.014 0.014 1.909 1.915
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.850 1.851
cp_fm_diag_elpa_base 50 14.0 1.694 1.756 1.848 1.848
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.836 1.837
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.757 1.784
wfi_extrapolate 11 7.9 0.001 0.001 1.709 1.709
grid_collocate_task_list 119 9.7 1.555 1.577 1.555 1.577
jit_kernel_multiply 8 15.5 1.000 1.534 1.000 1.534
mp_alltoall_d11v 2130 13.8 1.399 1.438 1.399 1.438
mp_alltoall_z22v 1201 16.6 1.414 1.430 1.414 1.430
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.346 1.391
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.355 1.368
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.165 1.179
mp_allgather_i34 2286 14.5 0.465 1.059 0.465 1.059
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.043 1.058
multiply_cannon_sync_h2d 9144 15.5 1.049 1.052 1.049 1.052
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.991 1.029
acc_transpose_blocks 9144 15.5 0.038 0.038 1.004 1.023
qs_create_task_list 11 7.9 0.049 0.049 0.994 1.009
generate_qs_task_list 11 8.9 0.368 0.387 0.946 0.961
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=47.973000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=707.636364, yerr=14.034197
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 502.624256E+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 1691127.
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 0.012 0.033 88.406 88.408
qs_mol_dyn_low 1 2.0 0.003 0.003 88.124 88.134
qs_forces 11 3.9 0.003 0.003 88.017 88.018
qs_energies 11 4.9 0.001 0.003 85.026 85.037
scf_env_do_scf 11 5.9 0.000 0.001 75.672 75.675
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 69.057 69.058
dbcsr_multiply_generic 2055 12.4 0.107 0.110 52.192 52.403
qs_scf_new_mos 99 7.5 0.000 0.001 50.160 50.295
qs_scf_loop_do_ot 99 8.5 0.001 0.001 50.160 50.294
ot_scf_mini 99 9.5 0.002 0.002 47.729 47.805
velocity_verlet 10 3.0 0.001 0.001 46.105 46.106
multiply_cannon 2055 13.4 0.185 0.189 42.701 43.648
multiply_cannon_loop 2055 14.4 1.496 1.533 41.503 42.498
ot_mini 99 10.5 0.001 0.001 27.563 27.625
qs_ot_get_derivative 99 11.5 0.001 0.001 20.698 20.781
multiply_cannon_multrec 49320 15.4 12.395 13.067 17.366 18.088
rebuild_ks_matrix 110 8.3 0.000 0.000 15.530 15.605
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 15.530 15.605
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.619 13.685
mp_waitall_1 220248 16.4 11.762 12.695 11.762 12.695
qs_ot_get_p 110 10.4 0.001 0.001 11.495 11.601
multiply_cannon_sync_h2d 49320 15.4 10.191 10.803 10.191 10.803
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.695 8.227
qs_ot_p2m_diag 48 11.0 0.012 0.019 7.987 8.042
sum_up_and_integrate 110 10.3 0.036 0.042 7.854 7.866
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.273 7.854
apply_single 110 13.6 0.000 0.001 7.273 7.853
integrate_v_rspace 110 11.3 0.003 0.004 7.817 7.838
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.504 7.651
calculate_rho_elec 110 8.6 0.021 0.026 7.504 7.650
multiply_cannon_metrocomm3 49320 15.4 0.079 0.083 6.514 7.577
init_scf_run 11 5.9 0.000 0.001 7.175 7.175
scf_env_initial_rho_setup 11 6.9 0.001 0.001 7.174 7.175
cp_dbcsr_syevd 48 12.0 0.002 0.003 7.021 7.022
ot_diis_step 99 11.5 0.006 0.006 6.624 6.625
init_scf_loop 11 6.9 0.000 0.000 6.571 6.572
cp_fm_diag_elpa 48 13.0 0.000 0.000 6.283 6.310
cp_fm_diag_elpa_base 48 14.0 6.264 6.295 6.281 6.308
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 6.018 6.059
mp_sum_l 6514 12.8 4.442 5.256 4.442 5.256
rs_pw_transfer 902 11.9 0.011 0.013 4.649 5.205
dbcsr_mm_accdrv_process 87628 16.1 1.894 2.011 4.847 5.166
density_rs2pw 110 9.6 0.005 0.005 4.318 4.788
make_m2s 4110 13.4 0.061 0.065 4.594 4.694
make_images 4110 14.4 0.178 0.191 4.498 4.601
wfi_extrapolate 11 7.9 0.001 0.001 4.394 4.394
prepare_preconditioner 11 7.9 0.000 0.000 4.297 4.312
make_preconditioner 11 8.9 0.000 0.000 4.297 4.312
pw_transfer 1331 11.6 0.056 0.067 4.035 4.147
make_full_inverse_cholesky 11 9.9 0.000 0.000 4.066 4.121
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.945 4.060
multiply_cannon_metrocomm1 49320 15.4 0.060 0.062 2.746 4.011
calculate_dm_sparse 110 9.5 0.001 0.001 3.829 3.942
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.641 3.646
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.375 3.449
potential_pw2rs 110 12.3 0.006 0.007 3.384 3.431
grid_integrate_task_list 110 12.3 3.222 3.382 3.222 3.382
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.257 3.299
fft_wrap_pw1pw2_140 451 13.1 0.448 0.497 3.157 3.282
fft3d_ps 1111 14.6 0.793 0.884 3.143 3.234
jit_kernel_multiply 13 15.9 2.679 2.731 2.679 2.731
calculate_first_density_matrix 1 7.0 0.000 0.000 2.688 2.692
mp_alltoall_d11v 2046 13.8 2.216 2.579 2.216 2.579
mp_waitany 14300 13.8 1.879 2.469 1.879 2.469
cp_fm_cholesky_invert 11 10.9 2.410 2.415 2.410 2.415
grid_collocate_task_list 110 9.6 2.082 2.283 2.082 2.283
acc_transpose_blocks 49320 15.4 0.209 0.219 2.091 2.186
make_images_data 4110 15.4 0.042 0.045 1.988 2.147
mp_alltoall_z22v 1111 16.6 1.955 2.132 1.955 2.132
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.043 2.054
mp_sum_d 3889 11.9 1.435 1.959 1.435 1.959
hybrid_alltoall_any 4261 16.3 0.082 0.475 1.702 1.955
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=88.408000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=477.545455, yerr=2.535696
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 589.094912E+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 2579496.
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 0.057 0.208 75.599 75.601
qs_mol_dyn_low 1 2.0 0.003 0.003 75.144 75.155
qs_forces 11 3.9 0.003 0.004 75.075 75.076
qs_energies 11 4.9 0.002 0.007 71.575 71.579
scf_env_do_scf 11 5.9 0.001 0.002 62.161 62.165
scf_env_do_scf_inner_loop 99 6.5 0.003 0.008 52.896 52.897
velocity_verlet 10 3.0 0.001 0.001 39.684 39.686
dbcsr_multiply_generic 2055 12.4 0.116 0.121 38.851 39.024
qs_scf_new_mos 99 7.5 0.001 0.001 35.208 35.357
qs_scf_loop_do_ot 99 8.5 0.001 0.001 35.208 35.356
ot_scf_mini 99 9.5 0.003 0.004 33.525 33.656
multiply_cannon 2055 13.4 0.221 0.244 31.500 32.557
multiply_cannon_loop 2055 14.4 0.921 0.952 30.032 30.762
ot_mini 99 10.5 0.001 0.001 18.953 19.099
multiply_cannon_multrec 24660 15.4 7.639 9.530 14.047 15.948
rebuild_ks_matrix 110 8.3 0.000 0.000 14.699 14.790
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.017 14.698 14.790
qs_ot_get_derivative 99 11.5 0.001 0.001 13.134 13.272
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.917 13.001
mp_waitall_1 176588 16.5 8.483 11.405 8.483 11.405
init_scf_loop 11 6.9 0.001 0.005 9.221 9.223
multiply_cannon_sync_h2d 24660 15.4 7.014 8.006 7.014 8.006
qs_ot_get_p 110 10.4 0.001 0.001 7.800 7.957
multiply_cannon_metrocomm3 24660 15.4 0.067 0.070 5.178 7.876
sum_up_and_integrate 110 10.3 0.052 0.059 7.479 7.488
integrate_v_rspace 110 11.3 0.003 0.003 7.427 7.438
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.520 7.160
apply_single 110 13.6 0.000 0.001 6.520 7.160
prepare_preconditioner 11 7.9 0.000 0.001 7.009 7.024
make_preconditioner 11 8.9 0.000 0.002 7.009 7.024
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.004 7.018
calculate_rho_elec 110 8.6 0.040 0.048 7.003 7.017
init_scf_run 11 5.9 0.000 0.002 6.892 6.892
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.892 6.892
make_full_inverse_cholesky 11 9.9 0.000 0.001 6.554 6.732
dbcsr_mm_accdrv_process 52282 16.1 4.894 5.883 6.247 6.518
ot_diis_step 99 11.5 0.010 0.010 5.768 5.769
qs_ot_p2m_diag 48 11.0 0.029 0.044 5.644 5.668
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.954 5.666
make_m2s 4110 13.4 0.056 0.060 4.839 5.337
make_images 4110 14.4 0.398 0.441 4.730 5.224
cp_dbcsr_syevd 48 12.0 0.003 0.006 5.173 5.174
rs_pw_transfer 902 11.9 0.012 0.013 4.145 4.686
density_rs2pw 110 9.6 0.005 0.011 4.148 4.658
cp_fm_diag_elpa 48 13.0 0.000 0.002 4.421 4.437
cp_fm_diag_elpa_base 48 14.0 4.364 4.393 4.415 4.430
pw_transfer 1331 11.6 0.067 0.075 4.180 4.338
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 4.072 4.233
wfi_extrapolate 11 7.9 0.001 0.001 3.891 3.891
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.561 3.563
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.422 3.502
grid_integrate_task_list 110 12.3 3.141 3.426 3.141 3.426
cp_fm_cholesky_invert 11 10.9 3.373 3.382 3.373 3.382
potential_pw2rs 110 12.3 0.008 0.009 3.245 3.263
fft_wrap_pw1pw2_140 451 13.1 0.519 0.538 3.102 3.262
fft3d_ps 1111 14.6 1.110 1.335 3.045 3.215
calculate_dm_sparse 110 9.5 0.001 0.001 3.148 3.176
make_images_data 4110 15.4 0.046 0.050 2.619 3.140
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.075 3.102
hybrid_alltoall_any 4261 16.3 0.102 0.439 2.264 2.994
calculate_first_density_matrix 1 7.0 0.000 0.002 2.879 2.881
mp_sum_l 6514 12.8 1.940 2.587 1.940 2.587
grid_collocate_task_list 110 9.6 2.051 2.485 2.051 2.485
mp_alltoall_d11v 2046 13.8 1.941 2.459 1.941 2.459
jit_kernel_multiply 10 16.3 1.002 2.087 1.002 2.087
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.031 2.058
mp_waitany 10164 13.8 1.386 1.988 1.386 1.988
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.942 1.957
cp_fm_cholesky_decompose 22 10.9 1.931 1.938 1.931 1.938
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.896 1.897
mp_allgather_i34 2055 14.4 0.796 1.823 0.796 1.823
dbcsr_complete_redistribute 325 12.2 0.281 0.363 1.426 1.734
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.720 1.730
mp_alltoall_z22v 1111 16.6 1.619 1.717 1.619 1.717
multiply_cannon_metrocomm4 22605 15.4 0.075 0.080 0.783 1.709
acc_transpose_blocks 24660 15.4 0.109 0.112 1.598 1.631
mp_irecv_dv 57340 16.2 0.657 1.588 0.657 1.588
rs_pw_transfer_RS2PW_140 121 11.5 0.206 0.218 1.018 1.573
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=75.601000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=556.818182, yerr=6.938931
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/11/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 404.681598E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 3346752 0.0% 0.0% 100.0%
average stack size 0.0 0.0 294.1
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 660.045824E+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 0.013 0.031 67.132 67.134
qs_mol_dyn_low 1 2.0 0.003 0.003 66.857 66.868
qs_forces 11 3.9 0.003 0.003 66.785 66.786
qs_energies 11 4.9 0.001 0.001 63.469 63.480
scf_env_do_scf 11 5.9 0.000 0.001 55.092 55.092
scf_env_do_scf_inner_loop 99 6.5 0.002 0.008 44.654 44.654
velocity_verlet 10 3.0 0.001 0.001 36.891 36.893
dbcsr_multiply_generic 2055 12.4 0.109 0.115 30.301 31.024
qs_scf_new_mos 99 7.5 0.001 0.001 28.538 28.649
qs_scf_loop_do_ot 99 8.5 0.001 0.001 28.537 28.648
ot_scf_mini 99 9.5 0.002 0.003 27.221 27.346
multiply_cannon 2055 13.4 0.211 0.220 22.745 24.654
multiply_cannon_loop 2055 14.4 0.616 0.630 21.286 23.225
ot_mini 99 10.5 0.001 0.001 15.052 15.200
rebuild_ks_matrix 110 8.3 0.000 0.000 13.010 13.299
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 13.010 13.299
mp_waitall_1 139946 16.5 7.999 12.326 7.999 12.326
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.458 11.723
multiply_cannon_multrec 16440 15.4 3.908 4.933 9.663 10.669
qs_ot_get_derivative 99 11.5 0.001 0.001 10.396 10.528
init_scf_loop 11 6.9 0.000 0.000 10.395 10.396
multiply_cannon_metrocomm3 16440 15.4 0.042 0.046 4.615 8.845
prepare_preconditioner 11 7.9 0.000 0.000 8.545 8.565
make_preconditioner 11 8.9 0.000 0.000 8.545 8.565
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.839 8.216
qs_ot_get_p 110 10.4 0.001 0.001 7.069 7.234
sum_up_and_integrate 110 10.3 0.060 0.061 7.126 7.140
integrate_v_rspace 110 11.3 0.003 0.003 7.066 7.081
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.601 6.622
calculate_rho_elec 110 8.6 0.059 0.060 6.601 6.621
init_scf_run 11 5.9 0.000 0.002 5.902 5.902
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.902 5.902
dbcsr_mm_accdrv_process 34862 16.1 4.662 5.300 5.614 5.777
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.017 5.422
apply_single 110 13.6 0.000 0.000 5.017 5.422
make_m2s 4110 13.4 0.050 0.051 4.690 5.068
qs_ot_p2m_diag 48 11.0 0.042 0.044 5.007 5.017
make_images 4110 14.4 0.390 0.509 4.575 4.954
density_rs2pw 110 9.6 0.005 0.005 3.659 4.867
rs_pw_transfer 902 11.9 0.010 0.011 3.600 4.777
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.614 4.615
ot_diis_step 99 11.5 0.011 0.011 4.576 4.577
multiply_cannon_sync_h2d 16440 15.4 3.663 4.378 3.663 4.378
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.566 4.225
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.907 3.914
cp_fm_diag_elpa_base 48 14.0 3.831 3.871 3.905 3.913
pw_transfer 1331 11.6 0.066 0.074 3.828 3.845
cp_fm_cholesky_invert 11 10.9 3.737 3.747 3.737 3.747
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.720 3.740
grid_integrate_task_list 110 12.3 3.183 3.393 3.183 3.393
wfi_extrapolate 11 7.9 0.001 0.001 3.384 3.384
make_images_data 4110 15.4 0.043 0.047 2.637 3.198
mp_sum_l 6514 12.8 2.245 3.075 2.245 3.075
hybrid_alltoall_any 4261 16.3 0.106 0.374 2.343 3.047
fft_wrap_pw1pw2_140 451 13.1 0.637 0.645 3.026 3.046
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.934 2.936
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.843 2.911
potential_pw2rs 110 12.3 0.011 0.011 2.832 2.848
calculate_dm_sparse 110 9.5 0.001 0.001 2.606 2.643
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.509 2.612
multiply_cannon_metrocomm4 14385 15.4 0.046 0.050 0.883 2.568
fft3d_ps 1111 14.6 1.089 1.100 2.541 2.555
mp_waitany 17072 13.8 1.316 2.550 1.316 2.550
grid_collocate_task_list 110 9.6 2.084 2.466 2.084 2.466
mp_irecv_dv 48980 15.7 0.812 2.439 0.812 2.439
calculate_first_density_matrix 1 7.0 0.000 0.000 2.424 2.425
mp_alltoall_d11v 2046 13.8 1.979 2.344 1.979 2.344
cp_fm_cholesky_decompose 22 10.9 2.232 2.255 2.232 2.255
rs_pw_transfer_RS2PW_140 121 11.5 0.176 0.180 0.932 2.115
dbcsr_complete_redistribute 325 12.2 0.311 0.330 1.630 2.110
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.971 1.972
mp_allgather_i34 2055 14.4 0.783 1.863 0.783 1.863
cp_fm_upper_to_full 70 14.2 1.423 1.827 1.423 1.827
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.747 1.773
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.075 1.558
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.499 1.542
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.489 1.507
jit_kernel_multiply 8 16.7 0.564 1.496 0.564 1.496
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.359 1.476
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=67.134000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=625.000000, yerr=8.377459
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/12/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 601.317074E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 4916280 0.0% 0.0% 100.0%
average stack size 0.0 0.0 200.2
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 724.705280E+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 0.020 0.045 72.115 72.118
qs_mol_dyn_low 1 2.0 0.003 0.004 71.788 71.797
qs_forces 11 3.9 0.003 0.003 71.675 71.679
qs_energies 11 4.9 0.003 0.022 68.123 68.129
scf_env_do_scf 11 5.9 0.001 0.001 59.397 59.400
scf_env_do_scf_inner_loop 99 6.5 0.003 0.006 45.557 45.559
velocity_verlet 10 3.0 0.001 0.001 41.075 41.080
dbcsr_multiply_generic 2055 12.4 0.117 0.122 30.804 30.993
qs_scf_new_mos 99 7.5 0.001 0.001 29.499 29.601
qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.499 29.601
ot_scf_mini 99 9.5 0.003 0.003 27.793 27.889
multiply_cannon 2055 13.4 0.242 0.258 22.826 24.092
multiply_cannon_loop 2055 14.4 0.881 0.907 21.166 21.667
ot_mini 99 10.5 0.001 0.001 15.157 15.288
init_scf_loop 11 6.9 0.001 0.005 13.781 13.782
multiply_cannon_multrec 24660 15.4 4.194 6.590 12.589 13.736
rebuild_ks_matrix 110 8.3 0.000 0.000 12.855 12.959
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.015 12.855 12.959
prepare_preconditioner 11 7.9 0.000 0.001 11.948 11.962
make_preconditioner 11 8.9 0.000 0.002 11.948 11.962
make_full_inverse_cholesky 11 9.9 0.000 0.000 10.112 11.610
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.336 11.429
qs_ot_get_derivative 99 11.5 0.001 0.001 10.840 10.944
dbcsr_mm_accdrv_process 52304 16.0 6.850 8.227 8.249 9.178
qs_ot_get_p 110 10.4 0.001 0.001 7.409 7.554
mp_waitall_1 121746 16.5 5.223 7.107 5.223 7.107
sum_up_and_integrate 110 10.3 0.068 0.071 7.064 7.080
integrate_v_rspace 110 11.3 0.003 0.003 6.996 7.012
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.659 6.668
calculate_rho_elec 110 8.6 0.078 0.081 6.659 6.667
make_m2s 4110 13.4 0.060 0.062 6.116 6.486
make_images 4110 14.4 0.575 0.698 5.973 6.340
init_scf_run 11 5.9 0.000 0.002 5.828 5.828
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.827 5.828
qs_ot_p2m_diag 48 11.0 0.055 0.064 5.308 5.322
cp_fm_upper_to_full 70 14.2 3.428 4.974 3.428 4.974
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.785 4.786
ot_diis_step 99 11.5 0.011 0.011 4.254 4.254
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.110 4.206
apply_single 110 13.6 0.000 0.000 4.109 4.206
pw_transfer 1331 11.6 0.066 0.078 4.043 4.087
dbcsr_complete_redistribute 325 12.2 0.410 0.445 2.883 4.068
cp_fm_cholesky_invert 11 10.9 4.047 4.061 4.047 4.061
density_rs2pw 110 9.6 0.005 0.007 3.552 4.059
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.026 4.035
cp_fm_diag_elpa_base 48 14.0 3.859 3.924 4.023 4.033
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.936 3.984
rs_pw_transfer 902 11.9 0.010 0.011 3.195 3.718
multiply_cannon_metrocomm3 24660 15.4 0.036 0.037 1.585 3.684
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.615 3.665
make_images_data 4110 15.4 0.046 0.050 3.101 3.588
hybrid_alltoall_any 4261 16.3 0.120 0.457 2.598 3.523
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.255 3.421
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.348 3.410
grid_integrate_task_list 110 12.3 3.271 3.408 3.271 3.408
wfi_extrapolate 11 7.9 0.001 0.001 3.296 3.297
multiply_cannon_sync_h2d 24660 15.4 3.164 3.289 3.164 3.289
fft_wrap_pw1pw2_140 451 13.1 0.671 0.694 3.210 3.260
calculate_dm_sparse 110 9.5 0.001 0.001 3.004 3.024
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.827 2.973
mp_alltoall_i22 605 13.7 1.725 2.904 1.725 2.904
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.866 2.867
fft3d_ps 1111 14.6 1.085 1.116 2.695 2.717
potential_pw2rs 110 12.3 0.013 0.013 2.677 2.700
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.463 2.502
grid_collocate_task_list 110 9.6 2.181 2.456 2.181 2.456
calculate_first_density_matrix 1 7.0 0.001 0.005 2.427 2.429
jit_kernel_multiply 13 15.8 1.069 2.342 1.069 2.342
qs_energies_init_hamiltonians 11 5.9 0.002 0.003 2.320 2.322
cp_fm_cholesky_decompose 22 10.9 2.253 2.297 2.253 2.297
mp_alltoall_d11v 2046 13.8 1.983 2.242 1.983 2.242
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.881 1.921
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.768 1.780
mp_waitany 13376 13.8 1.242 1.725 1.242 1.725
mp_allgather_i34 2055 14.4 0.763 1.712 0.763 1.712
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.600 1.700
mp_sum_l 6514 12.8 1.183 1.559 1.183 1.559
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.543 1.559
acc_transpose_blocks 24660 15.4 0.106 0.108 1.493 1.514
multiply_cannon_metrocomm4 20550 15.4 0.061 0.065 0.867 1.510
mp_alltoall_z22v 1111 16.6 1.395 1.459 1.395 1.459
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=72.118000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=689.363636, yerr=6.664325
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 818.454528E+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 0.017 0.058 61.204 61.217
qs_mol_dyn_low 1 2.0 0.003 0.003 60.946 60.953
qs_forces 11 3.9 0.003 0.003 60.867 60.867
qs_energies 11 4.9 0.001 0.001 57.069 57.077
scf_env_do_scf 11 5.9 0.000 0.001 48.444 48.445
scf_env_do_scf_inner_loop 99 6.5 0.003 0.009 39.000 39.001
velocity_verlet 10 3.0 0.001 0.001 34.921 34.925
dbcsr_multiply_generic 2055 12.4 0.105 0.109 24.077 24.209
qs_scf_new_mos 99 7.5 0.001 0.001 22.668 22.730
qs_scf_loop_do_ot 99 8.5 0.001 0.001 22.667 22.730
ot_scf_mini 99 9.5 0.002 0.002 21.322 21.337
multiply_cannon 2055 13.4 0.246 0.264 17.906 19.249
multiply_cannon_loop 2055 14.4 0.321 0.336 16.282 16.547
rebuild_ks_matrix 110 8.3 0.000 0.000 12.415 12.444
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 12.415 12.444
ot_mini 99 10.5 0.001 0.001 11.372 11.390
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.022 11.049
init_scf_loop 11 6.9 0.000 0.000 9.384 9.385
mp_waitall_1 103326 16.6 6.798 8.641 6.798 8.641
multiply_cannon_multrec 8220 15.4 3.224 4.416 7.508 8.374
prepare_preconditioner 11 7.9 0.000 0.000 7.670 7.675
make_preconditioner 11 8.9 0.000 0.000 7.670 7.675
qs_ot_get_derivative 99 11.5 0.001 0.001 7.485 7.502
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.212 7.307
sum_up_and_integrate 110 10.3 0.079 0.082 6.970 6.985
integrate_v_rspace 110 11.3 0.003 0.003 6.891 6.905
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.701 6.724
calculate_rho_elec 110 8.6 0.114 0.115 6.700 6.724
qs_ot_get_p 110 10.4 0.001 0.001 5.857 5.891
init_scf_run 11 5.9 0.000 0.002 5.413 5.413
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.412 5.412
dbcsr_mm_accdrv_process 17442 15.9 2.909 3.979 4.155 5.061
make_m2s 4110 13.4 0.039 0.040 4.682 5.014
make_images 4110 14.4 0.637 0.691 4.552 4.882
multiply_cannon_metrocomm3 8220 15.4 0.017 0.017 2.896 4.438
qs_ot_p2m_diag 48 11.0 0.081 0.084 4.296 4.303
pw_transfer 1331 11.6 0.066 0.073 4.236 4.258
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 4.128 4.154
cp_fm_cholesky_invert 11 10.9 4.056 4.063 4.056 4.063
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.944 3.944
ot_diis_step 99 11.5 0.012 0.012 3.847 3.847
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.757 3.790
apply_single 110 13.6 0.000 0.000 3.757 3.789
density_rs2pw 110 9.6 0.004 0.005 3.406 3.783
grid_integrate_task_list 110 12.3 3.343 3.564 3.343 3.564
fft_wrap_pw1pw2_140 451 13.1 0.833 0.844 3.434 3.468
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.311 3.318
cp_fm_diag_elpa_base 48 14.0 3.243 3.272 3.308 3.316
make_images_data 4110 15.4 0.038 0.044 2.638 3.175
rs_pw_transfer 902 11.9 0.010 0.010 2.722 3.136
hybrid_alltoall_any 4261 16.3 0.201 0.869 2.494 3.097
multiply_cannon_sync_h2d 8220 15.4 2.911 3.028 2.911 3.028
wfi_extrapolate 11 7.9 0.001 0.001 3.013 3.013
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 2.695 2.696
fft3d_ps 1111 14.6 1.138 1.167 2.671 2.693
calculate_dm_sparse 110 9.5 0.001 0.001 2.592 2.642
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.596 2.598
grid_collocate_task_list 110 9.6 2.280 2.536 2.280 2.536
potential_pw2rs 110 12.3 0.016 0.016 2.483 2.495
calculate_first_density_matrix 1 7.0 0.000 0.000 2.285 2.286
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.249 2.260
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.157 2.177
cp_fm_cholesky_decompose 22 10.9 2.133 2.151 2.133 2.151
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.115 2.123
mp_alltoall_d11v 2046 13.8 1.826 2.121 1.826 2.121
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.773 1.999
mp_allgather_i34 2055 14.4 0.806 1.954 0.806 1.954
dbcsr_complete_redistribute 325 12.2 0.569 0.607 1.643 1.762
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.725 1.728
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.531 1.657
mp_waitany 9240 13.8 1.182 1.615 1.182 1.615
multiply_cannon_metrocomm1 8220 15.4 0.020 0.021 1.197 1.608
jit_kernel_multiply 9 15.8 0.939 1.592 0.939 1.592
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.465 1.477
qs_create_task_list 11 7.9 0.001 0.001 1.232 1.338
generate_qs_task_list 11 8.9 0.375 0.444 1.232 1.338
mp_alltoall_z22v 1111 16.6 1.298 1.333 1.298 1.333
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.223 1.247
rs_gather_matrices 110 12.3 0.325 0.375 0.973 1.240
rs_pw_transfer_RS2PW_140 121 11.5 0.163 0.165 0.830 1.233
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=61.217000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=776.181818, yerr=6.820000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 1.342407E+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 0.018 0.036 95.529 95.530
qs_mol_dyn_low 1 2.0 0.003 0.003 95.125 95.137
qs_forces 11 3.9 0.003 0.003 95.049 95.050
qs_energies 11 4.9 0.001 0.001 90.705 90.708
scf_env_do_scf 11 5.9 0.001 0.001 80.212 80.212
velocity_verlet 10 3.0 0.001 0.001 61.126 61.133
scf_env_do_scf_inner_loop 99 6.5 0.003 0.008 48.570 48.571
init_scf_loop 11 6.9 0.000 0.000 31.559 31.561
dbcsr_multiply_generic 2055 12.4 0.121 0.125 30.053 30.157
prepare_preconditioner 11 7.9 0.000 0.000 29.422 29.429
make_preconditioner 11 8.9 0.000 0.000 29.422 29.429
qs_scf_new_mos 99 7.5 0.001 0.001 29.099 29.135
qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.099 29.134
make_full_inverse_cholesky 11 9.9 0.000 0.000 23.421 28.882
ot_scf_mini 99 9.5 0.002 0.002 27.251 27.272
multiply_cannon 2055 13.4 0.346 0.375 22.099 23.047
multiply_cannon_loop 2055 14.4 0.342 0.344 20.177 20.580
cp_fm_upper_to_full 70 14.2 12.681 18.405 12.681 18.405
rebuild_ks_matrix 110 8.3 0.000 0.001 14.903 14.944
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.014 14.903 14.944
ot_mini 99 10.5 0.001 0.001 14.664 14.686
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.464 13.502
dbcsr_complete_redistribute 325 12.2 1.009 1.033 7.802 11.115
qs_ot_get_derivative 99 11.5 0.001 0.001 10.018 10.039
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.661 9.976
multiply_cannon_multrec 8220 15.4 4.379 4.613 9.585 9.714
mp_waitall_1 84994 16.7 8.743 9.643 8.743 9.643
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.986 9.271
mp_alltoall_i22 605 13.7 5.615 8.945 5.615 8.945
qs_rho_update_rho_low 110 7.6 0.001 0.001 8.374 8.408
calculate_rho_elec 110 8.6 0.226 0.226 8.373 8.408
sum_up_and_integrate 110 10.3 0.150 0.151 7.907 7.920
integrate_v_rspace 110 11.3 0.004 0.004 7.757 7.770
cp_fm_cholesky_invert 11 10.9 7.543 7.549 7.543 7.549
qs_ot_get_p 110 10.4 0.001 0.001 7.129 7.160
make_m2s 4110 13.4 0.043 0.043 5.988 6.589
make_images 4110 14.4 0.876 0.915 5.798 6.397
init_scf_run 11 5.9 0.000 0.001 6.139 6.139
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.139 6.139
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 5.221 5.671
pw_transfer 1331 11.6 0.075 0.076 5.651 5.658
fft_wrap_pw1pw2 1111 12.6 0.009 0.010 5.534 5.542
dbcsr_mm_accdrv_process 11614 15.7 3.298 3.641 5.061 5.307
qs_ot_p2m_diag 48 11.0 0.151 0.156 5.247 5.257
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.734 5.174
apply_single 110 13.6 0.000 0.000 4.733 5.174
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.762 4.765
fft_wrap_pw1pw2_140 451 13.1 1.340 1.345 4.691 4.697
ot_diis_step 99 11.5 0.015 0.016 4.607 4.607
density_rs2pw 110 9.6 0.005 0.005 4.308 4.318
make_images_data 4110 15.4 0.041 0.045 3.366 4.222
hybrid_alltoall_any 4261 16.3 0.256 0.554 3.305 4.169
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.026 4.027
cp_fm_diag_elpa_base 48 14.0 3.457 3.673 4.024 4.025
multiply_cannon_sync_h2d 8220 15.4 3.952 3.957 3.952 3.957
grid_integrate_task_list 110 12.3 3.671 3.725 3.671 3.725
wfi_extrapolate 11 7.9 0.001 0.001 3.693 3.694
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.687 3.688
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.175 3.635
fft3d_ps 1111 14.6 1.305 1.312 3.459 3.470
calculate_dm_sparse 110 9.5 0.001 0.001 3.230 3.264
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.043 3.044
potential_pw2rs 110 12.3 0.021 0.022 2.891 2.896
rs_pw_transfer 902 11.9 0.010 0.011 2.852 2.884
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.708 2.724
cp_fm_cholesky_decompose 22 10.9 2.693 2.721 2.693 2.721
grid_collocate_task_list 110 9.6 2.641 2.661 2.641 2.661
mp_alltoall_d11v 2046 13.8 2.400 2.498 2.400 2.498
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.468 2.481
calculate_first_density_matrix 1 7.0 0.000 0.000 2.314 2.315
qs_env_update_s_mstruct 11 6.9 0.001 0.004 2.214 2.286
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.097 2.167
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.043 2.046
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.936 1.968
qs_create_task_list 11 7.9 0.000 0.001 1.903 1.949
generate_qs_task_list 11 8.9 0.733 0.785 1.903 1.949
mp_alltoall_z22v 1111 16.6 1.903 1.916 1.903 1.916
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=95.530000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1206.090909, yerr=57.994443
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/15/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 1.094965E+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 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 630.804480E+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 3175955383376
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 57767.
MP_Allreduce 11004 802.
MP_Sync 87
MP_Alltoall 2226 3570369.
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 0.018 0.047 216.401 216.404
qs_mol_dyn_low 1 2.0 0.006 0.027 215.949 215.963
qs_forces 11 3.9 0.005 0.006 215.829 215.832
qs_energies 11 4.9 0.001 0.002 210.046 210.062
scf_env_do_scf 11 5.9 0.001 0.001 193.038 193.044
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 170.586 170.590
velocity_verlet 10 3.0 0.001 0.001 129.725 129.728
qs_scf_new_mos 117 7.6 0.001 0.001 128.321 128.610
qs_scf_loop_do_ot 117 8.6 0.001 0.001 128.320 128.609
dbcsr_multiply_generic 2507 12.6 0.179 0.183 125.202 125.845
ot_scf_mini 117 9.6 0.003 0.003 121.705 121.963
multiply_cannon 2507 13.6 0.241 0.251 101.149 103.243
multiply_cannon_loop 2507 14.6 2.036 2.117 98.768 100.890
ot_mini 117 10.6 0.001 0.001 66.460 66.707
multiply_cannon_multrec 60168 15.6 33.407 35.589 41.607 43.423
qs_ot_get_derivative 117 11.6 0.001 0.001 41.721 41.971
rebuild_ks_matrix 128 8.3 0.001 0.001 34.979 35.503
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.017 34.978 35.503
qs_ot_get_p 128 10.4 0.001 0.001 33.276 33.574
mp_waitall_1 267128 16.5 29.386 32.470 29.386 32.470
qs_ks_update_qs_env 128 7.6 0.001 0.001 31.445 31.911
multiply_cannon_sync_h2d 60168 15.6 27.599 29.259 27.599 29.259
qs_ot_p2m_diag 83 11.4 0.079 0.092 26.332 26.405
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.402 25.122
apply_single 128 13.6 0.001 0.001 24.401 25.122
ot_diis_step 117 11.6 0.008 0.008 24.464 24.465
cp_dbcsr_syevd 83 12.4 0.004 0.005 23.792 23.794
init_scf_loop 11 6.9 0.000 0.000 22.370 22.371
cp_fm_diag_elpa 83 13.4 0.000 0.000 20.236 20.270
cp_fm_diag_elpa_base 83 14.4 20.138 20.193 20.232 20.266
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 19.393 19.546
prepare_preconditioner 11 7.9 0.000 0.000 17.616 17.662
make_preconditioner 11 8.9 0.000 0.000 17.616 17.662
multiply_cannon_metrocomm3 60168 15.6 0.115 0.120 15.449 17.354
make_full_inverse_cholesky 11 9.9 0.000 0.000 16.808 16.992
sum_up_and_integrate 128 10.3 0.089 0.108 15.338 15.351
integrate_v_rspace 128 11.3 0.004 0.005 15.248 15.265
make_m2s 5014 13.6 0.103 0.110 14.600 15.017
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.923 15.017
calculate_rho_elec 128 8.7 0.046 0.065 14.922 15.016
make_images 5014 14.6 0.404 0.422 14.420 14.847
init_scf_run 11 5.9 0.000 0.003 12.717 12.718
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.717 12.718
density_rs2pw 128 9.7 0.007 0.008 8.257 11.464
rs_pw_transfer 1046 11.9 0.016 0.019 7.345 10.648
cp_fm_cholesky_invert 11 10.9 10.203 10.211 10.203 10.211
mp_sum_l 7870 13.0 8.199 9.711 8.199 9.711
wfi_extrapolate 11 7.9 0.001 0.001 9.431 9.431
pw_transfer 1547 11.6 0.076 0.092 8.494 8.729
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.358 8.585
calculate_dm_sparse 128 9.5 0.001 0.001 8.420 8.533
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 8.288 8.521
dbcsr_mm_accdrv_process 124484 16.2 3.044 3.259 7.762 8.297
make_images_data 5014 15.6 0.065 0.072 7.176 8.182
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.830 7.898
multiply_cannon_metrocomm1 60168 15.6 0.087 0.091 6.280 7.785
hybrid_alltoall_any 5200 16.5 0.293 2.268 6.213 7.593
grid_integrate_task_list 128 12.3 7.051 7.583 7.051 7.583
fft_wrap_pw1pw2_140 523 13.2 1.261 1.324 6.897 7.129
cp_dbcsr_sm_fm_multiply 37 9.5 0.003 0.005 6.764 6.773
fft3d_ps 1291 14.7 2.149 2.779 6.068 6.447
mp_waitany 16020 13.9 2.823 6.111 2.823 6.111
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.861 6.001
grid_collocate_task_list 128 9.7 4.551 5.804 4.551 5.804
potential_pw2rs 128 12.3 0.009 0.011 5.649 5.727
mp_alltoall_d11v 2415 14.1 4.518 5.672 4.518 5.672
rs_pw_transfer_RS2PW_140 139 11.5 0.282 0.304 2.178 5.405
cp_fm_cholesky_decompose 22 10.9 5.155 5.169 5.155 5.169
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=216.404000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=596.181818, yerr=6.886326
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/16/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 2.183246E+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 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 829.816832E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100943E+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 1463834845264
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58220.
MP_Allreduce 11003 967.
MP_Sync 87
MP_Alltoall 1969 5618843.
MP_SendRecv 12032 47072.
MP_ISendRecv 12032 47072.
MP_Wait 25916
MP_ISend 11748 212467.
MP_IRecv 11748 212467.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.029 0.067 201.173 201.175
qs_mol_dyn_low 1 2.0 0.003 0.005 200.691 200.705
qs_forces 11 3.9 0.005 0.007 200.597 200.600
qs_energies 11 4.9 0.002 0.009 193.661 193.671
scf_env_do_scf 11 5.9 0.003 0.019 176.788 176.798
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 140.530 140.532
velocity_verlet 10 3.0 0.001 0.002 126.430 126.432
dbcsr_multiply_generic 2507 12.6 0.188 0.194 99.398 100.917
qs_scf_new_mos 117 7.6 0.001 0.001 99.054 99.467
qs_scf_loop_do_ot 117 8.6 0.001 0.002 99.053 99.466
ot_scf_mini 117 9.6 0.004 0.005 94.193 94.706
multiply_cannon 2507 13.6 0.475 0.529 78.022 82.312
multiply_cannon_loop 2507 14.6 1.251 1.286 74.472 78.048
ot_mini 117 10.6 0.001 0.001 51.201 51.654
mp_waitall_1 214728 16.6 26.326 41.621 26.326 41.621
multiply_cannon_multrec 30084 15.6 21.820 26.990 31.451 36.816
init_scf_loop 11 6.9 0.001 0.007 36.142 36.143
rebuild_ks_matrix 128 8.3 0.001 0.001 34.279 34.732
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.024 34.278 34.732
prepare_preconditioner 11 7.9 0.000 0.001 31.587 31.646
make_preconditioner 11 8.9 0.000 0.002 31.587 31.646
multiply_cannon_metrocomm3 30084 15.6 0.091 0.100 16.374 31.637
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.837 31.238
make_full_inverse_cholesky 11 9.9 0.000 0.001 30.214 30.782
qs_ot_get_derivative 117 11.6 0.001 0.002 29.031 29.537
qs_ot_get_p 128 10.4 0.001 0.001 24.428 25.043
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.142 23.404
apply_single 128 13.6 0.001 0.001 22.142 23.404
ot_diis_step 117 11.6 0.014 0.015 21.974 21.976
multiply_cannon_sync_h2d 30084 15.6 19.086 21.068 19.086 21.068
qs_ot_p2m_diag 83 11.4 0.187 0.216 19.329 19.413
cp_fm_cholesky_invert 11 10.9 18.524 18.538 18.524 18.538
cp_dbcsr_syevd 83 12.4 0.006 0.009 18.046 18.048
make_m2s 5014 13.6 0.089 0.094 15.131 16.758
make_images 5014 14.6 1.152 1.337 14.924 16.553
sum_up_and_integrate 128 10.3 0.116 0.132 15.438 15.468
integrate_v_rspace 128 11.3 0.004 0.004 15.322 15.357
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.823 14.876
calculate_rho_elec 128 8.7 0.088 0.104 14.823 14.876
cp_fm_diag_elpa 83 13.4 0.001 0.002 14.581 14.610
cp_fm_diag_elpa_base 83 14.4 14.293 14.425 14.572 14.601
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 11.591 11.967
init_scf_run 11 5.9 0.000 0.002 11.951 11.953
scf_env_initial_rho_setup 11 6.9 0.001 0.002 11.951 11.953
make_images_data 5014 15.6 0.063 0.071 9.252 11.154
multiply_cannon_metrocomm4 27577 15.6 0.098 0.112 3.796 10.918
density_rs2pw 128 9.7 0.007 0.009 8.181 10.620
hybrid_alltoall_any 5200 16.5 0.344 1.506 7.779 10.532
mp_irecv_dv 69486 16.3 3.599 10.529 3.599 10.529
dbcsr_mm_accdrv_process 62242 16.2 4.568 5.089 9.104 9.676
rs_pw_transfer 1046 11.9 0.014 0.017 6.936 9.422
pw_transfer 1547 11.6 0.087 0.097 9.102 9.171
fft_wrap_pw1pw2 1291 12.7 0.011 0.011 8.876 8.950
wfi_extrapolate 11 7.9 0.001 0.001 8.678 8.678
cp_fm_cholesky_decompose 22 10.9 8.125 8.206 8.125 8.206
grid_integrate_task_list 128 12.3 7.181 7.743 7.181 7.743
fft_wrap_pw1pw2_140 523 13.2 1.339 1.358 7.638 7.733
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.383 7.079
calculate_dm_sparse 128 9.5 0.001 0.001 6.530 6.656
mp_sum_l 7870 13.0 4.489 6.380 4.489 6.380
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.317 6.327
fft3d_ps 1291 14.7 2.809 2.986 6.178 6.224
grid_collocate_task_list 128 9.7 4.698 5.876 4.698 5.876
potential_pw2rs 128 12.3 0.015 0.017 5.686 5.712
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.514 5.608
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.431 5.573
mp_allgather_i34 2507 14.6 2.123 5.432 2.123 5.432
mp_alltoall_d11v 2415 14.1 4.252 5.162 4.252 5.162
mp_waitany 11748 13.9 2.600 5.097 2.600 5.097
rs_pw_transfer_RS2PW_140 139 11.5 0.352 0.376 2.127 4.563
dbcsr_complete_redistribute 395 12.7 0.767 0.844 3.320 4.224
mp_sum_d 4470 12.1 2.782 4.110 2.782 4.110
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=201.175000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=790.727273, yerr=4.024717
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430460020736 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613089636352 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228663E+12 0.0% 0.0% 100.0%
flops max/rank 2.950886E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806316384 0.0% 0.0% 100.0%
number of processed stacks 4015680 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1694.9
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 943.054848E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1051232
MPI messages size (bytes):
total size 2.737039E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.603648E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 284160 37245419520
131072 < size <= 4194304 665408 1004401590272
4194304 < size <= 16777216 66080 937889764224
16777216 < size 28896 757491302400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4024 57903.
MP_Allreduce 11057 1004.
MP_Sync 87
MP_Alltoall 1724 9394172.
MP_SendRecv 7998 75008.
MP_ISendRecv 7998 75008.
MP_Wait 21986
MP_ISend 11836 275177.
MP_IRecv 11836 275177.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.065 0.143 192.119 192.120
qs_mol_dyn_low 1 2.0 0.003 0.003 191.520 191.533
qs_forces 11 3.9 0.005 0.006 190.893 190.896
qs_energies 11 4.9 0.004 0.028 184.177 184.184
scf_env_do_scf 11 5.9 0.001 0.002 167.881 167.884
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 129.138 129.140
velocity_verlet 10 3.0 0.003 0.013 121.685 121.703
qs_scf_new_mos 118 7.6 0.001 0.001 89.218 89.640
qs_scf_loop_do_ot 118 8.6 0.001 0.001 89.217 89.640
dbcsr_multiply_generic 2527 12.6 0.182 0.187 86.258 87.364
ot_scf_mini 118 9.6 0.004 0.004 85.042 85.516
multiply_cannon 2527 13.6 0.508 0.523 63.839 69.017
multiply_cannon_loop 2527 14.6 0.866 0.896 60.173 64.298
ot_mini 118 10.6 0.001 0.001 46.736 47.207
init_scf_loop 11 6.9 0.001 0.003 38.632 38.634
mp_waitall_1 171878 16.6 26.106 37.002 26.106 37.002
prepare_preconditioner 11 7.9 0.000 0.001 34.490 34.546
make_preconditioner 11 8.9 0.000 0.001 34.490 34.546
make_full_inverse_cholesky 11 9.9 0.000 0.001 32.079 33.530
rebuild_ks_matrix 129 8.3 0.001 0.001 31.757 32.246
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.020 31.756 32.245
qs_ks_update_qs_env 129 7.6 0.001 0.001 28.629 29.079
multiply_cannon_metrocomm3 20216 15.6 0.058 0.062 15.472 28.167
multiply_cannon_multrec 20216 15.6 13.695 17.091 22.907 27.481
qs_ot_get_derivative 118 11.6 0.002 0.002 26.853 27.321
qs_ot_get_p 129 10.4 0.001 0.001 23.425 24.014
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 21.711 23.730
apply_single 129 13.6 0.001 0.001 21.711 23.730
ot_diis_step 118 11.6 0.018 0.018 19.764 19.764
qs_ot_p2m_diag 83 11.4 0.265 0.273 18.926 18.964
cp_dbcsr_syevd 83 12.4 0.006 0.011 17.785 17.787
cp_fm_cholesky_invert 11 10.9 16.838 16.852 16.838 16.852
multiply_cannon_sync_h2d 20216 15.6 14.558 16.523 14.558 16.523
make_m2s 5054 13.6 0.082 0.085 15.216 16.437
make_images 5054 14.6 1.178 1.265 14.981 16.201
sum_up_and_integrate 129 10.3 0.135 0.147 15.491 15.515
integrate_v_rspace 129 11.3 0.004 0.004 15.356 15.380
qs_rho_update_rho_low 129 7.7 0.001 0.001 15.223 15.258
calculate_rho_elec 129 8.7 0.131 0.147 15.222 15.257
cp_fm_diag_elpa 83 13.4 0.001 0.002 14.170 14.182
cp_fm_diag_elpa_base 83 14.4 13.710 13.873 14.162 14.172
init_scf_run 11 5.9 0.000 0.002 11.125 11.125
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.125 11.125
dbcsr_mm_accdrv_process 41830 16.2 4.800 7.775 8.663 11.093
make_images_data 5054 15.6 0.060 0.068 9.227 10.965
hybrid_alltoall_any 5240 16.5 0.440 2.013 8.124 10.018
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.510 9.865
density_rs2pw 129 9.7 0.007 0.009 8.013 9.854
pw_transfer 1559 11.6 0.088 0.105 9.236 9.340
fft_wrap_pw1pw2 1301 12.7 0.011 0.012 9.008 9.119
multiply_cannon_metrocomm4 17689 15.6 0.064 0.075 3.373 9.095
mp_irecv_dv 50625 16.2 3.249 8.834 3.249 8.834
cp_fm_cholesky_decompose 22 10.9 8.447 8.467 8.447 8.467
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 5.868 8.319
rs_pw_transfer 1054 12.0 0.014 0.015 6.474 8.317
grid_integrate_task_list 129 12.3 7.321 7.956 7.321 7.956
fft_wrap_pw1pw2_140 527 13.2 1.428 1.459 7.744 7.866
wfi_extrapolate 11 7.9 0.001 0.001 7.714 7.714
mp_sum_l 7930 13.1 5.388 7.566 5.388 7.566
cp_fm_upper_to_full 105 14.8 5.831 7.405 5.831 7.405
dbcsr_complete_redistribute 395 12.7 1.165 1.196 4.671 6.500
fft3d_ps 1301 14.7 2.760 2.983 6.100 6.178
calculate_dm_sparse 129 9.5 0.001 0.002 5.862 5.957
grid_collocate_task_list 129 9.7 4.940 5.873 4.940 5.873
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.723 5.733
potential_pw2rs 129 12.3 0.021 0.023 5.488 5.520
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.416 5.244
mp_alltoall_d11v 2423 14.1 4.463 5.121 4.463 5.121
mp_allgather_i34 2527 14.6 2.171 4.949 2.171 4.949
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.787 4.914
mp_waitany 11836 13.9 2.489 4.429 2.489 4.429
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.243 4.288
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.391 4.197
mp_alltoall_i22 718 14.1 2.033 3.976 2.033 3.976
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=192.120000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=890.272727, yerr=11.592246
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 4.387242E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 6026880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1129.3
marketing flops 145.651870E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 1.148752E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1153224
MPI messages size (bytes):
total size 2.039489E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.768511E+06
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 0 0
8192 < size <= 32768 396 8650752
32768 < size <= 131072 322096 36390305792
131072 < size <= 4194304 721976 792118951936
4194304 < size <= 16777216 70800 669922227920
16777216 < size 30960 541065216000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 57632.
MP_Allreduce 11104 1087.
MP_Sync 88
MP_Alltoall 1724 12509575.
MP_SendRecv 5934 75008.
MP_ISendRecv 5934 75008.
MP_Wait 22612
MP_ISend 15064 244788.
MP_IRecv 15064 244788.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.044 200.372 200.383
qs_mol_dyn_low 1 2.0 0.003 0.003 199.966 199.979
qs_forces 11 3.9 0.004 0.005 199.842 199.844
qs_energies 11 4.9 0.003 0.020 192.572 192.581
scf_env_do_scf 11 5.9 0.001 0.001 175.611 175.619
velocity_verlet 10 3.0 0.001 0.001 131.887 131.890
scf_env_do_scf_inner_loop 118 6.6 0.003 0.010 125.501 125.502
qs_scf_new_mos 118 7.6 0.001 0.001 86.940 87.252
qs_scf_loop_do_ot 118 8.6 0.001 0.001 86.939 87.251
ot_scf_mini 118 9.6 0.003 0.004 82.443 82.766
dbcsr_multiply_generic 2529 12.6 0.188 0.193 81.884 82.620
multiply_cannon 2529 13.6 0.563 0.599 56.934 59.305
multiply_cannon_loop 2529 14.6 1.187 1.219 52.948 54.637
init_scf_loop 11 6.9 0.001 0.003 49.980 49.982
prepare_preconditioner 11 7.9 0.000 0.000 45.848 45.871
make_preconditioner 11 8.9 0.000 0.001 45.848 45.871
ot_mini 118 10.6 0.001 0.001 44.918 45.249
make_full_inverse_cholesky 11 9.9 0.000 0.000 39.504 44.536
multiply_cannon_multrec 30348 15.6 14.181 20.080 27.339 33.241
rebuild_ks_matrix 129 8.3 0.001 0.001 30.493 30.744
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.020 30.492 30.744
mp_waitall_1 149172 16.7 18.196 27.766 18.196 27.766
qs_ks_update_qs_env 129 7.6 0.001 0.001 27.493 27.721
qs_ot_get_derivative 118 11.6 0.002 0.002 25.185 25.515
qs_ot_get_p 129 10.4 0.001 0.001 23.602 23.967
make_m2s 5058 13.6 0.097 0.101 20.472 21.440
make_images 5058 14.6 1.955 2.250 20.161 21.129
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.109 19.678
apply_single 129 13.6 0.001 0.001 19.108 19.678
ot_diis_step 118 11.6 0.018 0.019 19.608 19.609
qs_ot_p2m_diag 84 11.4 0.347 0.394 19.290 19.341
cp_fm_cholesky_invert 11 10.9 18.759 18.769 18.759 18.769
cp_dbcsr_syevd 84 12.4 0.005 0.006 17.919 17.921
cp_fm_upper_to_full 106 14.8 11.421 16.890 11.421 16.890
multiply_cannon_metrocomm3 30348 15.6 0.047 0.050 6.971 16.571
sum_up_and_integrate 129 10.3 0.141 0.154 15.308 15.328
integrate_v_rspace 129 11.3 0.004 0.004 15.166 15.191
qs_rho_update_rho_low 129 7.7 0.001 0.001 15.130 15.167
calculate_rho_elec 129 8.7 0.174 0.190 15.129 15.166
cp_fm_diag_elpa 84 13.4 0.000 0.001 14.425 14.436
cp_fm_diag_elpa_base 84 14.4 13.399 13.744 14.417 14.430
dbcsr_mm_accdrv_process 62780 16.2 8.131 10.293 12.728 13.731
dbcsr_complete_redistribute 397 12.7 1.551 1.704 9.093 12.839
make_images_data 5058 15.6 0.064 0.069 10.897 12.779
multiply_cannon_sync_h2d 30348 15.6 11.758 12.667 11.758 12.667
hybrid_alltoall_any 5245 16.5 0.532 2.231 9.783 11.534
copy_fm_to_dbcsr 210 11.7 0.002 0.002 7.671 11.440
init_scf_run 11 5.9 0.000 0.001 10.997 10.998
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.996 10.998
qs_ot_get_derivative_diag 78 12.4 0.002 0.002 9.837 10.086
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.322 10.011
mp_alltoall_i22 720 14.1 5.600 9.374 5.600 9.374
pw_transfer 1559 11.6 0.087 0.100 9.211 9.315
fft_wrap_pw1pw2 1301 12.7 0.011 0.011 8.982 9.092
density_rs2pw 129 9.7 0.006 0.007 7.548 8.847
cp_fm_cholesky_decompose 22 10.9 8.648 8.775 8.648 8.775
grid_integrate_task_list 129 12.3 7.544 8.133 7.544 8.133
fft_wrap_pw1pw2_140 527 13.2 1.576 1.612 7.863 7.993
multiply_cannon_metrocomm4 25290 15.6 0.084 0.097 2.908 7.903
wfi_extrapolate 11 7.9 0.001 0.001 7.775 7.775
mp_irecv_dv 76751 16.2 2.751 7.613 2.751 7.613
rs_pw_transfer 1054 12.0 0.013 0.014 5.707 7.104
calculate_dm_sparse 129 9.5 0.001 0.001 6.246 6.323
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.212 6.303
mp_alltoall_d11v 2429 14.1 4.913 6.065 4.913 6.065
fft3d_ps 1301 14.7 2.854 2.930 5.870 5.952
grid_collocate_task_list 129 9.7 5.119 5.851 5.119 5.851
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.439 5.485
potential_pw2rs 129 12.3 0.024 0.024 5.062 5.102
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 4.458 4.460
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.300 4.386
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.315 4.373
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=200.383000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1077.181818, yerr=21.451078
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/19/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 5.865089E+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 1960712 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3445.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.521086E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 240672
MPI messages size (bytes):
total size 1.331455E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.532238E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 113904 59718500352
4194304 < size <= 16777216 104976 550376570880
16777216 < size 20208 721350232272
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8931 51.
MP_Alltoall 9654 799394.
MP_ISend 40068 2102573.
MP_IRecv 40068 2101676.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58209.
MP_Allreduce 11002 1175.
MP_Sync 87
MP_Alltoall 1712 18838222.
MP_SendRecv 3840 122880.
MP_ISendRecv 3840 122880.
MP_Wait 16122
MP_ISend 10680 423556.
MP_IRecv 10680 423556.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.104 185.575 185.599
qs_mol_dyn_low 1 2.0 0.003 0.003 185.066 185.072
qs_forces 11 3.9 0.004 0.004 184.904 184.908
qs_energies 11 4.9 0.001 0.001 177.218 177.224
scf_env_do_scf 11 5.9 0.001 0.001 159.480 159.495
velocity_verlet 10 3.0 0.001 0.001 121.142 121.147
scf_env_do_scf_inner_loop 117 6.6 0.004 0.010 119.766 119.767
qs_scf_new_mos 117 7.6 0.001 0.001 80.216 80.312
qs_scf_loop_do_ot 117 8.6 0.001 0.001 80.215 80.312
ot_scf_mini 117 9.6 0.003 0.004 75.778 75.802
dbcsr_multiply_generic 2507 12.6 0.181 0.192 74.982 75.253
multiply_cannon 2507 13.6 0.585 0.619 54.675 59.992
multiply_cannon_loop 2507 14.6 0.446 0.457 49.897 50.407
ot_mini 117 10.6 0.001 0.001 40.491 40.531
init_scf_loop 11 6.9 0.000 0.000 39.552 39.553
prepare_preconditioner 11 7.9 0.000 0.000 35.564 35.589
make_preconditioner 11 8.9 0.000 0.000 35.564 35.589
make_full_inverse_cholesky 11 9.9 0.000 0.000 33.306 33.623
mp_waitall_1 125778 16.7 26.236 33.591 26.236 33.591
rebuild_ks_matrix 128 8.3 0.001 0.001 30.396 30.467
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.018 30.396 30.467
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.621 27.683
qs_ot_get_p 128 10.4 0.001 0.001 22.334 22.390
multiply_cannon_multrec 10028 15.6 10.423 14.182 17.859 21.233
cp_fm_cholesky_invert 11 10.9 20.824 20.833 20.824 20.833
qs_ot_get_derivative 117 11.6 0.001 0.002 20.534 20.560
ot_diis_step 117 11.6 0.020 0.020 19.855 19.855
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.440 19.687
apply_single 128 13.6 0.001 0.001 19.440 19.687
make_m2s 5014 13.6 0.067 0.071 16.483 18.713
multiply_cannon_metrocomm3 10028 15.6 0.023 0.024 12.205 18.505
qs_ot_p2m_diag 83 11.4 0.495 0.501 18.449 18.462
make_images 5014 14.6 2.293 2.603 16.177 18.409
cp_dbcsr_syevd 83 12.4 0.005 0.005 17.168 17.170
qs_rho_update_rho_low 128 7.7 0.001 0.001 15.906 15.948
calculate_rho_elec 128 8.7 0.256 0.267 15.905 15.947
sum_up_and_integrate 128 10.3 0.180 0.189 15.632 15.674
integrate_v_rspace 128 11.3 0.004 0.004 15.452 15.504
cp_fm_diag_elpa 83 13.4 0.000 0.000 13.658 13.666
cp_fm_diag_elpa_base 83 14.4 13.367 13.485 13.650 13.661
make_images_data 5014 15.6 0.051 0.061 9.860 12.497
multiply_cannon_sync_h2d 10028 15.6 11.630 12.462 11.630 12.462
hybrid_alltoall_any 5200 16.5 0.817 3.622 9.631 12.020
init_scf_run 11 5.9 0.000 0.002 10.832 10.833
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.832 10.832
pw_transfer 1547 11.6 0.087 0.097 9.727 9.754
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 9.502 9.537
cp_fm_cholesky_decompose 22 10.9 8.887 8.992 8.887 8.992
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 8.351 8.364
density_rs2pw 128 9.7 0.006 0.006 7.413 8.295
fft_wrap_pw1pw2_140 523 13.2 1.912 1.955 8.221 8.262
grid_integrate_task_list 128 12.3 7.784 8.114 7.784 8.114
mp_allgather_i34 2507 14.6 2.977 7.744 2.977 7.744
dbcsr_mm_accdrv_process 20762 16.1 2.685 3.467 7.067 7.686
wfi_extrapolate 11 7.9 0.001 0.001 7.660 7.660
multiply_cannon_metrocomm1 10028 15.6 0.028 0.029 4.565 7.187
calculate_dm_sparse 128 9.5 0.001 0.001 6.268 6.357
grid_collocate_task_list 128 9.7 5.394 6.056 5.394 6.056
fft3d_ps 1291 14.7 2.768 2.845 5.975 6.025
rs_pw_transfer 1046 11.9 0.013 0.013 5.021 5.941
dbcsr_complete_redistribute 395 12.7 2.118 2.225 5.432 5.912
mp_alltoall_d11v 2415 14.1 5.352 5.842 5.352 5.842
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.303 5.304
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.174 5.181
potential_pw2rs 128 12.3 0.027 0.028 4.988 5.017
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.054 4.078
multiply_cannon_metrocomm4 7521 15.6 0.024 0.027 1.863 4.054
mp_irecv_dv 28860 15.9 1.826 3.977 1.826 3.977
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.582 3.965
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.570 3.913
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.825 3.889
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.654 3.737
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=185.599000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1410.000000, yerr=49.248904
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430454546432 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1979979923456 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992006770688 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613065416704 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239182565376 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239182565376 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.250208E+12 0.0% 0.0% 100.0%
flops max/rank 11.788211E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806645728 0.0% 0.0% 100.0%
number of processed stacks 1983232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3432.1
marketing flops 145.668111E+12
-------------------------------------------------------------------------------
# multiplications 2537
max memory usage/rank 3.009446E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101480
MPI messages size (bytes):
total size 1.145238E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.285357E+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 45968 35701915648
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592540656
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4056 58383.
MP_Allreduce 11137 1510.
MP_Sync 87
MP_Alltoall 1724 36993632.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.043 315.586 315.589
qs_mol_dyn_low 1 2.0 0.003 0.003 315.014 315.028
qs_forces 11 3.9 0.004 0.005 314.840 314.845
qs_energies 11 4.9 0.001 0.001 305.367 305.376
scf_env_do_scf 11 5.9 0.001 0.001 282.536 282.555
velocity_verlet 10 3.0 0.001 0.001 226.807 226.826
scf_env_do_scf_inner_loop 118 6.6 0.005 0.010 151.511 151.514
init_scf_loop 11 6.9 0.000 0.000 130.752 130.755
prepare_preconditioner 11 7.9 0.000 0.000 125.525 125.547
make_preconditioner 11 8.9 0.000 0.000 125.525 125.547
make_full_inverse_cholesky 11 9.9 0.000 0.000 101.837 122.643
qs_scf_new_mos 118 7.6 0.001 0.001 98.592 98.748
qs_scf_loop_do_ot 118 8.6 0.001 0.001 98.592 98.747
ot_scf_mini 118 9.6 0.004 0.004 93.747 93.851
dbcsr_multiply_generic 2537 12.6 0.214 0.228 85.780 86.588
cp_fm_upper_to_full 105 14.8 52.126 74.938 52.126 74.938
multiply_cannon 2537 13.6 0.710 0.753 59.489 60.534
multiply_cannon_loop 2537 14.6 0.477 0.489 55.540 57.563
ot_mini 118 10.6 0.001 0.001 46.761 46.889
dbcsr_complete_redistribute 395 12.7 3.987 4.037 29.633 42.224
rebuild_ks_matrix 129 8.3 0.001 0.001 39.682 39.776
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 39.681 39.776
cp_fm_cholesky_invert 11 10.9 39.003 39.012 39.003 39.012
copy_fm_to_dbcsr 209 11.7 0.002 0.002 26.107 38.657
qs_ks_update_qs_env 129 7.6 0.001 0.001 36.578 36.660
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 23.641 36.114
mp_waitall_1 104900 16.8 29.656 34.522 29.656 34.522
mp_alltoall_i22 718 14.1 21.426 33.981 21.426 33.981
qs_ot_get_p 129 10.4 0.001 0.001 31.126 31.209
qs_ot_p2m_diag 83 11.4 0.878 0.884 26.568 26.613
qs_ot_get_derivative 118 11.6 0.002 0.002 26.152 26.265
cp_dbcsr_syevd 83 12.4 0.006 0.006 24.663 24.667
make_m2s 5074 13.6 0.078 0.083 21.152 22.204
make_images 5074 14.6 3.816 4.031 20.667 21.723
qs_rho_update_rho_low 129 7.7 0.001 0.001 21.636 21.659
calculate_rho_elec 129 8.7 0.485 0.485 21.635 21.659
cp_fm_diag_elpa 83 13.4 0.000 0.000 20.962 20.963
cp_fm_diag_elpa_base 83 14.4 16.404 18.172 20.957 20.958
multiply_cannon_metrocomm3 10148 15.6 0.023 0.024 19.207 20.921
sum_up_and_integrate 129 10.3 0.324 0.326 20.688 20.775
ot_diis_step 118 11.6 0.023 0.024 20.556 20.556
integrate_v_rspace 129 11.3 0.004 0.004 20.364 20.453
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.651 19.999
apply_single 129 13.6 0.001 0.001 19.651 19.999
multiply_cannon_multrec 10148 15.6 10.543 12.274 17.967 18.083
multiply_cannon_sync_h2d 10148 15.6 15.785 15.801 15.785 15.801
make_images_data 5074 15.6 0.060 0.065 11.487 13.624
hybrid_alltoall_any 5260 16.5 1.310 3.061 11.413 13.534
init_scf_run 11 5.9 0.000 0.001 12.595 12.595
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.594 12.595
pw_transfer 1559 11.6 0.094 0.094 12.508 12.520
fft_wrap_pw1pw2 1301 12.7 0.012 0.012 12.272 12.284
mp_alltoall_d11v 2423 14.1 9.403 11.074 9.403 11.074
fft_wrap_pw1pw2_140 527 13.2 3.107 3.172 10.769 10.782
cp_fm_cholesky_decompose 22 10.9 10.580 10.635 10.580 10.635
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 10.297 10.379
wfi_extrapolate 11 7.9 0.001 0.001 9.455 9.455
dbcsr_mm_accdrv_process 20966 16.1 3.794 5.766 7.181 9.012
density_rs2pw 129 9.7 0.006 0.006 8.837 8.972
grid_integrate_task_list 129 12.3 8.634 8.806 8.634 8.806
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 8.090 8.091
fft3d_ps 1301 14.7 2.862 2.878 7.297 7.337
calculate_dm_sparse 129 9.5 0.001 0.001 6.685 6.797
rs_scatter_matrices 140 9.7 3.709 4.650 6.349 6.604
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.371 6.487
grid_collocate_task_list 129 9.7 6.419 6.448 6.419 6.448
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.304 6.436
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=315.589000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2714.727273, yerr=148.204460
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 1.261576E+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
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 0.017 0.032 84.507 84.509
qs_energies 1 2.0 0.000 0.000 84.006 84.015
ls_scf 1 3.0 0.000 0.000 83.085 83.095
dbcsr_multiply_generic 111 6.7 0.014 0.015 72.016 72.162
multiply_cannon 111 7.7 0.018 0.021 55.588 56.688
multiply_cannon_loop 111 8.7 0.210 0.226 52.196 53.548
ls_scf_main 1 4.0 0.000 0.000 52.118 52.119
density_matrix_trs4 2 5.0 0.002 0.003 46.798 46.855
ls_scf_init_scf 1 4.0 0.000 0.000 28.019 28.020
ls_scf_init_matrix_S 1 5.0 0.000 0.000 26.942 26.987
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 24.862 24.877
mp_waitall_1 11031 10.9 21.295 23.677 21.295 23.677
multiply_cannon_multrec 2664 9.7 8.224 8.974 15.558 17.314
multiply_cannon_sync_h2d 2664 9.7 14.279 16.265 14.279 16.265
make_m2s 222 7.7 0.008 0.011 12.955 13.452
make_images 222 8.7 0.100 0.109 12.933 13.430
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 8.815 11.824
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.269 8.613
make_images_data 222 9.7 0.004 0.005 7.540 8.051
dbcsr_mm_accdrv_process 4760 10.4 0.515 0.646 6.946 7.919
hybrid_alltoall_any 227 10.6 0.216 1.835 6.480 7.743
dbcsr_mm_accdrv_process_sort 4760 11.4 6.229 7.139 6.229 7.139
calculate_norms 4752 9.8 5.621 6.252 5.621 6.252
apply_matrix_preconditioner 6 5.3 0.000 0.001 4.773 4.862
mp_sum_l 807 5.4 2.930 4.241 2.930 4.241
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.043 3.955
mp_irecv_dv 6231 10.9 2.027 3.932 2.027 3.932
arnoldi_extremal 4 6.8 0.000 0.000 3.450 3.477
arnoldi_normal_ev 4 7.8 0.001 0.002 3.450 3.477
build_subspace 16 8.4 0.009 0.012 3.355 3.357
make_images_sizes 222 9.7 0.000 0.000 0.679 3.240
mp_alltoall_i44 222 10.7 0.679 3.240 0.679 3.240
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.169 3.232
ls_scf_post 1 4.0 0.000 0.000 2.948 2.957
ls_scf_store_result 1 5.0 0.000 0.000 2.768 2.809
dbcsr_special_finalize 555 9.7 0.005 0.006 2.387 2.711
dbcsr_merge_single_wm 555 10.7 0.457 0.557 2.378 2.702
make_images_pack 222 9.7 2.202 2.624 2.204 2.625
dbcsr_matrix_vector_mult 304 9.0 0.006 0.013 2.379 2.625
dbcsr_sort_data 658 11.4 2.181 2.481 2.181 2.481
dbcsr_matrix_vector_mult_local 304 10.0 2.063 2.463 2.065 2.465
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.262 2.307
buffer_matrices_ensure_size 222 8.7 1.756 2.087 1.756 2.087
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.790 1.791
rebuild_ks_matrix 3 7.3 0.000 0.000 1.780 1.782
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.780 1.781
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.509000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1131.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 2.085241E+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 0.031 0.073 91.461 91.483
qs_energies 1 2.0 0.000 0.001 90.893 90.897
ls_scf 1 3.0 0.000 0.000 89.576 89.579
dbcsr_multiply_generic 111 6.7 0.015 0.017 75.427 75.734
multiply_cannon 111 7.7 0.028 0.038 53.224 57.317
ls_scf_main 1 4.0 0.000 0.000 55.149 55.156
multiply_cannon_loop 111 8.7 0.116 0.124 49.942 53.273
density_matrix_trs4 2 5.0 0.002 0.003 49.404 49.585
ls_scf_init_scf 1 4.0 0.000 0.000 30.895 30.896
mp_waitall_1 9105 10.9 21.277 30.548 21.277 30.548
ls_scf_init_matrix_S 1 5.0 0.000 0.001 29.692 29.796
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.254 27.271
multiply_cannon_multrec 1332 9.7 13.139 16.269 22.346 26.634
multiply_cannon_metrocomm3 1332 9.7 0.006 0.007 11.821 20.934
make_m2s 222 7.7 0.006 0.007 15.462 16.109
make_images 222 8.7 1.573 1.972 15.432 16.080
dbcsr_mm_accdrv_process 4041 10.4 0.297 0.452 8.807 10.396
dbcsr_mm_accdrv_process_sort 4041 11.4 8.400 9.944 8.400 9.944
make_images_data 222 9.7 0.004 0.004 8.871 9.941
hybrid_alltoall_any 227 10.6 0.521 2.478 8.277 9.291
mp_sum_l 807 5.4 5.464 8.521 5.464 8.521
multiply_cannon_metrocomm4 1221 9.7 0.006 0.008 3.234 7.801
mp_irecv_dv 3311 11.0 3.213 7.743 3.213 7.743
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.240 6.786
calculate_norms 2376 9.8 6.014 6.730 6.014 6.730
multiply_cannon_sync_h2d 1332 9.7 4.739 5.775 4.739 5.775
apply_matrix_preconditioner 6 5.3 0.001 0.001 5.053 5.235
arnoldi_extremal 4 6.8 0.000 0.000 4.761 4.785
arnoldi_normal_ev 4 7.8 0.001 0.004 4.761 4.785
build_subspace 16 8.4 0.014 0.021 4.498 4.501
ls_scf_post 1 4.0 0.000 0.000 3.531 3.536
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.200 3.399
ls_scf_store_result 1 5.0 0.000 0.000 3.235 3.357
dbcsr_matrix_vector_mult_local 304 10.0 2.737 3.220 2.739 3.222
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 1.209 2.722
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.619 2.706
mp_allgather_i34 111 8.7 0.977 2.475 0.977 2.475
make_images_pack 222 9.7 2.034 2.417 2.036 2.419
dbcsr_sort_data 436 11.2 1.845 2.081 1.845 2.081
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.888 1.890
rebuild_ks_matrix 3 7.3 0.000 0.000 1.875 1.878
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.875 1.878
dbcsr_data_new 4174 10.1 1.606 1.836 1.606 1.836
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=91.483000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1781.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 2.699071E+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 0.049 0.069 93.934 93.935
qs_energies 1 2.0 0.000 0.000 93.223 93.231
ls_scf 1 3.0 0.000 0.001 91.812 91.821
dbcsr_multiply_generic 111 6.7 0.016 0.016 76.517 76.774
ls_scf_main 1 4.0 0.000 0.003 57.729 57.734
multiply_cannon 111 7.7 0.040 0.090 52.721 57.144
multiply_cannon_loop 111 8.7 0.100 0.117 49.180 53.010
density_matrix_trs4 2 5.0 0.004 0.046 51.751 51.857
mp_waitall_1 7281 11.0 23.908 33.507 23.908 33.507
ls_scf_init_scf 1 4.0 0.001 0.005 30.519 30.521
ls_scf_init_matrix_S 1 5.0 0.000 0.003 29.337 29.406
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.951 26.962
multiply_cannon_multrec 888 9.7 12.677 15.209 21.189 24.280
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 10.813 22.556
make_m2s 222 7.7 0.006 0.007 17.225 18.428
make_images 222 8.7 1.971 2.289 17.186 18.390
hybrid_alltoall_any 227 10.6 0.622 2.892 9.528 10.981
make_images_data 222 9.7 0.003 0.004 9.949 10.856
dbcsr_mm_accdrv_process 3754 10.4 0.248 0.443 8.056 9.323
dbcsr_mm_accdrv_process_sort 3754 11.4 7.678 8.880 7.678 8.880
mp_sum_l 807 5.4 5.025 8.495 5.025 8.495
multiply_cannon_sync_h2d 888 9.7 6.027 7.669 6.027 7.669
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.445 6.874
mp_irecv_dv 2335 11.1 2.429 6.819 2.429 6.819
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 3.765 6.669
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.817 6.645
apply_matrix_preconditioner 6 5.3 0.001 0.001 4.975 5.148
arnoldi_extremal 4 6.8 0.000 0.000 5.086 5.101
arnoldi_normal_ev 4 7.8 0.001 0.005 5.086 5.101
build_subspace 16 8.4 0.014 0.020 4.790 4.795
calculate_norms 1584 9.8 4.302 4.766 4.302 4.766
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.448 3.790
mp_allgather_i34 111 8.7 1.382 3.776 1.382 3.776
dbcsr_matrix_vector_mult_local 304 10.0 3.016 3.591 3.018 3.593
ls_scf_post 1 4.0 0.001 0.006 3.564 3.575
ls_scf_store_result 1 5.0 0.000 0.000 3.291 3.386
ls_scf_dm_to_ks 2 5.0 0.001 0.033 2.826 2.918
dbcsr_sort_data 325 11.1 1.917 2.174 1.917 2.174
make_images_pack 222 9.7 1.820 2.118 1.823 2.121
dbcsr_data_release 9322 10.9 1.297 2.023 1.297 2.023
make_images_sizes 222 9.7 0.000 0.000 0.921 1.981
mp_alltoall_i44 222 10.7 0.920 1.981 0.920 1.981
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.911 1.913
dbcsr_finalize 304 7.8 0.026 0.032 1.617 1.896
rebuild_ks_matrix 3 7.3 0.000 0.000 1.893 1.896
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.003 1.893 1.896
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=93.935000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2177.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 3.343573E+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 0.034 0.128 97.089 97.141
qs_energies 1 2.0 0.000 0.000 96.463 96.469
ls_scf 1 3.0 0.000 0.000 94.808 94.815
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.510 78.875
ls_scf_main 1 4.0 0.000 0.000 58.856 58.857
multiply_cannon 111 7.7 0.051 0.108 51.605 56.220
density_matrix_trs4 2 5.0 0.002 0.003 52.704 52.847
multiply_cannon_loop 111 8.7 0.115 0.126 46.517 49.932
ls_scf_init_scf 1 4.0 0.000 0.000 32.730 32.731
ls_scf_init_matrix_S 1 5.0 0.000 0.001 31.517 31.598
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.063 29.078
mp_waitall_1 6369 11.0 22.756 28.908 22.756 28.908
multiply_cannon_multrec 1332 9.7 14.169 17.548 22.013 24.616
make_m2s 222 7.7 0.007 0.008 21.178 22.655
make_images 222 8.7 3.144 3.611 21.127 22.607
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.206 17.237
make_images_data 222 9.7 0.004 0.004 11.804 13.562
hybrid_alltoall_any 227 10.6 0.797 3.743 11.108 12.879
dbcsr_mm_accdrv_process 3641 10.4 0.212 0.402 7.481 9.004
dbcsr_mm_accdrv_process_sort 3641 11.4 7.108 8.591 7.108 8.591
mp_sum_l 807 5.4 4.177 7.593 4.177 7.593
multiply_cannon_sync_h2d 1332 9.7 5.514 6.030 5.514 6.030
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.081 5.958
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.168 5.886
mp_irecv_dv 3229 10.9 2.058 5.878 2.058 5.878
arnoldi_extremal 4 6.8 0.000 0.000 5.247 5.260
arnoldi_normal_ev 4 7.8 0.001 0.005 5.247 5.260
build_subspace 16 8.4 0.014 0.021 4.904 4.911
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.574 4.783
apply_matrix_preconditioner 6 5.3 0.001 0.001 4.575 4.719
calculate_norms 2376 9.8 4.190 4.587 4.190 4.587
mp_allgather_i34 111 8.7 2.211 4.581 2.211 4.581
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.591 3.880
dbcsr_matrix_vector_mult_local 304 10.0 3.182 3.692 3.184 3.693
dbcsr_sort_data 658 11.4 3.021 3.394 3.021 3.394
ls_scf_post 1 4.0 0.000 0.000 3.222 3.228
dbcsr_special_finalize 555 9.7 0.006 0.007 2.760 3.215
dbcsr_merge_single_wm 555 10.7 0.529 0.659 2.752 3.207
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.000 3.067
ls_scf_store_result 1 5.0 0.000 0.000 2.955 3.027
dbcsr_data_release 10477 10.7 1.587 2.406 1.587 2.406
dbcsr_finalize 304 7.8 0.049 0.061 1.799 1.979
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.948 1.950
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.141000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2742.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 4.624978E+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 0.039 0.053 99.250 99.253
qs_energies 1 2.0 0.000 0.000 98.303 98.308
ls_scf 1 3.0 0.000 0.000 96.377 96.388
dbcsr_multiply_generic 111 6.7 0.017 0.018 77.911 78.133
ls_scf_main 1 4.0 0.000 0.000 62.009 62.010
multiply_cannon 111 7.7 0.089 0.145 55.261 60.817
density_matrix_trs4 2 5.0 0.002 0.003 54.941 55.030
multiply_cannon_loop 111 8.7 0.070 0.076 50.879 52.486
mp_waitall_1 5436 11.0 26.691 32.322 26.691 32.322
ls_scf_init_scf 1 4.0 0.000 0.000 30.735 30.741
ls_scf_init_matrix_S 1 5.0 0.000 0.003 29.446 29.511
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.269 27.276
multiply_cannon_multrec 444 9.7 13.950 16.061 20.976 23.231
make_m2s 222 7.7 0.004 0.005 17.811 20.185
make_images 222 8.7 3.732 4.433 17.750 20.125
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.096 16.922
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.378 14.244
make_images_data 222 9.7 0.003 0.004 10.018 12.354
hybrid_alltoall_any 227 10.6 0.788 3.756 9.860 12.250
dbcsr_mm_accdrv_process 3003 10.4 0.187 0.341 6.732 7.879
multiply_cannon_sync_h2d 444 9.7 6.560 7.551 6.560 7.551
dbcsr_mm_accdrv_process_sort 3003 11.4 6.417 7.533 6.417 7.533
mp_allgather_i34 111 8.7 2.623 7.014 2.623 7.014
arnoldi_extremal 4 6.8 0.000 0.000 5.808 5.817
arnoldi_normal_ev 4 7.8 0.001 0.004 5.808 5.817
build_subspace 16 8.4 0.015 0.020 5.416 5.424
apply_matrix_preconditioner 6 5.3 0.001 0.001 4.530 4.689
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.686 4.661
mp_irecv_dv 1241 11.2 1.666 4.628 1.666 4.628
mp_sum_l 807 5.4 2.891 4.545 2.891 4.545
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.160 4.366
dbcsr_matrix_vector_mult_local 304 10.0 3.687 4.163 3.689 4.165
calculate_norms 792 9.8 3.545 3.650 3.545 3.650
ls_scf_post 1 4.0 0.000 0.000 3.633 3.637
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.528 3.607
make_images_sizes 222 9.7 0.000 0.000 1.139 3.498
mp_alltoall_i44 222 10.7 1.139 3.498 1.139 3.498
ls_scf_store_result 1 5.0 0.000 0.000 3.410 3.460
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.936 3.248
dbcsr_finalize 304 7.8 0.062 0.078 2.202 2.295
dbcsr_merge_all 275 8.9 0.475 0.527 2.051 2.129
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.079 2.082
rebuild_ks_matrix 3 7.3 0.000 0.000 2.046 2.048
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.046 2.048
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=99.253000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3686.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_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 8.771764E+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 0.075 0.097 107.900 107.901
qs_energies 1 2.0 0.000 0.000 106.528 106.542
ls_scf 1 3.0 0.000 0.000 103.608 103.623
dbcsr_multiply_generic 111 6.7 0.023 0.028 77.319 77.458
ls_scf_main 1 4.0 0.000 0.000 65.870 65.872
density_matrix_trs4 2 5.0 0.002 0.003 56.716 56.765
multiply_cannon 111 7.7 0.164 0.285 49.900 51.877
multiply_cannon_loop 111 8.7 0.068 0.070 46.417 47.279
ls_scf_init_scf 1 4.0 0.000 0.000 34.052 34.053
ls_scf_init_matrix_S 1 5.0 0.000 0.001 32.440 32.465
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.632 29.643
mp_waitall_1 4527 11.1 22.109 25.937 22.109 25.937
make_m2s 222 7.7 0.005 0.005 23.831 24.825
make_images 222 8.7 4.620 4.999 23.724 24.716
multiply_cannon_multrec 444 9.7 17.873 18.616 22.474 22.995
hybrid_alltoall_any 227 10.6 1.663 3.626 12.843 15.470
make_images_data 222 9.7 0.003 0.003 13.037 15.453
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.493 10.899
multiply_cannon_sync_h2d 444 9.7 8.849 8.897 8.849 8.897
arnoldi_extremal 4 6.8 0.000 0.000 7.309 7.324
arnoldi_normal_ev 4 7.8 0.003 0.009 7.309 7.324
build_subspace 16 8.4 0.026 0.036 6.754 6.767
dbcsr_matrix_vector_mult 304 9.0 0.017 0.033 5.435 5.581
dbcsr_matrix_vector_mult_local 304 10.0 5.019 5.327 5.021 5.329
apply_matrix_preconditioner 6 5.3 0.001 0.001 5.003 5.241
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.128 5.218
dbcsr_mm_accdrv_process 1814 10.4 0.231 0.316 4.424 4.554
dbcsr_mm_accdrv_process_sort 1814 11.4 4.124 4.257 4.124 4.257
ls_scf_post 1 4.0 0.000 0.000 3.686 3.699
mp_allgather_i34 111 8.7 1.113 3.532 1.113 3.532
make_images_sizes 222 9.7 0.000 0.000 1.427 3.475
mp_alltoall_i44 222 10.7 1.427 3.475 1.427 3.475
ls_scf_store_result 1 5.0 0.000 0.000 3.433 3.448
calculate_norms 792 9.8 3.238 3.281 3.238 3.281
dbcsr_finalize 304 7.8 0.082 0.089 3.092 3.174
dbcsr_merge_all 275 8.9 0.891 0.922 2.878 2.955
qs_energies_init_hamiltonians 1 3.0 0.001 0.002 2.889 2.889
dbcsr_complete_redistribute 5 7.6 1.432 1.469 2.757 2.876
dbcsr_data_release 12724 10.6 2.324 2.861 2.324 2.861
matrix_ls_to_qs 2 6.0 0.000 0.000 2.409 2.537
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.518 2.519
dbcsr_sort_data 325 11.1 2.450 2.504 2.450 2.504
rebuild_ks_matrix 3 7.3 0.000 0.000 2.452 2.454
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.452 2.454
dbcsr_new_transposed 4 7.5 0.243 0.251 2.257 2.270
dbcsr_frobenius_norm 74 6.6 2.056 2.133 2.194 2.242
dbcsr_add_d 103 6.2 0.000 0.000 2.132 2.212
dbcsr_add_anytype 103 7.2 0.859 0.895 2.132 2.212
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=107.901000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6867.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d4020ab16046caa29083b8d6b10b5b5cd0a3b986_performance_tests/27/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 7009386627072 0.0% 0.0% 100.0%
flops 9 x 9 x 32 7335108845568 0.0% 0.0% 100.0%
flops 9 x 22 x 32 9866241589248 0.0% 0.0% 100.0%
flops 22 x 9 x 32 9884108906496 0.0% 0.0% 100.0%
flops 22 x 22 x 32 13354440523776 0.0% 0.0% 100.0%
flops 32 x 32 x 9 20607185977344 0.0% 0.0% 100.0%
flops 32 x 32 x 22 25186560638976 0.0% 0.0% 100.0%
flops 9 x 32 x 32 28458319085568 0.0% 0.0% 100.0%
flops 22 x 32 x 32 34782389993472 0.0% 0.0% 100.0%
flops 9 x 32 x 9 42881542373376 0.0% 0.0% 100.0%
flops 22 x 32 x 9 55680402235392 0.0% 0.0% 100.0%
flops 9 x 32 x 22 55680402235392 0.0% 0.0% 100.0%
flops 22 x 32 x 22 72328573419520 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 383.054662E+12 0.0% 0.0% 100.0%
flops max/rank 733.641090E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 26899403712 0.0% 0.0% 100.0%
number of processed stacks 118860288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 226.3
marketing flops 780.439111E+12
-------------------------------------------------------------------------------
# multiplications 1445
max memory usage/rank 590.467072E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 76559.
MP_Allreduce 13152 2094.
MP_Sync 1064
MP_Alltoall 2588 4037197.
MP_SendRecv 168740 11136.
MP_ISendRecv 92040 11136.
MP_Wait 102830
MP_comm_split 40
MP_ISend 26090 85106.
MP_IRecv 37890 59644.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.038 0.099 255.238 255.242
qs_mol_dyn_low 1 2.0 0.003 0.030 254.290 254.313
qs_forces 5 3.8 0.005 0.006 254.186 254.191
qs_energies 5 4.8 0.002 0.066 250.987 251.010
scf_env_do_scf 5 5.8 0.000 0.004 235.856 235.861
scf_env_do_scf_inner_loop 105 6.6 0.002 0.006 205.735 205.737
qs_scf_new_mos 105 7.6 0.000 0.001 159.796 160.023
qs_scf_loop_do_ot 105 8.6 0.001 0.002 159.796 160.022
ot_scf_mini 105 9.6 0.003 0.005 149.917 150.067
dbcsr_multiply_generic 1445 12.2 0.125 0.133 134.543 135.271
multiply_cannon 1445 13.2 0.276 0.292 114.450 117.327
multiply_cannon_loop 1445 14.2 2.358 2.488 112.310 114.282
velocity_verlet 4 3.0 0.000 0.001 112.241 112.242
ot_mini 105 10.6 0.001 0.001 61.009 61.142
qs_ot_get_p 112 10.4 0.001 0.001 58.862 59.198
qs_ot_p2m_diag 40 11.0 0.020 0.030 47.118 47.216
mp_waitall_1 488190 16.1 36.539 44.615 36.539 44.615
multiply_cannon_multrec 69360 15.2 31.878 36.368 39.615 44.612
cp_dbcsr_syevd 40 12.0 0.002 0.002 43.880 43.882
qs_ot_get_derivative 55 11.6 0.001 0.001 39.293 39.432
cp_fm_syevd 40 13.0 0.000 0.001 38.352 38.500
multiply_cannon_metrocomm3 69360 15.2 0.192 0.203 25.635 35.425
multiply_cannon_sync_h2d 69360 15.2 30.429 34.931 30.429 34.931
cp_fm_redistribute_end 40 14.0 16.659 33.226 16.667 33.229
cp_fm_syevd_base 40 14.0 16.552 33.128 16.552 33.128
rebuild_ks_matrix 110 8.4 0.000 0.001 31.134 31.346
qs_ks_build_kohn_sham_matrix 110 9.4 0.012 0.014 31.134 31.346
init_scf_loop 7 6.6 0.000 0.007 30.074 30.076
qs_ks_update_qs_env 112 7.6 0.001 0.001 28.567 28.759
prepare_preconditioner 7 7.6 0.000 0.000 24.830 24.870
make_preconditioner 7 8.6 0.000 0.001 24.830 24.870
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 22.997 23.274
apply_single 62 13.6 0.000 0.000 22.997 23.274
qs_rho_update_rho_low 110 7.6 0.001 0.001 21.190 21.510
calculate_rho_elec 110 8.6 0.030 0.033 21.189 21.509
ot_new_cg_direction 55 11.6 0.001 0.001 20.981 20.981
rs_pw_transfer 690 11.5 0.011 0.013 16.447 17.779
make_full_inverse_cholesky 7 9.6 0.000 0.000 17.156 17.215
density_rs2pw 110 9.6 0.006 0.007 14.984 16.294
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.847 14.988
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 14.032 14.153
mp_sum_l 4684 12.4 12.375 13.494 12.375 13.494
pw_transfer 1645 12.4 0.084 0.107 12.453 12.677
fft_wrap_pw1pw2 1425 13.5 0.013 0.017 12.309 12.534
init_scf_run 5 5.8 0.000 0.002 12.395 12.396
scf_env_initial_rho_setup 5 6.8 0.002 0.003 12.395 12.396
cp_fm_cholesky_invert 7 10.6 11.967 11.987 11.967 11.987
calculate_dm_sparse 110 9.5 0.000 0.001 11.540 11.692
qs_vxc_create 110 10.4 0.002 0.009 10.589 10.644
fft_wrap_pw1pw2_240 915 15.0 1.196 1.288 10.370 10.604
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 10.473 10.546
sum_up_and_integrate 60 10.3 0.028 0.031 10.503 10.512
integrate_v_rspace 60 11.3 0.002 0.002 10.475 10.485
check_diag 80 13.5 8.585 8.854 10.080 10.230
fft3d_pb 915 16.0 2.391 2.677 8.417 8.712
dbcsr_mm_accdrv_process 154766 15.8 3.934 4.148 7.607 8.412
multiply_cannon_metrocomm1 69360 15.2 0.089 0.096 4.773 8.018
make_m2s 2890 13.2 0.076 0.084 7.251 7.800
make_images 2890 14.2 0.242 0.260 7.149 7.695
xc_rho_set_and_dset_create 110 12.4 0.076 0.097 7.349 7.603
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.002 7.569 7.584
make_full_single_inverse 7 9.6 0.001 0.001 7.371 7.410
acc_transpose_blocks 69360 15.2 0.350 0.375 6.685 7.298
calculate_first_density_matrix 1 7.0 0.000 0.003 7.237 7.252
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 7.124 7.186
xc_vxc_pw_create 60 11.3 0.038 0.049 7.101 7.151
potential_pw2rs 60 12.3 0.003 0.003 6.495 6.554
xc_pw_derive 510 13.4 0.005 0.007 6.317 6.383
rs_pw_transfer_RS2PW_30 110 11.6 1.582 1.682 5.403 5.942
mp_alltoall_z22v 2340 17.7 5.504 5.941 5.504 5.941
mp_waitany 7680 13.5 4.502 5.775 4.502 5.775
acc_transpose_blocks_kernels 69360 16.2 0.846 0.898 4.845 5.345
multiply_cannon_metrocomm4 67915 15.2 0.182 0.195 2.082 5.203
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=255.242000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=560.200000, yerr=3.429286
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: d4020ab16046caa29083b8d6b10b5b5cd0a3b986
Summary: empty
Status: OK