=== 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: e64daf41ce02fb1dd7f72831acc068f4c092e916
################# 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 2021.11.002, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 5.2.3, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.0,
# SIRIUS 7.3.2, 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 (15.09.2022)
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 --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=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2021.11.002
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 5.2.3
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.0
#USE_QUIP := b4336484fb65b0e73211a8f920ae4361c7c353fd
USE_SIRIUS := 7.3.2
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MPI_VERSION=3
DFLAGS += -D__MAX_CONTR=$(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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/01
job id: 41993869
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/02
job id: 41993874
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/03
job id: 41993875
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/04
job id: 41993876
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/05
job id: 41993877
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/06
job id: 41993878
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/07
job id: 41993879
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/08
job id: 41993880
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/09
job id: 41993881
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/10
job id: 41993882
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/11
job id: 41993883
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/12
job id: 41993884
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/13
job id: 41993885
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/14
job id: 41993886
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/15
job id: 41993887
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/16
job id: 41993888
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/17
job id: 41993889
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/18
job id: 41993890
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/19
job id: 41993891
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/20
job id: 41993893
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/21
job id: 41993894
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/22
job id: 41993895
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/23
job id: 41993896
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/24
job id: 41993897
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/25
job id: 41993898
--- 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/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/26
job id: 41993899
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 0.035 133.254 133.255
farming_run 1 2.0 132.721 132.727 133.226 133.229
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.456517E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 228 1113141.
MP_Allreduce 489 2263609.
MP_Sync 27
MP_Alltoall 38 9316958.
MP_SendRecv 30 829726.
MP_ISendRecv 135 235435.
MP_Wait 281
MP_comm_split 8
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.024 116.008 116.008
qs_energies 1 2.0 0.000 0.000 115.778 115.795
mp2_main 1 3.0 0.000 0.000 113.846 113.863
mp2_gpw_main 1 4.0 0.020 0.026 113.004 113.022
mp2_ri_gpw_compute_in 1 5.0 0.173 0.174 94.120 94.511
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 55.543 55.935
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.169 41.867 47.210
get_2c_integrals 1 6.0 0.000 0.002 37.558 38.404
integrate_v_rspace 273 8.0 0.439 0.451 25.181 30.289
pw_transfer 6555 10.6 0.376 0.393 27.455 27.962
grid_integrate_task_list 273 9.0 20.979 26.579 20.979 26.579
fft_wrap_pw1pw2 5465 11.4 0.044 0.047 26.115 26.482
fft_wrap_pw1pw2_100 2178 12.4 1.203 1.460 23.676 24.044
compute_2c_integrals 1 7.0 0.003 0.003 19.747 19.747
compute_2c_integrals_loop_lm 1 8.0 0.003 0.004 18.965 19.473
mp2_eri_2c_integrate_gpw 1 9.0 2.401 2.446 18.962 19.471
rpa_ri_compute_en 1 5.0 0.007 0.008 18.775 19.027
cp_fm_cholesky_decompose 12 8.2 17.773 18.617 17.773 18.617
cholesky_decomp 1 7.0 0.000 0.000 16.653 17.508
fft3d_s 5443 13.4 16.198 16.593 16.221 16.614
ao_to_mo_and_store_B_mult_1 272 7.0 10.844 15.554 10.844 15.554
calculate_wavefunction 272 8.0 5.447 5.597 12.594 13.171
rpa_num_int 1 6.0 0.002 0.033 10.570 10.590
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.538 10.559
calc_mat_Q 8 8.0 0.000 0.000 9.369 9.465
contract_S_to_Q 8 9.0 0.000 0.000 8.793 8.885
calc_potential_gpw 544 9.5 0.005 0.006 8.217 8.702
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.002 0.002 8.237 8.568
parallel_gemm_fm 14 9.1 0.000 0.000 8.380 8.454
parallel_gemm_fm_cosma 14 10.1 8.380 8.454 8.380 8.454
potential_pw2rs 545 10.0 0.107 0.109 7.771 8.362
collocate_single_gaussian 272 10.0 0.040 0.042 7.506 7.784
create_integ_mat 1 6.0 0.006 0.008 7.711 7.711
array2fm 1 7.0 0.000 0.000 6.679 7.161
pw_scatter_s 2720 13.7 4.425 4.663 4.425 4.663
pw_gather_s 2722 13.2 3.919 4.427 3.919 4.427
array2fm_buffer_send 1 8.0 2.892 3.145 2.892 3.145
pw_poisson_solve 545 10.5 1.145 1.259 2.190 2.360
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=113.021735, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2727.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 344 10.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.029 0.038 396.177 396.178
farming_run 1 2.0 395.571 395.577 396.142 396.146
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.221706E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 703 408373.
MP_Allreduce 1825 23678.
MP_Sync 38
MP_Alltoall 77
MP_SendRecv 2171 2843495.
MP_ISendRecv 1739 144022.
MP_Wait 2051
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.036 210.425 210.430
qs_energies 1 2.0 0.000 0.000 210.208 210.213
scf_env_do_scf 1 3.0 0.000 0.000 107.539 107.540
qs_ks_update_qs_env 5 5.0 0.000 0.000 106.677 106.687
rebuild_ks_matrix 4 6.0 0.000 0.000 106.676 106.686
qs_ks_build_kohn_sham_matrix 4 7.0 0.054 0.064 106.676 106.686
hfx_ks_matrix 4 8.0 0.001 0.001 106.302 106.305
integrate_four_center 4 9.0 0.143 0.461 106.301 106.305
mp2_main 1 3.0 0.000 0.000 102.380 102.386
mp2_gpw_main 1 4.0 0.033 0.045 101.541 101.550
integrate_four_center_main 4 10.0 0.107 0.503 96.820 100.897
integrate_four_center_bin 267 11.0 96.713 100.887 96.713 100.887
init_scf_loop 1 4.0 0.000 0.000 93.034 93.034
mp2_ri_gpw_compute_in 1 5.0 0.075 0.109 74.905 75.993
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 54.489 55.583
mp2_eri_3c_integrate_gpw 91 7.0 0.144 0.159 42.127 47.391
integrate_v_rspace 95 8.0 0.397 0.577 28.490 33.563
pw_transfer 2240 10.6 0.145 0.165 29.915 30.375
fft_wrap_pw1pw2 1868 11.4 0.018 0.021 28.914 29.396
grid_integrate_task_list 95 9.0 23.780 29.057 23.780 29.057
mp2_ri_gpw_compute_en 1 5.0 0.055 0.065 26.490 28.129
fft_wrap_pw1pw2_100 730 12.4 1.288 1.436 26.638 27.145
ao_to_mo_and_store_B_mult_1 91 7.0 10.678 26.088 10.678 26.088
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.831 1.883 24.765 24.774
get_2c_integrals 1 6.0 0.000 0.000 20.307 20.345
compute_2c_integrals 1 7.0 0.004 0.010 19.298 19.304
compute_2c_integrals_loop_lm 1 8.0 0.003 0.008 18.932 19.166
mp2_eri_2c_integrate_gpw 1 9.0 1.740 1.855 18.930 19.165
fft3d_s 1823 13.4 18.401 18.792 18.414 18.806
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.504 14.504
calculate_wavefunction 91 8.0 2.025 2.059 9.749 10.000
mp2_ri_gpw_compute_en_expansio 172 7.0 0.556 0.585 8.753 9.431
potential_pw2rs 186 10.0 0.034 0.035 8.625 9.251
local_gemm 172 8.0 8.196 8.859 8.196 8.859
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.279 8.613
calc_potential_gpw 182 9.5 0.002 0.003 7.963 8.168
mp2_ri_gpw_compute_en_comm 22 7.0 0.502 0.523 7.795 8.147
collocate_single_gaussian 91 10.0 0.017 0.024 7.896 8.138
mp_sync 38 10.4 4.123 6.954 4.123 6.954
mp2_ri_gpw_compute_en_ener 172 7.0 6.341 6.438 6.341 6.438
mp_sendrecv_dm3 2068 8.0 5.819 6.197 5.819 6.197
pw_gather_s 912 13.2 4.906 5.497 4.906 5.497
pw_scatter_s 910 13.7 3.950 4.384 3.950 4.384
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=101.540751, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1504.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 450.244608E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62385.
MP_Allreduce 10249 271.
MP_Sync 580
MP_Alltoall 2083 279165.
MP_ISendRecv 45220 5520.
MP_Wait 60486
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 0.011 0.028 51.544 51.545
qs_mol_dyn_low 1 2.0 0.003 0.003 51.300 51.308
qs_forces 11 3.9 0.002 0.002 51.219 51.233
qs_energies 11 4.9 0.001 0.002 49.791 49.813
scf_env_do_scf 11 5.9 0.000 0.001 43.851 43.852
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 41.783 41.783
dbcsr_multiply_generic 2286 12.5 0.092 0.097 33.297 33.776
qs_scf_new_mos 108 7.5 0.000 0.001 31.908 32.209
qs_scf_loop_do_ot 108 8.5 0.000 0.001 31.908 32.209
ot_scf_mini 108 9.5 0.002 0.002 30.263 30.427
multiply_cannon 2286 13.5 0.186 0.195 25.771 27.430
multiply_cannon_loop 2286 14.5 1.496 1.573 25.072 26.718
velocity_verlet 10 3.0 0.001 0.001 25.633 25.634
ot_mini 108 10.5 0.001 0.001 19.358 19.620
qs_ot_get_derivative 108 11.5 0.001 0.001 16.434 16.617
mp_waitall_1 267858 16.1 8.671 15.061 8.671 15.061
multiply_cannon_metrocomm3 54864 15.5 0.066 0.072 5.992 13.240
multiply_cannon_multrec 54864 15.5 4.200 6.527 7.472 10.877
rebuild_ks_matrix 119 8.3 0.000 0.000 7.784 7.944
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 7.783 7.944
multiply_cannon_sync_h2d 54864 15.5 5.844 7.740 5.844 7.740
mp_sum_l 7207 12.9 5.599 7.475 5.599 7.475
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.880 7.024
qs_ot_get_p 119 10.4 0.001 0.001 6.288 6.625
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.457 5.898
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.622 5.746
init_scf_run 11 5.9 0.000 0.001 4.704 4.704
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.703 4.704
sum_up_and_integrate 119 10.3 0.012 0.015 4.435 4.441
dbcsr_mm_accdrv_process 76910 16.1 1.185 1.823 3.195 4.430
integrate_v_rspace 119 11.3 0.002 0.002 4.423 4.430
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.010 4.108
calculate_rho_elec 119 8.7 0.011 0.016 4.009 4.107
qs_ot_p2m_diag 50 11.0 0.004 0.006 3.446 3.507
multiply_cannon_metrocomm1 54864 15.5 0.052 0.059 1.828 3.331
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.871 3.080
apply_single 119 13.6 0.000 0.000 2.871 3.079
calculate_dm_sparse 119 9.5 0.000 0.000 2.884 3.075
rs_pw_transfer 974 11.9 0.012 0.013 2.736 2.804
ot_diis_step 108 11.5 0.006 0.006 2.663 2.663
cp_dbcsr_syevd 50 12.0 0.002 0.003 2.622 2.622
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.442 2.531
jit_kernel_multiply 13 15.8 1.948 2.525 1.948 2.525
calculate_first_density_matrix 1 7.0 0.000 0.000 2.480 2.484
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.407 2.407
cp_fm_redistribute_end 50 14.0 2.188 2.387 2.192 2.388
cp_fm_diag_elpa_base 50 14.0 0.194 2.343 0.195 2.354
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.229 2.231
density_rs2pw 119 9.7 0.004 0.004 2.129 2.199
acc_transpose_blocks 54864 15.5 0.225 0.243 1.740 2.167
wfi_extrapolate 11 7.9 0.001 0.001 2.164 2.164
grid_integrate_task_list 119 12.3 2.019 2.120 2.019 2.120
init_scf_loop 11 6.9 0.000 0.000 2.049 2.050
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.937 1.981
mp_sum_d 4129 12.0 1.287 1.900 1.287 1.900
potential_pw2rs 119 12.3 0.004 0.004 1.809 1.819
pw_transfer 1439 11.6 0.051 0.056 1.596 1.665
make_m2s 4572 13.5 0.054 0.056 1.616 1.665
fft_wrap_pw1pw2 1201 12.6 0.006 0.007 1.521 1.592
make_images 4572 14.5 0.132 0.137 1.534 1.582
mp_alltoall_d11v 2130 13.8 1.315 1.493 1.315 1.493
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.396 1.414
mp_waitany 12084 13.8 1.245 1.399 1.245 1.399
fft3d_ps 1201 14.6 0.359 0.467 1.299 1.364
grid_collocate_task_list 119 9.7 1.283 1.359 1.283 1.359
fft_wrap_pw1pw2_140 487 13.2 0.080 0.093 1.164 1.235
dbcsr_dot_sd 1205 11.9 0.047 0.057 0.699 1.118
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=51.545000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=429.181818, yerr=1.113404
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 486.350848E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62664.
MP_Allreduce 10226 305.
MP_Sync 104
MP_Alltoall 2060 1568197.
MP_ISendRecv 33558 37093.
MP_Wait 40318
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 0.011 0.026 38.192 38.193
qs_mol_dyn_low 1 2.0 0.003 0.003 38.000 38.008
qs_forces 11 3.9 0.002 0.003 37.939 37.940
qs_energies 11 4.9 0.001 0.001 36.253 36.257
scf_env_do_scf 11 5.9 0.000 0.001 31.196 31.196
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 28.704 28.705
dbcsr_multiply_generic 2286 12.5 0.100 0.104 21.298 21.643
qs_scf_new_mos 108 7.5 0.001 0.001 19.710 19.959
qs_scf_loop_do_ot 108 8.5 0.001 0.001 19.710 19.958
ot_scf_mini 108 9.5 0.002 0.003 18.816 18.992
velocity_verlet 10 3.0 0.001 0.001 18.192 18.193
multiply_cannon 2286 13.5 0.208 0.214 16.296 17.842
multiply_cannon_loop 2286 14.5 0.902 0.981 15.167 16.516
ot_mini 108 10.5 0.001 0.001 11.752 11.990
mp_waitall_1 217478 16.2 6.052 11.177 6.052 11.177
multiply_cannon_metrocomm3 27432 15.5 0.067 0.070 4.132 9.489
qs_ot_get_derivative 108 11.5 0.001 0.001 9.298 9.475
multiply_cannon_multrec 27432 15.5 1.970 4.380 5.861 8.780
rebuild_ks_matrix 119 8.3 0.000 0.000 7.160 7.292
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 7.159 7.292
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.325 6.446
dbcsr_mm_accdrv_process 47894 16.0 3.008 5.187 3.822 5.650
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.589 4.443
qs_ot_get_p 119 10.4 0.001 0.001 4.158 4.393
sum_up_and_integrate 119 10.3 0.024 0.027 4.196 4.203
integrate_v_rspace 119 11.3 0.002 0.002 4.172 4.179
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.046 4.116
apply_single 119 13.6 0.000 0.000 3.046 4.116
mp_sum_l 7207 12.9 2.035 3.988 2.035 3.988
init_scf_run 11 5.9 0.000 0.001 3.862 3.862
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.861 3.862
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.621 3.652
calculate_rho_elec 119 8.7 0.021 0.024 3.620 3.651
multiply_cannon_sync_h2d 27432 15.5 2.201 2.829 2.201 2.829
make_m2s 4572 13.5 0.052 0.055 2.515 2.737
qs_ot_p2m_diag 50 11.0 0.008 0.013 2.685 2.704
rs_pw_transfer 974 11.9 0.010 0.011 2.586 2.689
make_images 4572 14.5 0.201 0.240 2.427 2.647
init_scf_loop 11 6.9 0.000 0.000 2.468 2.468
ot_diis_step 108 11.5 0.010 0.011 2.405 2.406
calculate_first_density_matrix 1 7.0 0.000 0.000 2.344 2.346
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.244 2.244
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.132 2.225
calculate_dm_sparse 119 9.5 0.000 0.001 2.032 2.105
density_rs2pw 119 9.7 0.004 0.004 1.980 2.057
potential_pw2rs 119 12.3 0.006 0.006 1.936 1.945
grid_integrate_task_list 119 12.3 1.840 1.938 1.840 1.938
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.923 1.926
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.901 1.902
cp_fm_redistribute_end 50 14.0 1.568 1.877 1.572 1.879
cp_fm_diag_elpa_base 50 14.0 0.298 1.821 0.305 1.851
pw_transfer 1439 11.6 0.063 0.066 1.777 1.809
jit_kernel_multiply 9 16.0 0.762 1.789 0.762 1.789
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.711 1.750
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 1.688 1.720
prepare_preconditioner 11 7.9 0.000 0.000 1.568 1.594
make_preconditioner 11 8.9 0.000 0.000 1.568 1.594
make_images_data 4572 15.5 0.045 0.052 1.148 1.558
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.472 1.527
acc_transpose_blocks 27432 15.5 0.107 0.112 1.193 1.495
hybrid_alltoall_any 4725 16.4 0.051 0.112 0.993 1.489
wfi_extrapolate 11 7.9 0.001 0.001 1.466 1.466
fft3d_ps 1201 14.6 0.499 0.552 1.395 1.423
mp_alltoall_d11v 2130 13.8 1.204 1.348 1.204 1.348
fft_wrap_pw1pw2_140 487 13.2 0.076 0.083 1.311 1.344
grid_collocate_task_list 119 9.7 1.241 1.325 1.241 1.325
mp_allgather_i34 2286 14.5 0.572 1.314 0.572 1.314
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.243 1.250
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.182 1.229
mp_sum_d 4129 12.0 0.568 1.006 0.568 1.006
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.932 0.933
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.891 0.904
acc_transpose_blocks_kernels 27432 16.5 0.181 0.270 0.663 0.877
rs_pw_transfer_PW2RS_50 119 14.3 0.588 0.608 0.773 0.816
make_images_sizes 4572 15.5 0.005 0.005 0.549 0.799
mp_alltoall_i44 4572 16.5 0.544 0.794 0.544 0.794
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=38.193000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=462.000000, yerr=1.279204
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 520.876032E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62660.
MP_Allreduce 10225 303.
MP_Sync 104
MP_Alltoall 1821 1607811.
MP_ISendRecv 22134 57667.
MP_Wait 33054
MP_comm_split 50
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.042 0.108 32.601 32.603
qs_mol_dyn_low 1 2.0 0.003 0.003 32.178 32.187
qs_forces 11 3.9 0.002 0.002 32.117 32.118
qs_energies 11 4.9 0.010 0.050 30.539 30.542
scf_env_do_scf 11 5.9 0.005 0.012 25.660 25.660
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 22.993 22.998
dbcsr_multiply_generic 2286 12.5 0.093 0.098 16.467 16.553
velocity_verlet 10 3.0 0.001 0.001 15.421 15.423
qs_scf_new_mos 108 7.5 0.001 0.001 14.873 14.895
qs_scf_loop_do_ot 108 8.5 0.001 0.001 14.872 14.895
ot_scf_mini 108 9.5 0.003 0.005 14.148 14.168
multiply_cannon 2286 13.5 0.194 0.204 13.046 13.830
multiply_cannon_loop 2286 14.5 0.640 0.669 12.229 13.055
ot_mini 108 10.5 0.001 0.001 8.791 8.808
qs_ot_get_derivative 108 11.5 0.001 0.001 7.295 7.315
multiply_cannon_multrec 18288 15.5 1.944 3.007 6.752 7.135
rebuild_ks_matrix 119 8.3 0.000 0.000 6.437 6.452
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.015 6.437 6.451
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.699 5.713
dbcsr_mm_accdrv_process 38222 16.0 4.042 5.080 4.724 5.517
mp_waitall_1 169478 16.3 3.105 4.204 3.105 4.204
sum_up_and_integrate 119 10.3 0.029 0.030 4.032 4.037
integrate_v_rspace 119 11.3 0.002 0.003 4.003 4.011
init_scf_run 11 5.9 0.000 0.001 3.542 3.542
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.542 3.542
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.731 3.367
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.332 3.338
calculate_rho_elec 119 8.7 0.030 0.031 3.331 3.337
qs_ot_get_p 119 10.4 0.001 0.001 3.171 3.195
init_scf_loop 11 6.9 0.001 0.004 2.644 2.645
rs_pw_transfer 974 11.9 0.009 0.010 2.455 2.593
multiply_cannon_metrocomm3 18288 15.5 0.045 0.046 1.361 2.358
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.041 2.354
apply_single 119 13.6 0.000 0.000 2.041 2.354
calculate_first_density_matrix 1 7.0 0.001 0.003 2.312 2.313
make_m2s 4572 13.5 0.045 0.046 1.903 2.066
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.038 2.044
density_rs2pw 119 9.7 0.004 0.004 1.910 2.030
make_images 4572 14.5 0.190 0.203 1.818 1.978
grid_integrate_task_list 119 12.3 1.791 1.889 1.791 1.889
calculate_dm_sparse 119 9.5 0.000 0.000 1.808 1.823
potential_pw2rs 119 12.3 0.007 0.008 1.802 1.810
prepare_preconditioner 11 7.9 0.000 0.000 1.800 1.803
make_preconditioner 11 8.9 0.000 0.001 1.800 1.803
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.762 1.764
pw_transfer 1439 11.6 0.063 0.066 1.749 1.762
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.656 1.737
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.732 1.733
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.659 1.674
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.609 1.617
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.558 1.565
multiply_cannon_sync_h2d 18288 15.5 1.378 1.522 1.378 1.522
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.505 1.506
mp_sum_l 7207 12.9 1.169 1.500 1.169 1.500
jit_kernel_multiply 8 16.0 0.631 1.497 0.631 1.497
cp_fm_redistribute_end 50 14.0 1.116 1.481 1.118 1.482
ot_diis_step 108 11.5 0.011 0.011 1.473 1.473
cp_fm_diag_elpa_base 50 14.0 0.348 1.411 0.362 1.449
fft3d_ps 1201 14.6 0.506 0.522 1.346 1.360
fft_wrap_pw1pw2_140 487 13.2 0.086 0.090 1.299 1.312
grid_collocate_task_list 119 9.7 1.206 1.309 1.206 1.309
acc_transpose_blocks 18288 15.5 0.074 0.076 1.255 1.283
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.179 1.183
wfi_extrapolate 11 7.9 0.001 0.001 1.181 1.181
make_images_data 4572 15.5 0.045 0.048 0.853 1.040
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.027 1.028
hybrid_alltoall_any 4725 16.4 0.055 0.115 0.738 0.949
mp_alltoall_d11v 2130 13.8 0.759 0.923 0.759 0.923
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.894 0.919
acc_transpose_blocks_kernels 18288 16.5 0.210 0.220 0.817 0.830
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.809 0.811
cp_fm_cholesky_invert 11 10.9 0.767 0.771 0.767 0.771
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.665 0.743
mp_alltoall_z22v 1201 16.6 0.681 0.742 0.681 0.742
mp_waitany 9880 13.7 0.581 0.717 0.581 0.717
rs_pw_transfer_RS2PW_140 130 11.5 0.119 0.122 0.574 0.712
mp_allgather_i34 2286 14.5 0.281 0.657 0.281 0.657
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=32.603000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=494.636364, yerr=1.553455
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 554.487808E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62659.
MP_Allreduce 10224 344.
MP_Sync 104
MP_Alltoall 1582 2412273.
MP_ISendRecv 16422 74133.
MP_Wait 24482
MP_comm_split 50
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.027 36.463 36.463
qs_mol_dyn_low 1 2.0 0.003 0.003 36.255 36.264
qs_forces 11 3.9 0.002 0.002 36.197 36.198
qs_energies 11 4.9 0.001 0.002 34.440 34.445
scf_env_do_scf 11 5.9 0.000 0.001 29.145 29.146
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 25.490 25.490
dbcsr_multiply_generic 2286 12.5 0.100 0.102 18.927 19.094
velocity_verlet 10 3.0 0.001 0.001 18.239 18.241
qs_scf_new_mos 108 7.5 0.001 0.001 17.015 17.074
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.014 17.074
ot_scf_mini 108 9.5 0.002 0.003 16.053 16.106
multiply_cannon 2286 13.5 0.226 0.268 15.055 15.980
multiply_cannon_loop 2286 14.5 0.938 0.971 14.056 14.867
ot_mini 108 10.5 0.001 0.001 10.001 10.060
multiply_cannon_multrec 27432 15.5 2.346 3.013 8.878 9.432
qs_ot_get_derivative 108 11.5 0.001 0.001 8.168 8.217
dbcsr_mm_accdrv_process 47916 15.9 5.431 7.343 6.438 7.740
rebuild_ks_matrix 119 8.3 0.000 0.000 6.664 6.717
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.664 6.717
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.901 5.949
sum_up_and_integrate 119 10.3 0.034 0.037 3.926 3.935
integrate_v_rspace 119 11.3 0.002 0.003 3.892 3.902
init_scf_run 11 5.9 0.000 0.001 3.865 3.865
scf_env_initial_rho_setup 11 6.9 0.026 0.026 3.865 3.865
init_scf_loop 11 6.9 0.000 0.000 3.633 3.633
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.170 3.582
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.489 3.521
calculate_rho_elec 119 8.7 0.040 0.046 3.488 3.520
mp_waitall_1 145218 16.4 2.761 3.508 2.761 3.508
qs_ot_get_p 119 10.4 0.001 0.001 3.406 3.472
prepare_preconditioner 11 7.9 0.000 0.000 2.778 2.786
make_preconditioner 11 8.9 0.000 0.000 2.778 2.786
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.374 2.706
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.063 2.590
apply_single 119 13.6 0.000 0.000 2.063 2.590
make_m2s 4572 13.5 0.054 0.057 2.368 2.459
calculate_first_density_matrix 1 7.0 0.000 0.000 2.435 2.437
rs_pw_transfer 974 11.9 0.009 0.009 2.275 2.392
make_images 4572 14.5 0.273 0.334 2.260 2.351
jit_kernel_multiply 10 15.6 0.947 2.257 0.947 2.257
calculate_dm_sparse 119 9.5 0.000 0.000 2.084 2.146
density_rs2pw 119 9.7 0.003 0.004 1.963 2.082
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.051 2.061
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.025 2.051
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.927 1.929
grid_integrate_task_list 119 12.3 1.819 1.916 1.819 1.916
pw_transfer 1439 11.6 0.063 0.066 1.838 1.873
ot_diis_step 108 11.5 0.012 0.013 1.792 1.792
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.749 1.787
multiply_cannon_metrocomm3 27432 15.5 0.037 0.038 1.068 1.774
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.723 1.737
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.714 1.714
potential_pw2rs 119 12.3 0.008 0.009 1.697 1.704
mp_sum_l 7207 12.9 1.078 1.701 1.078 1.701
acc_transpose_blocks 27432 15.5 0.112 0.114 1.480 1.527
fft3d_ps 1201 14.6 0.534 0.586 1.429 1.457
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.450 1.452
fft_wrap_pw1pw2_140 487 13.2 0.084 0.091 1.403 1.444
cp_fm_redistribute_end 50 14.0 0.956 1.422 0.957 1.423
cp_fm_diag_elpa_base 50 14.0 0.442 1.354 0.463 1.394
wfi_extrapolate 11 7.9 0.001 0.001 1.353 1.353
grid_collocate_task_list 119 9.7 1.221 1.283 1.221 1.283
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.244 1.255
cp_fm_upper_to_full 72 13.5 0.830 1.147 0.830 1.147
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.126 1.140
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.130 1.130
dbcsr_complete_redistribute 329 12.2 0.121 0.156 0.827 1.104
make_images_data 4572 15.5 0.045 0.049 0.937 1.048
multiply_cannon_sync_h2d 27432 15.5 0.990 1.047 0.990 1.047
hybrid_alltoall_any 4725 16.4 0.062 0.151 0.793 0.968
mp_alltoall_d11v 2130 13.8 0.823 0.939 0.823 0.939
acc_transpose_blocks_kernels 27432 16.5 0.268 0.275 0.856 0.889
cp_fm_cholesky_invert 11 10.9 0.882 0.887 0.882 0.887
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.612 0.881
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.797 0.879
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.854 0.858
mp_alltoall_z22v 1201 16.6 0.777 0.804 0.777 0.804
mp_alltoall_i22 627 13.8 0.446 0.747 0.446 0.747
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=36.463000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=526.636364, yerr=2.422348
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 602.107904E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10224 344.
MP_Sync 104
MP_Alltoall 1582 3682667.
MP_ISendRecv 10710 94533.
MP_Wait 16690
MP_comm_split 50
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.026 29.308 29.309
qs_mol_dyn_low 1 2.0 0.003 0.003 29.091 29.098
qs_forces 11 3.9 0.002 0.002 29.026 29.028
qs_energies 11 4.9 0.009 0.018 27.306 27.310
scf_env_do_scf 11 5.9 0.000 0.001 22.223 22.224
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.457 19.457
velocity_verlet 10 3.0 0.001 0.001 14.754 14.758
dbcsr_multiply_generic 2286 12.5 0.092 0.095 12.934 13.005
qs_scf_new_mos 108 7.5 0.001 0.001 11.619 11.650
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.618 11.649
ot_scf_mini 108 9.5 0.002 0.002 10.909 10.936
multiply_cannon 2286 13.5 0.231 0.240 10.111 10.550
multiply_cannon_loop 2286 14.5 0.332 0.345 9.173 9.424
multiply_cannon_multrec 9144 15.5 1.635 1.865 6.157 6.471
ot_mini 108 10.5 0.001 0.001 6.119 6.153
rebuild_ks_matrix 119 8.3 0.000 0.000 5.965 5.993
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.965 5.993
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.323 5.349
qs_ot_get_derivative 108 11.5 0.001 0.001 4.800 4.827
dbcsr_mm_accdrv_process 12550 15.8 3.313 4.059 4.420 4.543
sum_up_and_integrate 119 10.3 0.037 0.040 3.673 3.678
integrate_v_rspace 119 11.3 0.003 0.003 3.636 3.640
init_scf_run 11 5.9 0.000 0.001 3.574 3.574
scf_env_initial_rho_setup 11 6.9 0.016 0.017 3.573 3.574
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.343 3.350
calculate_rho_elec 119 8.7 0.059 0.061 3.342 3.349
qs_ot_get_p 119 10.4 0.001 0.001 2.942 2.978
init_scf_loop 11 6.9 0.000 0.000 2.743 2.744
calculate_first_density_matrix 1 7.0 0.000 0.000 2.418 2.418
mp_waitall_1 121218 16.5 1.940 2.402 1.940 2.402
make_m2s 4572 13.5 0.035 0.036 1.879 2.019
prepare_preconditioner 11 7.9 0.000 0.000 1.988 1.992
make_preconditioner 11 8.9 0.000 0.000 1.988 1.992
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.938 1.940
make_images 4572 14.5 0.272 0.308 1.790 1.928
grid_integrate_task_list 119 12.3 1.848 1.925 1.848 1.925
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.795 1.896
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.856 1.892
rs_pw_transfer 974 11.9 0.008 0.008 1.782 1.859
pw_transfer 1439 11.6 0.063 0.066 1.833 1.840
calculate_dm_sparse 119 9.5 0.000 0.000 1.819 1.837
density_rs2pw 119 9.7 0.003 0.004 1.734 1.807
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.742 1.750
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.720 1.721
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.712 1.713
jit_kernel_multiply 8 15.6 1.069 1.614 1.069 1.614
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.539 1.545
potential_pw2rs 119 12.3 0.010 0.011 1.442 1.445
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.265 1.433
apply_single 119 13.6 0.000 0.000 1.265 1.433
fft_wrap_pw1pw2_140 487 13.2 0.083 0.089 1.418 1.426
fft3d_ps 1201 14.6 0.542 0.555 1.409 1.417
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.392 1.393
cp_fm_redistribute_end 50 14.0 0.694 1.365 0.695 1.366
cp_fm_diag_elpa_base 50 14.0 0.624 1.290 0.669 1.347
grid_collocate_task_list 119 9.7 1.273 1.345 1.273 1.345
ot_diis_step 108 11.5 0.012 0.013 1.307 1.307
qs_energies_init_hamiltonians 11 5.9 0.008 0.012 1.270 1.273
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.204 1.210
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.182 1.194
wfi_extrapolate 11 7.9 0.001 0.001 1.096 1.096
hybrid_alltoall_any 4725 16.4 0.063 0.176 0.860 1.088
make_images_data 4572 15.5 0.039 0.042 0.892 1.066
cp_fm_cholesky_invert 11 10.9 1.037 1.044 1.037 1.044
acc_transpose_blocks 9144 15.5 0.039 0.040 0.982 0.990
mp_alltoall_d11v 2130 13.8 0.869 0.950 0.869 0.950
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.873 0.923
multiply_cannon_sync_h2d 9144 15.5 0.715 0.791 0.715 0.791
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.772 0.779
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.766 0.768
mp_alltoall_z22v 1201 16.6 0.740 0.767 0.740 0.767
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.714 0.759
acc_transpose_blocks_kernels 9144 16.5 0.117 0.120 0.741 0.745
multiply_cannon_metrocomm3 9144 15.5 0.018 0.019 0.359 0.730
mp_allgather_i34 2286 14.5 0.271 0.727 0.271 0.727
mp_sum_l 7207 12.9 0.512 0.699 0.512 0.699
dbcsr_complete_redistribute 329 12.2 0.166 0.182 0.588 0.629
jit_kernel_transpose 5 15.6 0.624 0.629 0.624 0.629
qs_create_task_list 11 7.9 0.038 0.038 0.584 0.606
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=29.309000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=569.181818, yerr=6.379202
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/08/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 235.585836E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 1388964 0.0% 0.0% 100.0%
average stack size 0.0 0.0 105.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 751.972352E+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_ISendRecv 4998 189067.
MP_Wait 8898
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 0.030 0.121 41.783 41.784
qs_mol_dyn_low 1 2.0 0.003 0.003 41.430 41.438
qs_forces 11 3.9 0.023 0.026 41.367 41.369
qs_energies 11 4.9 0.001 0.001 39.352 39.371
scf_env_do_scf 11 5.9 0.001 0.001 33.641 33.641
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 25.489 25.490
velocity_verlet 10 3.0 0.001 0.001 23.606 23.612
dbcsr_multiply_generic 2286 12.5 0.117 0.119 17.844 18.111
qs_scf_new_mos 108 7.5 0.001 0.001 16.428 16.525
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.427 16.524
ot_scf_mini 108 9.5 0.002 0.002 15.325 15.428
multiply_cannon 2286 13.5 0.306 0.314 13.735 14.733
multiply_cannon_loop 2286 14.5 0.343 0.348 12.447 13.496
ot_mini 108 10.5 0.001 0.001 9.076 9.199
multiply_cannon_multrec 9144 15.5 3.325 4.688 8.596 8.711
init_scf_loop 11 6.9 0.000 0.000 8.121 8.123
prepare_preconditioner 11 7.9 0.000 0.000 7.159 7.173
make_preconditioner 11 8.9 0.000 0.000 7.159 7.173
qs_ot_get_derivative 108 11.5 0.001 0.001 7.008 7.111
rebuild_ks_matrix 119 8.3 0.000 0.000 6.961 7.109
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 6.960 7.109
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.677 7.049
dbcsr_mm_accdrv_process 12550 15.8 3.932 5.330 5.147 6.458
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.282 6.417
cp_fm_upper_to_full 72 14.2 3.266 4.689 3.266 4.689
mp_waitall_1 97218 16.6 2.743 3.835 2.743 3.835
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.780 3.788
calculate_rho_elec 119 8.7 0.118 0.121 3.780 3.788
sum_up_and_integrate 119 10.3 0.064 0.065 3.780 3.786
integrate_v_rspace 119 11.3 0.003 0.003 3.716 3.722
init_scf_run 11 5.9 0.000 0.001 3.680 3.680
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.679 3.680
qs_ot_get_p 119 10.4 0.001 0.001 3.443 3.591
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.547 2.952
dbcsr_complete_redistribute 329 12.2 0.287 0.290 2.079 2.914
make_m2s 4572 13.5 0.038 0.038 2.426 2.604
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.725 2.561
make_images 4572 14.5 0.353 0.384 2.306 2.485
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.190 2.459
apply_single 119 13.6 0.000 0.000 2.190 2.459
multiply_cannon_metrocomm3 9144 15.5 0.019 0.019 1.388 2.401
mp_alltoall_i22 627 13.8 1.459 2.332 1.459 2.332
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.477 2.306
calculate_dm_sparse 119 9.5 0.000 0.000 2.238 2.258
calculate_first_density_matrix 1 7.0 0.000 0.000 2.203 2.204
pw_transfer 1439 11.6 0.066 0.067 2.110 2.114
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.084 2.085
grid_integrate_task_list 119 12.3 2.026 2.052 2.026 2.052
ot_diis_step 108 11.5 0.014 0.014 2.036 2.036
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.015 2.020
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.888 1.889
mp_sum_l 7207 12.9 1.128 1.888 1.128 1.888
density_rs2pw 119 9.7 0.003 0.003 1.853 1.872
jit_kernel_multiply 10 15.5 1.188 1.862 1.188 1.862
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.790 1.843
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.800 1.801
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.753 1.776
fft_wrap_pw1pw2_140 487 13.2 0.088 0.090 1.670 1.675
fft3d_ps 1201 14.6 0.567 0.577 1.649 1.657
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.601 1.644
cp_fm_cholesky_invert 11 10.9 1.575 1.579 1.575 1.579
rs_pw_transfer 974 11.9 0.009 0.010 1.466 1.497
grid_collocate_task_list 119 9.7 1.464 1.493 1.464 1.493
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.482 1.484
cp_fm_diag_elpa_base 50 14.0 1.331 1.386 1.480 1.481
wfi_extrapolate 11 7.9 0.001 0.001 1.414 1.414
hybrid_alltoall_any 4725 16.4 0.087 0.147 1.193 1.412
make_images_data 4572 15.5 0.043 0.046 1.134 1.365
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.284 1.297
potential_pw2rs 119 12.3 0.014 0.015 1.275 1.279
mp_alltoall_d11v 2130 13.8 1.249 1.274 1.249 1.274
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.185 1.209
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.092 1.107
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 1.004 1.061
multiply_cannon_sync_h2d 9144 15.5 1.043 1.047 1.043 1.047
mp_alltoall_z22v 1201 16.6 0.946 0.971 0.946 0.971
qs_create_task_list 11 7.9 0.001 0.001 0.945 0.955
generate_qs_task_list 11 8.9 0.372 0.391 0.944 0.954
acc_transpose_blocks 9144 15.5 0.038 0.039 0.915 0.917
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.890 0.905
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=41.784000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=704.181818, yerr=13.716733
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 498.319360E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65372.
MP_Allreduce 9840 486.
MP_Sync 100
MP_Alltoall 1938 1545165.
MP_ISendRecv 41800 9096.
MP_Wait 58168
MP_comm_split 48
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.051 84.070 84.072
qs_mol_dyn_low 1 2.0 0.003 0.003 83.687 83.697
qs_forces 11 3.9 0.002 0.003 83.616 83.617
qs_energies 11 4.9 0.003 0.011 80.772 80.784
scf_env_do_scf 11 5.9 0.000 0.001 71.836 71.839
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 65.949 65.949
dbcsr_multiply_generic 2055 12.4 0.104 0.107 52.890 53.267
qs_scf_new_mos 99 7.5 0.000 0.001 48.858 48.963
qs_scf_loop_do_ot 99 8.5 0.000 0.001 48.858 48.963
ot_scf_mini 99 9.5 0.002 0.002 46.410 46.511
multiply_cannon 2055 13.4 0.180 0.186 42.910 43.826
velocity_verlet 10 3.0 0.001 0.001 43.223 43.223
multiply_cannon_loop 2055 14.4 1.566 1.603 41.884 42.663
ot_mini 99 10.5 0.001 0.001 28.520 28.612
qs_ot_get_derivative 99 11.5 0.001 0.001 21.600 21.688
multiply_cannon_multrec 49320 15.4 12.043 12.984 17.078 17.828
rebuild_ks_matrix 110 8.3 0.000 0.001 14.529 14.703
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.528 14.702
mp_waitall_1 241148 16.1 12.349 13.332 12.349 13.332
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.758 12.925
multiply_cannon_sync_h2d 49320 15.4 9.895 10.691 9.895 10.691
qs_ot_get_p 110 10.4 0.001 0.001 9.383 9.525
multiply_cannon_metrocomm3 49320 15.4 0.077 0.082 7.218 8.431
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.442 8.017
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.213 7.867
apply_single 110 13.6 0.000 0.000 7.213 7.866
sum_up_and_integrate 110 10.3 0.037 0.043 7.086 7.100
integrate_v_rspace 110 11.3 0.002 0.003 7.049 7.072
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 6.904 6.974
init_scf_run 11 5.9 0.000 0.001 6.785 6.786
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.785 6.786
ot_diis_step 99 11.5 0.005 0.006 6.609 6.610
qs_rho_update_rho_low 110 7.6 0.000 0.001 6.308 6.430
calculate_rho_elec 110 8.6 0.020 0.024 6.307 6.430
mp_sum_l 6514 12.8 5.533 6.260 5.533 6.260
qs_ot_p2m_diag 48 11.0 0.012 0.019 6.012 6.036
init_scf_loop 11 6.9 0.000 0.000 5.860 5.860
dbcsr_mm_accdrv_process 87628 16.1 2.093 2.197 4.914 5.200
cp_dbcsr_syevd 48 12.0 0.002 0.003 4.994 4.994
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.511 4.512
cp_fm_redistribute_end 48 14.0 3.928 4.484 3.932 4.486
cp_fm_diag_elpa_base 48 14.0 0.547 4.388 0.551 4.413
make_m2s 4110 13.4 0.061 0.066 4.013 4.116
rs_pw_transfer 902 11.9 0.012 0.014 3.874 4.091
wfi_extrapolate 11 7.9 0.001 0.001 4.089 4.089
make_images 4110 14.4 0.178 0.192 3.917 4.024
calculate_dm_sparse 110 9.5 0.000 0.001 3.810 3.910
multiply_cannon_metrocomm1 49320 15.4 0.060 0.064 2.867 3.849
prepare_preconditioner 11 7.9 0.000 0.000 3.680 3.702
make_preconditioner 11 8.9 0.000 0.000 3.680 3.702
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.554 3.558
density_rs2pw 110 9.6 0.004 0.004 3.334 3.499
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.443 3.488
grid_integrate_task_list 110 12.3 3.250 3.410 3.250 3.410
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.244 3.305
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.221 3.275
pw_transfer 1331 11.6 0.054 0.064 2.948 3.007
fft_wrap_pw1pw2 1111 12.6 0.007 0.008 2.860 2.923
potential_pw2rs 110 12.3 0.006 0.007 2.667 2.699
mp_alltoall_d11v 2046 13.8 2.270 2.668 2.270 2.668
calculate_first_density_matrix 1 7.0 0.000 0.003 2.613 2.615
jit_kernel_multiply 13 15.9 2.538 2.550 2.538 2.550
fft3d_ps 1111 14.6 0.752 0.837 2.399 2.456
fft_wrap_pw1pw2_140 451 13.1 0.168 0.187 2.376 2.437
acc_transpose_blocks 49320 15.4 0.214 0.224 2.183 2.254
grid_collocate_task_list 110 9.6 2.085 2.194 2.085 2.194
mp_sum_d 3883 11.9 1.569 2.169 1.569 2.169
mp_waitany 14300 13.8 1.859 2.112 1.859 2.112
make_images_data 4110 15.4 0.044 0.047 1.839 1.997
cp_fm_cholesky_invert 11 10.9 1.957 1.961 1.957 1.961
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.910 1.925
hybrid_alltoall_any 4261 16.3 0.082 0.478 1.639 1.890
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=84.072000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=474.363636, yerr=1.720081
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 586.526720E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65587.
MP_Allreduce 9839 562.
MP_Sync 100
MP_Alltoall 1717 2754997.
MP_ISendRecv 20680 26400.
MP_Wait 32692
MP_comm_split 48
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.028 68.403 68.404
qs_mol_dyn_low 1 2.0 0.003 0.003 68.072 68.081
qs_forces 11 3.9 0.002 0.003 68.003 68.005
qs_energies 11 4.9 0.003 0.016 64.678 64.682
scf_env_do_scf 11 5.9 0.001 0.001 56.120 56.123
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 48.431 48.431
dbcsr_multiply_generic 2055 12.4 0.113 0.118 37.652 37.795
velocity_verlet 10 3.0 0.001 0.001 36.162 36.163
qs_scf_new_mos 99 7.5 0.001 0.001 32.520 32.649
qs_scf_loop_do_ot 99 8.5 0.001 0.001 32.520 32.648
multiply_cannon 2055 13.4 0.221 0.239 30.977 32.094
ot_scf_mini 99 9.5 0.004 0.011 30.865 30.987
multiply_cannon_loop 2055 14.4 0.928 0.951 29.652 30.496
ot_mini 99 10.5 0.001 0.001 18.185 18.306
multiply_cannon_multrec 24660 15.4 7.649 9.326 13.834 15.716
rebuild_ks_matrix 110 8.3 0.000 0.001 13.667 13.735
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 13.667 13.735
qs_ot_get_derivative 99 11.5 0.001 0.001 12.399 12.515
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.043 12.105
mp_waitall_1 186928 16.3 7.836 10.534 7.836 10.534
multiply_cannon_sync_h2d 24660 15.4 7.048 7.979 7.048 7.979
multiply_cannon_metrocomm3 24660 15.4 0.070 0.073 5.084 7.825
init_scf_loop 11 6.9 0.001 0.005 7.649 7.650
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.517 7.179
apply_single 110 13.6 0.000 0.001 6.516 7.179
sum_up_and_integrate 110 10.3 0.052 0.058 6.562 6.575
integrate_v_rspace 110 11.3 0.002 0.002 6.510 6.525
dbcsr_mm_accdrv_process 52282 16.1 4.665 5.464 6.025 6.333
qs_ot_get_p 110 10.4 0.001 0.001 6.025 6.173
init_scf_run 11 5.9 0.000 0.001 6.135 6.136
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.135 6.136
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.812 5.819
calculate_rho_elec 110 8.6 0.039 0.047 5.811 5.818
ot_diis_step 99 11.5 0.010 0.010 5.741 5.742
prepare_preconditioner 11 7.9 0.000 0.000 5.628 5.643
make_preconditioner 11 8.9 0.000 0.000 5.628 5.643
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.679 5.395
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.167 5.327
make_m2s 4110 13.4 0.057 0.060 4.263 4.771
make_images 4110 14.4 0.400 0.444 4.154 4.660
qs_ot_p2m_diag 48 11.0 0.028 0.044 4.098 4.121
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.666 3.667
wfi_extrapolate 11 7.9 0.001 0.001 3.533 3.533
pw_transfer 1331 11.6 0.066 0.070 3.365 3.510
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.259 3.402
grid_integrate_task_list 110 12.3 3.157 3.341 3.157 3.341
density_rs2pw 110 9.6 0.004 0.004 3.096 3.329
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.294 3.295
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.189 3.267
rs_pw_transfer 902 11.9 0.012 0.014 3.008 3.176
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.125 3.127
cp_fm_redistribute_end 48 14.0 2.330 3.099 2.332 3.099
cp_fm_diag_elpa_base 48 14.0 0.733 2.973 0.764 3.062
calculate_dm_sparse 110 9.5 0.001 0.001 2.958 2.986
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.919 2.956
make_images_data 4110 15.4 0.048 0.052 2.386 2.920
fft_wrap_pw1pw2_140 451 13.1 0.200 0.217 2.729 2.871
hybrid_alltoall_any 4261 16.3 0.102 0.443 2.098 2.869
fft3d_ps 1111 14.6 1.074 1.269 2.617 2.747
cp_fm_cholesky_invert 11 10.9 2.618 2.625 2.618 2.625
calculate_first_density_matrix 1 7.0 0.000 0.000 2.518 2.519
mp_sum_l 6514 12.8 1.792 2.437 1.792 2.437
potential_pw2rs 110 12.3 0.008 0.009 2.417 2.429
grid_collocate_task_list 110 9.6 2.116 2.254 2.116 2.254
mp_alltoall_d11v 2046 13.8 1.844 2.021 1.844 2.021
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.954 1.972
jit_kernel_multiply 10 16.3 1.011 1.951 1.011 1.951
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 1.897 1.898
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.789 1.801
mp_allgather_i34 2055 14.4 0.656 1.692 0.656 1.692
multiply_cannon_metrocomm4 22605 15.4 0.075 0.078 0.779 1.603
acc_transpose_blocks 24660 15.4 0.106 0.110 1.544 1.567
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.537 1.548
dbcsr_complete_redistribute 325 12.2 0.252 0.315 1.226 1.499
mp_irecv_dv 57340 16.2 0.654 1.486 0.654 1.486
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.378 1.485
mp_waitany 10164 13.8 1.205 1.394 1.205 1.394
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=68.404000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=553.909091, yerr=7.415641
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 657.547264E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65578.
MP_Allreduce 9838 559.
MP_Sync 100
MP_Alltoall 1496 4511006.
MP_ISendRecv 13640 27424.
MP_Wait 32318
MP_comm_split 48
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.029 60.014 60.015
qs_mol_dyn_low 1 2.0 0.003 0.003 59.693 59.704
qs_forces 11 3.9 0.002 0.002 59.626 59.628
qs_energies 11 4.9 0.001 0.002 56.411 56.423
scf_env_do_scf 11 5.9 0.000 0.001 48.258 48.258
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 39.579 39.580
velocity_verlet 10 3.0 0.001 0.001 32.688 32.689
dbcsr_multiply_generic 2055 12.4 0.112 0.116 28.534 28.782
qs_scf_new_mos 99 7.5 0.001 0.001 24.924 25.032
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.924 25.032
ot_scf_mini 99 9.5 0.002 0.003 23.692 23.823
multiply_cannon 2055 13.4 0.211 0.223 22.059 23.210
multiply_cannon_loop 2055 14.4 0.621 0.634 20.853 21.928
ot_mini 99 10.5 0.001 0.001 13.572 13.701
rebuild_ks_matrix 110 8.3 0.000 0.000 12.276 12.402
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 12.276 12.401
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.817 10.933
multiply_cannon_multrec 16440 15.4 3.871 4.911 9.650 10.629
mp_waitall_1 146766 16.3 7.320 10.284 7.320 10.284
qs_ot_get_derivative 99 11.5 0.001 0.001 9.145 9.276
init_scf_loop 11 6.9 0.000 0.000 8.640 8.642
multiply_cannon_metrocomm3 16440 15.4 0.042 0.043 4.257 7.065
prepare_preconditioner 11 7.9 0.000 0.000 6.880 6.899
make_preconditioner 11 8.9 0.000 0.000 6.880 6.899
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.206 6.555
sum_up_and_integrate 110 10.3 0.060 0.060 6.504 6.519
integrate_v_rspace 110 11.3 0.002 0.003 6.444 6.460
dbcsr_mm_accdrv_process 34862 16.1 4.694 5.379 5.632 5.771
init_scf_run 11 5.9 0.000 0.001 5.674 5.674
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.674 5.674
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.634 5.644
calculate_rho_elec 110 8.6 0.058 0.058 5.634 5.643
qs_ot_get_p 110 10.4 0.001 0.001 5.295 5.434
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.948 5.383
apply_single 110 13.6 0.000 0.000 4.948 5.383
make_m2s 4110 13.4 0.050 0.052 4.356 4.717
make_images 4110 14.4 0.393 0.513 4.241 4.601
ot_diis_step 99 11.5 0.010 0.011 4.400 4.400
multiply_cannon_sync_h2d 16440 15.4 3.669 4.233 3.669 4.233
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.103 3.743
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.619 3.622
grid_integrate_task_list 110 12.3 3.178 3.361 3.178 3.361
pw_transfer 1331 11.6 0.065 0.071 3.280 3.288
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.276 3.277
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.173 3.183
density_rs2pw 110 9.6 0.004 0.004 2.895 3.052
wfi_extrapolate 11 7.9 0.001 0.001 3.015 3.015
make_images_data 4110 15.4 0.044 0.048 2.501 3.001
hybrid_alltoall_any 4261 16.3 0.105 0.373 2.230 2.920
rs_pw_transfer 902 11.9 0.011 0.012 2.752 2.911
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.901 2.902
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.754 2.756
cp_fm_cholesky_invert 11 10.9 2.730 2.736 2.730 2.736
cp_fm_redistribute_end 48 14.0 1.714 2.730 1.716 2.731
cp_fm_diag_elpa_base 48 14.0 0.951 2.587 1.009 2.699
fft_wrap_pw1pw2_140 451 13.1 0.210 0.213 2.673 2.684
calculate_dm_sparse 110 9.5 0.001 0.001 2.550 2.584
calculate_first_density_matrix 1 7.0 0.000 0.000 2.569 2.569
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.487 2.533
fft3d_ps 1111 14.6 1.064 1.072 2.488 2.496
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.368 2.443
multiply_cannon_metrocomm4 14385 15.4 0.046 0.048 0.898 2.435
potential_pw2rs 110 12.3 0.011 0.011 2.315 2.328
mp_irecv_dv 48980 15.7 0.827 2.314 0.827 2.314
grid_collocate_task_list 110 9.6 2.176 2.309 2.176 2.309
mp_alltoall_d11v 2046 13.8 1.893 2.208 1.893 2.208
mp_sum_l 6514 12.8 1.501 2.097 1.501 2.097
qs_energies_init_hamiltonians 11 5.9 0.003 0.020 2.009 2.010
dbcsr_complete_redistribute 325 12.2 0.346 0.380 1.499 1.959
cp_fm_upper_to_full 70 13.6 1.381 1.856 1.381 1.856
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.657 1.674
cp_fm_cholesky_decompose 22 10.9 1.652 1.671 1.652 1.671
mp_allgather_i34 2055 14.4 0.531 1.582 0.531 1.582
jit_kernel_multiply 8 16.6 0.551 1.539 0.551 1.539
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.511 1.526
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.046 1.508
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.367 1.501
qs_ot_get_orbitals 99 10.5 0.000 0.001 1.264 1.271
mp_waitany 17072 13.8 1.112 1.270 1.112 1.270
acc_transpose_blocks 16440 15.4 0.072 0.075 1.231 1.243
mp_alltoall_z22v 1111 16.6 1.211 1.231 1.211 1.231
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=60.015000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=624.090909, yerr=9.049908
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 728.682496E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65576.
MP_Allreduce 9838 600.
MP_Sync 100
MP_Alltoall 1496 5863162.
MP_ISendRecv 10120 43184.
MP_Wait 25102
MP_comm_split 48
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.048 0.138 68.619 68.619
qs_mol_dyn_low 1 2.0 0.003 0.003 68.083 68.093
qs_forces 11 3.9 0.002 0.004 67.886 67.887
qs_energies 11 4.9 0.003 0.011 64.475 64.479
scf_env_do_scf 11 5.9 0.001 0.001 55.621 55.623
scf_env_do_scf_inner_loop 99 6.5 0.003 0.015 42.855 42.856
velocity_verlet 10 3.0 0.001 0.001 38.502 38.510
dbcsr_multiply_generic 2055 12.4 0.123 0.133 30.339 30.555
qs_scf_new_mos 99 7.5 0.001 0.001 27.966 28.066
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.965 28.065
ot_scf_mini 99 9.5 0.003 0.004 26.298 26.392
multiply_cannon 2055 13.4 0.243 0.264 22.881 23.982
multiply_cannon_loop 2055 14.4 0.883 0.915 21.278 21.895
ot_mini 99 10.5 0.001 0.001 14.774 14.895
multiply_cannon_multrec 24660 15.4 4.237 6.956 12.740 13.939
init_scf_loop 11 6.9 0.003 0.020 12.721 12.722
rebuild_ks_matrix 110 8.3 0.000 0.000 12.290 12.391
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.014 12.290 12.391
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.910 11.005
prepare_preconditioner 11 7.9 0.000 0.000 10.870 10.885
make_preconditioner 11 8.9 0.000 0.000 10.870 10.885
qs_ot_get_derivative 99 11.5 0.001 0.001 10.564 10.662
make_full_inverse_cholesky 11 9.9 0.000 0.000 9.046 10.537
dbcsr_mm_accdrv_process 52304 16.0 6.860 8.103 8.356 9.253
mp_waitall_1 126806 16.4 5.041 7.016 5.041 7.016
sum_up_and_integrate 110 10.3 0.067 0.069 6.591 6.614
qs_ot_get_p 110 10.4 0.001 0.001 6.428 6.568
integrate_v_rspace 110 11.3 0.003 0.003 6.524 6.549
make_m2s 4110 13.4 0.060 0.061 5.662 6.044
init_scf_run 11 5.9 0.000 0.001 5.923 5.923
scf_env_initial_rho_setup 11 6.9 0.006 0.043 5.923 5.923
make_images 4110 14.4 0.578 0.704 5.522 5.901
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.868 5.879
calculate_rho_elec 110 8.6 0.077 0.081 5.867 5.879
cp_fm_upper_to_full 70 13.8 3.381 4.799 3.381 4.799
qs_ot_p2m_diag 48 11.0 0.055 0.064 4.470 4.487
ot_diis_step 99 11.5 0.011 0.011 4.158 4.158
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.062 4.148
apply_single 110 13.6 0.000 0.000 4.062 4.148
dbcsr_complete_redistribute 325 12.2 0.411 0.456 2.815 3.977
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.969 3.970
multiply_cannon_metrocomm3 24660 15.4 0.035 0.036 1.606 3.636
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.452 3.511
pw_transfer 1331 11.6 0.064 0.071 3.466 3.504
grid_integrate_task_list 110 12.3 3.254 3.444 3.254 3.444
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.361 3.402
multiply_cannon_sync_h2d 24660 15.4 3.179 3.392 3.179 3.392
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.367 3.370
copy_fm_to_dbcsr 174 11.2 0.001 0.002 2.218 3.362
cp_fm_redistribute_end 48 14.0 1.670 3.317 1.672 3.318
cp_fm_cholesky_invert 11 10.9 3.303 3.314 3.303 3.314
cp_fm_diag_elpa_base 48 14.0 1.560 3.175 1.642 3.287
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.192 3.247
make_images_data 4110 15.4 0.046 0.051 2.845 3.201
hybrid_alltoall_any 4261 16.3 0.120 0.456 2.463 3.153
density_rs2pw 110 9.6 0.004 0.004 2.988 3.124
wfi_extrapolate 11 7.9 0.001 0.001 3.046 3.046
calculate_dm_sparse 110 9.5 0.001 0.001 3.000 3.032
rs_pw_transfer 902 11.9 0.010 0.011 2.807 2.968
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.815 2.937
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.923 2.924
mp_alltoall_i22 605 13.7 1.707 2.909 1.707 2.909
fft_wrap_pw1pw2_140 451 13.1 0.201 0.212 2.795 2.837
calculate_first_density_matrix 1 7.0 0.000 0.003 2.768 2.777
fft3d_ps 1111 14.6 1.069 1.103 2.664 2.690
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.522 2.563
grid_collocate_task_list 110 9.6 2.220 2.394 2.220 2.394
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 2.390 2.391
potential_pw2rs 110 12.3 0.012 0.013 2.346 2.364
mp_alltoall_d11v 2046 13.8 1.990 2.257 1.990 2.257
jit_kernel_multiply 12 15.9 1.167 2.232 1.167 2.232
cp_fm_cholesky_decompose 22 10.9 1.910 1.967 1.910 1.967
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.767 1.797
mp_allgather_i34 2055 14.4 0.703 1.720 0.703 1.720
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.606 1.705
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.645 1.654
mp_sum_l 6514 12.8 1.112 1.654 1.112 1.654
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.579 1.597
acc_transpose_blocks 24660 15.4 0.104 0.107 1.498 1.527
mp_waitany 13376 13.8 1.202 1.483 1.202 1.483
multiply_cannon_metrocomm4 20550 15.4 0.057 0.060 0.835 1.451
qs_env_update_s_mstruct 11 6.9 0.000 0.001 1.332 1.444
mp_alltoall_z22v 1111 16.6 1.378 1.441 1.378 1.441
mp_irecv_dv 62702 16.1 0.738 1.376 0.738 1.376
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=68.619000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=688.818182, yerr=8.726326
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 827.019264E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65574.
MP_Allreduce 9838 640.
MP_Sync 100
MP_Alltoall 1496 8504061.
MP_ISendRecv 6600 54848.
MP_Wait 17226
MP_comm_split 48
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.030 55.473 55.474
qs_mol_dyn_low 1 2.0 0.003 0.003 55.063 55.073
qs_forces 11 3.9 0.008 0.009 54.997 54.998
qs_energies 11 4.9 0.001 0.001 51.376 51.382
scf_env_do_scf 11 5.9 0.000 0.001 42.875 42.875
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 34.908 34.908
velocity_verlet 10 3.0 0.001 0.001 30.851 30.854
dbcsr_multiply_generic 2055 12.4 0.104 0.106 23.220 23.429
qs_scf_new_mos 99 7.5 0.001 0.001 20.791 20.848
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.790 20.848
ot_scf_mini 99 9.5 0.002 0.002 19.551 19.599
multiply_cannon 2055 13.4 0.245 0.262 17.761 18.913
multiply_cannon_loop 2055 14.4 0.322 0.334 16.385 16.716
rebuild_ks_matrix 110 8.3 0.000 0.000 11.489 11.552
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 11.489 11.551
ot_mini 99 10.5 0.001 0.001 10.546 10.593
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.239 10.295
multiply_cannon_multrec 8220 15.4 3.221 5.039 7.682 9.049
mp_waitall_1 106626 16.5 6.235 7.958 6.235 7.958
init_scf_loop 11 6.9 0.000 0.000 7.919 7.920
qs_ot_get_derivative 99 11.5 0.001 0.001 6.661 6.710
prepare_preconditioner 11 7.9 0.000 0.000 6.243 6.250
make_preconditioner 11 8.9 0.000 0.000 6.243 6.250
sum_up_and_integrate 110 10.3 0.080 0.081 6.206 6.219
integrate_v_rspace 110 11.3 0.003 0.003 6.126 6.141
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.825 5.901
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.624 5.641
calculate_rho_elec 110 8.6 0.115 0.116 5.624 5.640
init_scf_run 11 5.9 0.000 0.001 5.392 5.392
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.391 5.392
dbcsr_mm_accdrv_process 17442 15.9 2.972 4.180 4.331 5.322
qs_ot_get_p 110 10.4 0.001 0.001 5.077 5.118
multiply_cannon_metrocomm3 8220 15.4 0.017 0.017 3.125 4.664
make_m2s 4110 13.4 0.039 0.040 4.174 4.424
make_images 4110 14.4 0.636 0.698 4.043 4.290
ot_diis_step 99 11.5 0.012 0.012 3.861 3.861
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.760 3.829
apply_single 110 13.6 0.000 0.000 3.760 3.829
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.696 3.700
grid_integrate_task_list 110 12.3 3.359 3.507 3.359 3.507
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.391 3.392
pw_transfer 1331 11.6 0.065 0.071 3.269 3.279
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.162 3.176
cp_fm_cholesky_invert 11 10.9 3.070 3.074 3.070 3.074
multiply_cannon_sync_h2d 8220 15.4 2.899 3.043 2.899 3.043
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.857 2.858
density_rs2pw 110 9.6 0.004 0.004 2.676 2.858
cp_fm_redistribute_end 48 14.0 0.726 2.828 0.730 2.828
cp_fm_diag_elpa_base 48 14.0 1.922 2.636 2.090 2.792
make_images_data 4110 15.4 0.038 0.043 2.373 2.756
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.731 2.733
fft_wrap_pw1pw2_140 451 13.1 0.211 0.216 2.705 2.725
wfi_extrapolate 11 7.9 0.001 0.001 2.718 2.718
hybrid_alltoall_any 4261 16.3 0.200 0.858 2.285 2.664
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.659 2.660
calculate_dm_sparse 110 9.5 0.001 0.001 2.570 2.624
calculate_first_density_matrix 1 7.0 0.000 0.000 2.587 2.588
grid_collocate_task_list 110 9.6 2.330 2.436 2.330 2.436
fft3d_ps 1111 14.6 1.121 1.134 2.422 2.432
rs_pw_transfer 902 11.9 0.010 0.011 2.217 2.408
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.355 2.378
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 1.995 2.022
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.779 1.999
potential_pw2rs 110 12.3 0.015 0.015 1.954 1.961
mp_alltoall_d11v 2046 13.8 1.647 1.870 1.647 1.870
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.816 1.837
cp_fm_cholesky_decompose 22 10.9 1.719 1.739 1.719 1.739
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.505 1.623
mp_allgather_i34 2055 14.4 0.538 1.614 0.538 1.614
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.580 1.588
dbcsr_complete_redistribute 325 12.2 0.563 0.583 1.488 1.565
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.531 1.546
multiply_cannon_metrocomm4 6165 15.4 0.018 0.019 0.483 1.465
mp_irecv_dv 24056 15.7 0.459 1.424 0.459 1.424
qs_create_task_list 11 7.9 0.000 0.001 1.224 1.324
generate_qs_task_list 11 8.9 0.378 0.446 1.224 1.323
mp_waitany 9240 13.8 1.071 1.281 1.071 1.281
multiply_cannon_metrocomm1 8220 15.4 0.021 0.022 0.748 1.223
jit_kernel_multiply 6 15.7 1.052 1.208 1.052 1.208
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.152 1.177
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=55.474000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=780.181818, yerr=11.247773
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 1.340137E+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_ISendRecv 4620 360267.
MP_Wait 7524
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 0.023 0.084 86.453 86.454
qs_mol_dyn_low 1 2.0 0.003 0.003 86.101 86.110
qs_forces 11 3.9 0.002 0.002 86.028 86.030
qs_energies 11 4.9 0.002 0.002 81.982 81.986
scf_env_do_scf 11 5.9 0.001 0.001 72.117 72.117
velocity_verlet 10 3.0 0.001 0.001 55.495 55.501
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 43.282 43.284
dbcsr_multiply_generic 2055 12.4 0.120 0.122 28.866 28.910
init_scf_loop 11 6.9 0.000 0.000 28.761 28.765
qs_scf_new_mos 99 7.5 0.001 0.001 26.801 26.836
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.801 26.835
prepare_preconditioner 11 7.9 0.000 0.000 26.811 26.818
make_preconditioner 11 8.9 0.000 0.000 26.811 26.818
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.912 26.272
ot_scf_mini 99 9.5 0.002 0.002 25.025 25.052
multiply_cannon 2055 13.4 0.355 0.381 21.567 22.353
multiply_cannon_loop 2055 14.4 0.342 0.345 19.731 20.128
cp_fm_upper_to_full 70 14.2 12.795 18.404 12.795 18.404
ot_mini 99 10.5 0.001 0.001 13.803 13.824
rebuild_ks_matrix 110 8.3 0.000 0.001 13.246 13.299
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.014 13.246 13.299
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.029 12.077
dbcsr_complete_redistribute 325 12.2 1.024 1.040 7.524 10.834
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.466 9.781
multiply_cannon_multrec 8220 15.4 4.385 4.562 9.568 9.662
qs_ot_get_derivative 99 11.5 0.001 0.001 9.170 9.198
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.884 9.175
mp_waitall_1 87304 16.6 8.065 8.998 8.065 8.998
mp_alltoall_i22 605 13.7 5.498 8.815 5.498 8.815
sum_up_and_integrate 110 10.3 0.151 0.153 6.672 6.688
integrate_v_rspace 110 11.3 0.004 0.004 6.520 6.537
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.498 6.535
calculate_rho_elec 110 8.6 0.227 0.227 6.498 6.535
make_m2s 4110 13.4 0.043 0.044 5.499 6.068
qs_ot_get_p 110 10.4 0.001 0.001 5.949 5.976
make_images 4110 14.4 0.878 0.917 5.310 5.879
init_scf_run 11 5.9 0.000 0.001 5.621 5.622
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.621 5.621
cp_fm_cholesky_invert 11 10.9 5.461 5.466 5.461 5.466
dbcsr_mm_accdrv_process 11614 15.7 3.273 3.637 5.041 5.261
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.735 5.183
apply_single 110 13.6 0.000 0.000 4.735 5.183
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 4.804 5.179
ot_diis_step 99 11.5 0.015 0.015 4.608 4.609
qs_ot_p2m_diag 48 11.0 0.151 0.155 4.303 4.310
multiply_cannon_sync_h2d 8220 15.4 3.951 3.957 3.951 3.957
hybrid_alltoall_any 4261 16.3 0.256 0.554 3.065 3.894
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.859 3.860
pw_transfer 1331 11.6 0.074 0.074 3.823 3.827
make_images_data 4110 15.4 0.042 0.045 3.089 3.799
grid_integrate_task_list 110 12.3 3.658 3.712 3.658 3.712
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 3.707 3.711
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.639 3.642
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.926 3.374
wfi_extrapolate 11 7.9 0.001 0.001 3.263 3.263
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.236 3.237
cp_fm_diag_elpa_base 48 14.0 2.701 2.899 3.234 3.234
fft_wrap_pw1pw2_140 451 13.1 0.214 0.217 3.192 3.200
calculate_dm_sparse 110 9.5 0.001 0.001 3.137 3.157
density_rs2pw 110 9.6 0.004 0.004 2.996 3.014
fft3d_ps 1111 14.6 1.262 1.267 2.918 2.925
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.887 2.891
grid_collocate_task_list 110 9.6 2.628 2.647 2.628 2.647
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.435 2.453
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.374 2.391
calculate_first_density_matrix 1 7.0 0.000 0.000 2.254 2.254
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.176 2.234
rs_pw_transfer 902 11.9 0.011 0.011 2.152 2.205
mp_alltoall_d11v 2046 13.8 2.103 2.168 2.103 2.168
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.097 2.145
cp_fm_cholesky_decompose 22 10.9 2.124 2.140 2.124 2.140
potential_pw2rs 110 12.3 0.021 0.021 2.100 2.105
qs_create_task_list 11 7.9 0.001 0.001 1.903 1.947
generate_qs_task_list 11 8.9 0.739 0.793 1.902 1.947
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.793 1.833
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.821 1.826
jit_kernel_multiply 10 15.3 1.566 1.732 1.566 1.732
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=86.454000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1201.272727, yerr=61.357508
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420243808256 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528908111872 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514772E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755942624 0.0% 0.0% 100.0%
number of processed stacks 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 623.960064E+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 434.009031E+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 3175955597216
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4103 56513.
MP_Allreduce 11298 784.
MP_Sync 170
MP_Alltoall 2226 2452168.
MP_ISendRecv 48640 18752.
MP_Wait 66796
MP_comm_split 83
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.035 205.356 205.359
qs_mol_dyn_low 1 2.0 0.004 0.012 204.859 204.873
qs_forces 11 3.9 0.004 0.006 204.756 204.758
qs_energies 11 4.9 0.002 0.003 199.215 199.225
scf_env_do_scf 11 5.9 0.001 0.001 182.592 182.596
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 161.388 161.390
dbcsr_multiply_generic 2507 12.6 0.175 0.179 124.255 124.999
velocity_verlet 10 3.0 0.001 0.001 123.395 123.396
qs_scf_new_mos 117 7.6 0.001 0.001 122.221 122.584
qs_scf_loop_do_ot 117 8.6 0.001 0.001 122.221 122.583
ot_scf_mini 117 9.6 0.003 0.003 115.582 115.953
multiply_cannon 2507 13.6 0.238 0.246 100.919 103.463
multiply_cannon_loop 2507 14.6 2.116 2.186 98.664 101.169
ot_mini 117 10.6 0.001 0.001 65.707 66.125
multiply_cannon_multrec 60168 15.6 33.084 35.447 41.461 43.797
qs_ot_get_derivative 117 11.6 0.001 0.001 40.744 41.127
rebuild_ks_matrix 128 8.3 0.001 0.001 33.522 33.993
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.017 33.521 33.992
mp_waitall_1 291448 16.2 28.905 32.394 28.905 32.394
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.169 30.615
multiply_cannon_sync_h2d 60168 15.6 27.490 30.229 27.490 30.229
qs_ot_get_p 128 10.4 0.001 0.001 27.986 28.412
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.323 24.909
apply_single 128 13.6 0.001 0.001 24.323 24.909
ot_diis_step 117 11.6 0.008 0.008 24.584 24.586
qs_ot_p2m_diag 83 11.4 0.077 0.091 21.209 21.297
init_scf_loop 11 6.9 0.000 0.000 21.126 21.127
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 18.968 19.289
cp_dbcsr_syevd 83 12.4 0.004 0.005 18.630 18.631
multiply_cannon_metrocomm3 60168 15.6 0.112 0.117 15.455 18.290
prepare_preconditioner 11 7.9 0.000 0.000 16.517 16.558
make_preconditioner 11 8.9 0.000 0.000 16.517 16.558
make_full_inverse_cholesky 11 9.9 0.000 0.000 15.756 15.925
cp_fm_diag_elpa 83 13.4 0.000 0.001 15.697 15.704
cp_fm_redistribute_end 83 14.4 12.385 15.612 12.400 15.617
cp_fm_diag_elpa_base 83 14.4 3.168 15.307 3.202 15.424
make_m2s 5014 13.6 0.103 0.111 13.882 14.373
make_images 5014 14.6 0.403 0.421 13.702 14.205
sum_up_and_integrate 128 10.3 0.090 0.107 14.154 14.169
integrate_v_rspace 128 11.3 0.003 0.004 14.064 14.082
qs_rho_update_rho_low 128 7.7 0.001 0.001 12.854 12.971
calculate_rho_elec 128 8.7 0.045 0.063 12.853 12.970
init_scf_run 11 5.9 0.000 0.001 12.484 12.485
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.484 12.485
cp_fm_cholesky_invert 11 10.9 9.559 9.567 9.559 9.567
mp_sum_l 7870 13.0 8.197 9.538 8.197 9.538
wfi_extrapolate 11 7.9 0.001 0.001 9.123 9.123
multiply_cannon_metrocomm1 60168 15.6 0.088 0.092 6.242 8.739
calculate_dm_sparse 128 9.5 0.001 0.001 8.464 8.580
dbcsr_mm_accdrv_process 124484 16.2 3.308 3.451 7.932 8.420
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.007 8.179
make_images_data 5014 15.6 0.070 0.077 6.855 7.873
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.745 7.844
grid_integrate_task_list 128 12.3 7.107 7.552 7.107 7.552
hybrid_alltoall_any 5200 16.5 0.291 2.252 5.983 7.323
pw_transfer 1547 11.6 0.075 0.108 6.919 7.200
density_rs2pw 128 9.7 0.006 0.007 6.623 7.130
fft_wrap_pw1pw2 1291 12.7 0.010 0.013 6.715 6.967
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.782 6.790
rs_pw_transfer 1046 11.9 0.017 0.020 5.868 6.457
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.880 6.066
fft_wrap_pw1pw2_140 523 13.2 0.443 0.506 5.726 5.890
fft3d_ps 1291 14.7 2.096 2.579 5.511 5.715
mp_alltoall_d11v 2415 14.1 4.305 5.377 4.305 5.377
grid_collocate_task_list 128 9.7 4.726 5.130 4.726 5.130
cp_fm_cholesky_decompose 22 10.9 4.778 4.792 4.778 4.792
potential_pw2rs 128 12.3 0.009 0.011 4.585 4.618
mp_sum_d 4473 12.2 3.727 4.553 3.727 4.553
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=205.359000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=591.818182, yerr=6.191169
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 2.183246E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5975232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.7
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 834.859008E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100942E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703955E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70860 2317615104
32768 < size <= 131072 722992 55511613440
131072 < size <= 4194304 1375664 1398181724160
4194304 < size <= 16777216 154704 1463834332048
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4075 57341.
MP_Allreduce 11227 947.
MP_Sync 170
MP_Alltoall 1969 6666003.
MP_ISendRecv 24064 47072.
MP_Wait 37948
MP_comm_split 83
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 0.018 0.031 193.515 193.516
qs_mol_dyn_low 1 2.0 0.003 0.003 193.074 193.088
qs_forces 11 3.9 0.003 0.003 192.988 192.989
qs_energies 11 4.9 0.002 0.002 186.207 186.218
scf_env_do_scf 11 5.9 0.001 0.001 169.576 169.586
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 135.882 135.884
velocity_verlet 10 3.0 0.001 0.001 121.580 121.582
dbcsr_multiply_generic 2507 12.6 0.185 0.190 98.160 99.309
qs_scf_new_mos 117 7.6 0.001 0.001 96.535 97.039
qs_scf_loop_do_ot 117 8.6 0.001 0.001 96.534 97.038
ot_scf_mini 117 9.6 0.004 0.004 91.699 92.267
multiply_cannon 2507 13.6 0.480 0.531 77.394 82.151
multiply_cannon_loop 2507 14.6 1.257 1.297 73.983 76.382
ot_mini 117 10.6 0.001 0.001 50.173 50.690
mp_waitall_1 226760 16.4 25.700 38.545 25.700 38.545
multiply_cannon_multrec 30084 15.6 21.940 26.806 31.536 36.916
rebuild_ks_matrix 128 8.3 0.001 0.001 33.276 33.800
qs_ks_build_kohn_sham_matrix 128 9.3 0.021 0.049 33.276 33.799
init_scf_loop 11 6.9 0.000 0.000 33.597 33.598
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.008 30.503
prepare_preconditioner 11 7.9 0.000 0.000 29.135 29.200
make_preconditioner 11 8.9 0.000 0.000 29.135 29.200
qs_ot_get_derivative 117 11.6 0.001 0.002 28.240 28.784
multiply_cannon_metrocomm3 30084 15.6 0.095 0.101 15.760 28.568
make_full_inverse_cholesky 11 9.9 0.000 0.000 27.829 28.380
qs_ot_get_p 128 10.4 0.001 0.001 23.215 23.964
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 21.987 23.122
apply_single 128 13.6 0.001 0.001 21.986 23.121
ot_diis_step 117 11.6 0.014 0.015 21.763 21.766
multiply_cannon_sync_h2d 30084 15.6 19.173 21.122 19.173 21.122
qs_ot_p2m_diag 83 11.4 0.187 0.215 18.201 18.233
cp_fm_cholesky_invert 11 10.9 17.094 17.108 17.094 17.108
cp_dbcsr_syevd 83 12.4 0.005 0.006 17.017 17.019
make_m2s 5014 13.6 0.089 0.094 14.927 16.512
make_images 5014 14.6 1.157 1.353 14.721 16.308
sum_up_and_integrate 128 10.3 0.114 0.133 14.746 14.776
integrate_v_rspace 128 11.3 0.003 0.003 14.632 14.667
cp_fm_diag_elpa 83 13.4 0.000 0.001 13.816 13.823
cp_fm_redistribute_end 83 14.4 8.065 13.734 8.077 13.738
cp_fm_diag_elpa_base 83 14.4 5.420 13.277 5.643 13.626
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.288 13.324
calculate_rho_elec 128 8.7 0.088 0.105 13.287 13.323
init_scf_run 11 5.9 0.000 0.001 11.851 11.852
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.850 11.852
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 11.206 11.622
make_images_data 5014 15.6 0.067 0.075 9.039 10.881
multiply_cannon_metrocomm4 27577 15.6 0.097 0.112 3.760 10.800
mp_irecv_dv 69486 16.3 3.567 10.415 3.567 10.415
hybrid_alltoall_any 5200 16.5 0.343 1.526 7.642 10.194
dbcsr_mm_accdrv_process 62242 16.2 4.625 5.281 9.058 9.646
wfi_extrapolate 11 7.9 0.001 0.001 8.513 8.513
pw_transfer 1547 11.6 0.085 0.104 8.033 8.099
fft_wrap_pw1pw2 1291 12.7 0.010 0.010 7.810 7.870
grid_integrate_task_list 128 12.3 7.156 7.589 7.156 7.589
density_rs2pw 128 9.7 0.006 0.006 7.024 7.392
cp_fm_cholesky_decompose 22 10.9 7.169 7.244 7.169 7.244
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.245 6.961
fft_wrap_pw1pw2_140 523 13.2 0.468 0.521 6.729 6.798
calculate_dm_sparse 128 9.5 0.001 0.001 6.499 6.642
rs_pw_transfer 1046 11.9 0.014 0.016 6.002 6.448
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.397 6.407
fft3d_ps 1291 14.7 2.773 2.928 6.185 6.229
mp_sum_l 7870 13.0 4.061 6.028 4.061 6.028
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.515 5.651
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.349 5.416
grid_collocate_task_list 128 9.7 4.907 5.350 4.907 5.350
mp_allgather_i34 2507 14.6 1.985 5.262 1.985 5.262
potential_pw2rs 128 12.3 0.015 0.017 5.147 5.176
mp_alltoall_d11v 2415 14.1 4.317 4.686 4.317 4.686
dbcsr_complete_redistribute 395 12.7 0.770 0.845 3.213 4.094
mp_sum_d 4459 12.1 2.691 4.012 2.691 4.012
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=193.516000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=793.818182, yerr=3.785575
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 2.906045E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 3951168 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1697.1
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 949.002240E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1033760
MPI messages size (bytes):
total size 2.695213E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.607194E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 279168 36591108096
131072 < size <= 4194304 654272 987691483136
4194304 < size <= 16777216 65184 925172905552
16777216 < size 28448 745747251200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4079 57277.
MP_Allreduce 11236 986.
MP_Sync 168
MP_Alltoall 1700 9383497.
MP_ISendRecv 15748 75008.
MP_Wait 29528
MP_comm_split 82
MP_ISend 11660 275234.
MP_IRecv 11660 275234.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.029 0.050 177.353 177.353
qs_mol_dyn_low 1 2.0 0.003 0.003 176.830 176.845
qs_forces 11 3.9 0.003 0.003 176.720 176.727
qs_energies 11 4.9 0.001 0.002 170.180 170.192
scf_env_do_scf 11 5.9 0.001 0.001 154.603 154.604
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 118.835 118.836
velocity_verlet 10 3.0 0.001 0.001 113.343 113.345
qs_scf_new_mos 116 7.6 0.001 0.001 82.342 82.656
qs_scf_loop_do_ot 116 8.6 0.001 0.001 82.341 82.655
dbcsr_multiply_generic 2485 12.5 0.180 0.184 81.332 82.548
ot_scf_mini 116 9.6 0.003 0.004 78.214 78.599
multiply_cannon 2485 13.5 0.494 0.515 61.391 66.166
multiply_cannon_loop 2485 14.5 0.856 0.888 57.986 60.856
ot_mini 116 10.6 0.001 0.001 42.395 42.810
init_scf_loop 11 6.9 0.000 0.000 35.662 35.663
mp_waitall_1 176908 16.5 25.059 34.651 25.059 34.651
prepare_preconditioner 11 7.9 0.000 0.000 31.639 31.694
make_preconditioner 11 8.9 0.000 0.000 31.639 31.694
rebuild_ks_matrix 127 8.3 0.001 0.001 30.078 30.656
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.017 30.078 30.655
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.269 30.649
qs_ks_update_qs_env 127 7.6 0.001 0.001 27.122 27.656
multiply_cannon_multrec 19880 15.5 13.450 16.457 22.067 25.128
multiply_cannon_metrocomm3 19880 15.5 0.057 0.062 14.757 24.460
qs_ot_get_derivative 116 11.6 0.001 0.002 22.911 23.306
qs_ot_get_p 127 10.4 0.001 0.001 21.283 21.771
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.618 20.512
apply_single 127 13.6 0.001 0.001 19.618 20.511
ot_diis_step 116 11.6 0.017 0.018 19.382 19.383
qs_ot_p2m_diag 82 11.4 0.261 0.269 16.915 16.921
multiply_cannon_sync_h2d 19880 15.5 14.183 15.995 14.183 15.995
make_m2s 4970 13.5 0.079 0.086 14.792 15.936
cp_dbcsr_syevd 82 12.4 0.005 0.005 15.890 15.891
make_images 4970 14.5 1.181 1.287 14.562 15.704
cp_fm_cholesky_invert 11 10.9 14.962 14.972 14.962 14.972
sum_up_and_integrate 127 10.3 0.130 0.143 14.197 14.224
integrate_v_rspace 127 11.3 0.003 0.004 14.066 14.098
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.042 13.072
calculate_rho_elec 127 8.7 0.131 0.145 13.041 13.071
cp_fm_diag_elpa 82 13.4 0.000 0.001 12.708 12.712
cp_fm_redistribute_end 82 14.4 4.815 12.634 4.832 12.638
cp_fm_diag_elpa_base 82 14.4 7.372 11.998 7.781 12.515
init_scf_run 11 5.9 0.000 0.001 10.598 10.598
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.597 10.598
make_images_data 4970 15.5 0.060 0.068 9.064 10.522
hybrid_alltoall_any 5155 16.4 0.427 1.963 8.015 9.789
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 8.996 9.291
multiply_cannon_metrocomm4 17395 15.5 0.062 0.072 3.389 9.012
mp_irecv_dv 49801 16.2 3.267 8.770 3.267 8.770
dbcsr_mm_accdrv_process 41158 16.2 4.437 5.234 8.067 8.182
pw_transfer 1535 11.6 0.084 0.105 7.889 7.994
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 7.668 7.779
grid_integrate_task_list 127 12.3 7.225 7.697 7.225 7.697
cp_fm_cholesky_decompose 22 10.9 7.557 7.585 7.557 7.585
wfi_extrapolate 11 7.9 0.001 0.001 7.461 7.461
cp_fm_upper_to_full 104 14.5 5.699 7.436 5.699 7.436
fft_wrap_pw1pw2_140 519 13.2 0.474 0.525 6.696 6.814
density_rs2pw 127 9.7 0.006 0.006 6.551 6.809
dbcsr_complete_redistribute 393 12.7 1.162 1.192 4.657 6.424
fft3d_ps 1281 14.7 2.673 2.877 5.951 6.016
calculate_dm_sparse 127 9.5 0.001 0.001 5.743 5.863
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.467 5.471
rs_pw_transfer 1038 11.9 0.014 0.014 5.138 5.468
grid_collocate_task_list 127 9.7 5.041 5.443 5.041 5.443
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.618 5.293
copy_fm_to_dbcsr 208 11.6 0.002 0.002 3.434 5.194
mp_alltoall_d11v 2401 14.1 4.377 5.152 4.377 5.152
mp_allgather_i34 2485 14.5 1.941 4.965 1.941 4.965
mp_sum_l 7804 13.0 3.347 4.889 3.347 4.889
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.570 4.694
potential_pw2rs 127 12.3 0.020 0.022 4.616 4.639
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.350 4.054
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.987 4.015
mp_alltoall_i22 712 14.1 1.934 3.782 1.934 3.782
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.712 3.712
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=177.353000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=894.818182, yerr=11.953076
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 4.320339E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 5927808 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.2
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.137410E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1133160
MPI messages size (bytes):
total size 2.008142E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.772161E+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 315952 35695099904
131072 < size <= 4194304 709496 778939400192
4194304 < size <= 16777216 69840 660837789680
16777216 < size 30480 532676608000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4085 57191.
MP_Allreduce 11251 1068.
MP_Sync 168
MP_Alltoall 1700 12496381.
MP_ISendRecv 11684 75008.
MP_Wait 28114
MP_comm_split 82
MP_ISend 14840 244848.
MP_IRecv 14840 244848.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.060 0.148 187.704 187.706
qs_mol_dyn_low 1 2.0 0.003 0.003 187.174 187.187
qs_forces 11 3.9 0.003 0.003 187.060 187.069
qs_energies 11 4.9 0.013 0.064 179.981 179.991
scf_env_do_scf 11 5.9 0.009 0.048 163.207 163.221
velocity_verlet 10 3.0 0.001 0.001 123.922 123.925
scf_env_do_scf_inner_loop 116 6.6 0.006 0.018 115.983 115.996
qs_scf_new_mos 116 7.6 0.001 0.001 80.362 80.671
qs_scf_loop_do_ot 116 8.6 0.001 0.001 80.361 80.670
dbcsr_multiply_generic 2485 12.5 0.189 0.193 78.986 79.592
ot_scf_mini 116 9.6 0.004 0.005 75.890 76.196
multiply_cannon 2485 13.5 0.548 0.575 54.195 57.491
multiply_cannon_loop 2485 14.5 1.170 1.199 50.601 52.166
init_scf_loop 11 6.9 0.001 0.003 47.088 47.109
prepare_preconditioner 11 7.9 0.000 0.000 42.862 42.888
make_preconditioner 11 8.9 0.000 0.001 42.862 42.888
ot_mini 116 10.6 0.001 0.001 42.017 42.320
make_full_inverse_cholesky 11 9.9 0.000 0.000 36.426 41.497
multiply_cannon_multrec 29820 15.5 14.158 18.797 25.954 30.358
rebuild_ks_matrix 127 8.3 0.001 0.001 29.166 29.467
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.020 29.166 29.466
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.298 26.565
mp_waitall_1 152434 16.5 17.734 26.547 17.734 26.547
qs_ot_get_derivative 116 11.6 0.001 0.002 22.710 23.015
make_m2s 4970 13.5 0.095 0.097 20.434 21.301
make_images 4970 14.5 1.936 2.240 20.138 21.006
qs_ot_get_p 127 10.4 0.002 0.012 20.291 20.596
apply_preconditioner_dbcsr 127 12.6 0.000 0.001 18.746 19.226
apply_single 127 13.6 0.001 0.001 18.746 19.226
ot_diis_step 116 11.6 0.017 0.018 19.169 19.171
cp_fm_upper_to_full 104 14.7 11.260 16.658 11.260 16.658
cp_fm_cholesky_invert 11 10.9 16.317 16.326 16.317 16.326
qs_ot_p2m_diag 82 11.4 0.338 0.384 16.055 16.106
multiply_cannon_metrocomm3 29820 15.5 0.046 0.049 6.235 14.996
cp_dbcsr_syevd 82 12.4 0.005 0.005 14.790 14.792
sum_up_and_integrate 127 10.3 0.140 0.150 14.194 14.232
integrate_v_rspace 127 11.3 0.003 0.004 14.054 14.096
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.205 13.236
calculate_rho_elec 127 8.7 0.173 0.188 13.205 13.236
dbcsr_complete_redistribute 393 12.7 1.504 1.616 9.121 12.919
make_images_data 4970 15.5 0.063 0.068 11.088 12.749
multiply_cannon_sync_h2d 29820 15.5 11.666 12.493 11.666 12.493
dbcsr_mm_accdrv_process 61748 16.2 7.337 8.283 11.381 11.902
cp_fm_diag_elpa 82 13.4 0.000 0.000 11.733 11.736
cp_fm_redistribute_end 82 14.4 2.016 11.661 2.034 11.667
hybrid_alltoall_any 5155 16.4 0.525 2.204 9.774 11.608
cp_fm_diag_elpa_base 82 14.4 8.992 11.049 9.593 11.531
copy_fm_to_dbcsr 208 11.6 0.001 0.002 7.699 11.498
init_scf_run 11 5.9 0.001 0.004 10.745 10.746
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.744 10.746
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.413 10.132
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.341 9.543
mp_alltoall_i22 712 14.1 5.591 9.390 5.591 9.390
pw_transfer 1535 11.6 0.085 0.098 8.044 8.125
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 7.821 7.903
cp_fm_cholesky_decompose 22 10.9 7.759 7.845 7.759 7.845
grid_integrate_task_list 127 12.3 7.445 7.833 7.445 7.833
wfi_extrapolate 11 7.9 0.001 0.001 7.618 7.618
fft_wrap_pw1pw2_140 519 13.2 0.477 0.488 6.882 6.978
density_rs2pw 127 9.7 0.005 0.006 6.331 6.848
multiply_cannon_metrocomm4 24850 15.5 0.075 0.085 2.698 6.541
mp_irecv_dv 75445 16.2 2.555 6.281 2.555 6.281
calculate_dm_sparse 127 9.5 0.001 0.001 6.132 6.228
fft3d_ps 1281 14.7 2.769 2.831 6.072 6.131
mp_alltoall_d11v 2401 14.1 5.058 5.647 5.058 5.647
grid_collocate_task_list 127 9.7 5.169 5.526 5.169 5.526
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.458 5.495
rs_pw_transfer 1038 11.9 0.013 0.014 4.643 5.146
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.422 4.529
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.372 4.445
potential_pw2rs 127 12.3 0.022 0.023 4.411 4.444
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 4.418 4.422
qs_ot_get_orbitals 116 10.6 0.001 0.001 4.144 4.209
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=187.706000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1071.454545, yerr=18.282449
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 5.820059E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 1944496 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3448.5
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.522504E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 238560
MPI messages size (bytes):
total size 1.321104E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.537828E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 112800 59139686400
4194304 < size <= 16777216 104112 545846722560
16777216 < size 20064 716108638608
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8852 52.
MP_Alltoall 9584 804353.
MP_ISend 39716 2104723.
MP_IRecv 39716 2103824.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3997 58279.
MP_Allreduce 10990 1175.
MP_Sync 86
MP_Alltoall 1700 18828160.
MP_ISendRecv 7620 122880.
MP_Wait 19810
MP_ISend 10600 423612.
MP_IRecv 10600 423612.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.060 0.065 170.671 170.671
qs_mol_dyn_low 1 2.0 0.003 0.003 170.086 170.099
qs_forces 11 3.9 0.003 0.003 169.976 169.983
qs_energies 11 4.9 0.001 0.002 162.665 162.672
scf_env_do_scf 11 5.9 0.001 0.001 145.616 145.625
velocity_verlet 10 3.0 0.001 0.001 111.231 111.235
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 109.621 109.622
qs_scf_new_mos 116 7.6 0.001 0.001 74.237 74.337
qs_scf_loop_do_ot 116 8.6 0.001 0.001 74.236 74.336
dbcsr_multiply_generic 2485 12.5 0.179 0.183 73.452 73.902
ot_scf_mini 116 9.6 0.003 0.004 69.823 69.871
multiply_cannon 2485 13.5 0.580 0.618 53.753 58.023
multiply_cannon_loop 2485 14.5 0.445 0.454 49.123 49.992
ot_mini 116 10.6 0.001 0.001 39.050 39.096
init_scf_loop 11 6.9 0.000 0.000 35.843 35.845
mp_waitall_1 128490 16.6 25.891 32.983 25.891 32.983
prepare_preconditioner 11 7.9 0.000 0.000 32.002 32.031
make_preconditioner 11 8.9 0.000 0.000 32.002 32.031
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.884 30.174
rebuild_ks_matrix 127 8.3 0.001 0.001 28.428 28.549
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.018 28.427 28.549
qs_ks_update_qs_env 127 7.6 0.001 0.001 25.919 26.036
multiply_cannon_multrec 9940 15.5 10.293 14.230 17.524 20.470
ot_diis_step 116 11.6 0.019 0.020 19.592 19.593
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.162 19.447
apply_single 127 13.6 0.001 0.001 19.162 19.446
qs_ot_get_derivative 116 11.6 0.002 0.002 19.391 19.443
multiply_cannon_metrocomm3 9940 15.5 0.022 0.023 12.104 19.336
cp_fm_cholesky_invert 11 10.9 18.291 18.297 18.291 18.297
make_m2s 4970 13.5 0.065 0.070 16.177 18.285
qs_ot_get_p 127 10.4 0.001 0.001 17.973 18.056
make_images 4970 14.5 2.294 2.837 15.874 17.977
qs_ot_p2m_diag 82 11.4 0.489 0.494 14.230 14.246
sum_up_and_integrate 127 10.3 0.179 0.189 13.999 14.049
integrate_v_rspace 127 11.3 0.004 0.004 13.820 13.877
cp_dbcsr_syevd 82 12.4 0.005 0.005 13.087 13.089
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.057 13.084
calculate_rho_elec 127 8.7 0.256 0.267 13.057 13.083
make_images_data 4970 15.5 0.052 0.060 9.872 12.503
multiply_cannon_sync_h2d 9940 15.5 11.530 12.100 11.530 12.100
hybrid_alltoall_any 5155 16.4 0.777 3.570 9.560 12.036
init_scf_run 11 5.9 0.000 0.001 10.306 10.306
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.305 10.306
cp_fm_diag_elpa 82 13.4 0.000 0.000 10.057 10.060
cp_fm_diag_elpa_base 82 14.4 9.801 9.887 10.049 10.052
cp_fm_cholesky_decompose 22 10.9 8.070 8.165 8.070 8.165
grid_integrate_task_list 127 12.3 7.702 7.981 7.702 7.981
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 7.795 7.830
pw_transfer 1535 11.6 0.083 0.091 7.810 7.825
multiply_cannon_metrocomm1 9940 15.5 0.029 0.030 4.380 7.648
fft_wrap_pw1pw2 1281 12.7 0.010 0.010 7.591 7.614
dbcsr_mm_accdrv_process 20590 16.1 2.696 3.456 6.865 7.470
wfi_extrapolate 11 7.9 0.001 0.001 7.299 7.299
mp_allgather_i34 2485 14.5 2.851 7.176 2.851 7.176
fft_wrap_pw1pw2_140 519 13.2 0.498 0.519 6.666 6.693
density_rs2pw 127 9.7 0.005 0.006 5.933 6.163
calculate_dm_sparse 127 9.5 0.001 0.001 6.068 6.146
fft3d_ps 1281 14.7 2.698 2.773 5.774 5.809
dbcsr_complete_redistribute 393 12.7 2.150 2.213 5.296 5.683
grid_collocate_task_list 127 9.7 5.446 5.641 5.446 5.641
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.247 5.283
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 5.206 5.206
mp_alltoall_d11v 2401 14.1 4.721 5.197 4.721 5.197
rs_pw_transfer 1038 11.9 0.013 0.013 4.031 4.272
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.091 4.127
multiply_cannon_metrocomm4 7455 15.5 0.023 0.026 1.851 4.119
potential_pw2rs 127 12.3 0.026 0.026 4.080 4.094
mp_irecv_dv 28618 15.9 1.813 4.044 1.813 4.044
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.576 3.845
copy_fm_to_dbcsr 208 11.6 0.002 0.002 3.460 3.791
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.639 3.663
copy_dbcsr_to_fm 185 11.7 0.004 0.004 3.566 3.637
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.435 3.457
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=170.671000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1426.909091, yerr=56.295413
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1389588037632 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1887638126592 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1929499951104 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1935094689792 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2675276685312 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4300336005120 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5255966228480 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5299016564736 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6476575801344 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11276331687936 0.0% 0.0% 100.0%
flops 22 x 32 x 9 14799245598720 0.0% 0.0% 100.0%
flops 9 x 32 x 22 14799245598720 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19338589388800 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 91.362404E+12 0.0% 0.0% 100.0%
flops max/rank 11.427044E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6604657536 0.0% 0.0% 100.0%
number of processed stacks 1915696 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3447.7
marketing flops 141.366701E+12
-------------------------------------------------------------------------------
# multiplications 2442
max memory usage/rank 2.966249E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 97680
MPI messages size (bytes):
total size 1.109909E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.362705E+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 43928 34091302912
4194304 < size <= 16777216 43248 370189271040
16777216 < size 9888 705625925088
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4013 58983.
MP_Allreduce 11030 1515.
MP_Sync 86
MP_Alltoall 1676 36913873.
MP_ISendRecv 3500 218624.
MP_Wait 11330
MP_ISend 6264 1080795.
MP_IRecv 6264 1080795.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.067 296.644 296.644
qs_mol_dyn_low 1 2.0 0.003 0.003 296.021 296.033
qs_forces 11 3.9 0.003 0.003 295.917 295.922
qs_energies 11 4.9 0.002 0.002 287.144 287.152
scf_env_do_scf 11 5.9 0.001 0.001 264.740 264.759
velocity_verlet 10 3.0 0.001 0.001 213.284 213.293
scf_env_do_scf_inner_loop 114 6.6 0.003 0.008 135.928 135.930
init_scf_loop 11 6.9 0.000 0.000 128.543 128.546
prepare_preconditioner 11 7.9 0.000 0.000 123.723 123.751
make_preconditioner 11 8.9 0.000 0.000 123.723 123.751
make_full_inverse_cholesky 11 9.9 0.000 0.000 99.045 120.896
qs_scf_new_mos 114 7.6 0.001 0.001 94.263 94.404
qs_scf_loop_do_ot 114 8.6 0.001 0.001 94.262 94.403
ot_scf_mini 114 9.6 0.004 0.004 89.479 89.566
dbcsr_multiply_generic 2442 12.5 0.208 0.215 84.574 85.433
cp_fm_upper_to_full 104 14.8 54.262 77.997 54.262 77.997
multiply_cannon 2442 13.5 0.684 0.740 59.471 60.321
multiply_cannon_loop 2442 14.5 0.460 0.466 55.653 57.548
ot_mini 114 10.6 0.001 0.001 45.531 45.643
dbcsr_complete_redistribute 393 12.7 3.997 4.024 30.553 44.574
copy_fm_to_dbcsr 208 11.6 0.001 0.002 27.055 41.088
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 24.632 38.479
mp_alltoall_i22 708 14.1 22.441 36.615 22.441 36.615
cp_fm_cholesky_invert 11 10.9 35.087 35.094 35.087 35.094
mp_waitall_1 102746 16.7 30.018 34.321 30.018 34.321
rebuild_ks_matrix 125 8.3 0.001 0.001 33.479 33.609
qs_ks_build_kohn_sham_matrix 125 9.3 0.017 0.018 33.478 33.608
qs_ks_update_qs_env 125 7.6 0.001 0.001 31.108 31.235
qs_ot_get_p 125 10.4 0.001 0.001 28.501 28.595
qs_ot_get_derivative 114 11.6 0.001 0.002 24.869 24.958
qs_ot_p2m_diag 82 11.4 0.867 0.872 24.314 24.349
cp_dbcsr_syevd 82 12.4 0.005 0.006 22.562 22.564
multiply_cannon_metrocomm3 9768 15.5 0.022 0.023 19.919 21.740
make_m2s 4884 13.5 0.074 0.076 20.212 21.291
make_images 4884 14.5 3.709 3.845 19.744 20.826
ot_diis_step 114 11.6 0.022 0.022 20.616 20.616
apply_preconditioner_dbcsr 125 12.6 0.000 0.000 19.866 20.232
apply_single 125 13.6 0.001 0.001 19.865 20.232
cp_fm_diag_elpa 82 13.4 0.000 0.000 19.252 19.259
cp_fm_diag_elpa_base 82 14.4 14.769 16.437 19.248 19.255
multiply_cannon_multrec 9768 15.5 10.272 12.059 17.836 17.944
sum_up_and_integrate 125 10.3 0.314 0.318 15.618 15.708
integrate_v_rspace 125 11.3 0.004 0.004 15.304 15.393
multiply_cannon_sync_h2d 9768 15.5 15.302 15.329 15.302 15.329
qs_rho_update_rho_low 125 7.7 0.001 0.001 14.938 14.962
calculate_rho_elec 125 8.7 0.472 0.473 14.938 14.961
hybrid_alltoall_any 5069 16.4 1.271 2.962 11.132 12.984
make_images_data 4884 15.5 0.058 0.062 11.035 12.912
init_scf_run 11 5.9 0.000 0.001 12.276 12.276
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.275 12.276
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.785 9.852
cp_fm_cholesky_decompose 22 10.9 9.511 9.600 9.511 9.600
dbcsr_mm_accdrv_process 20250 16.0 3.887 5.714 7.333 9.121
wfi_extrapolate 11 7.9 0.001 0.001 8.964 8.964
pw_transfer 1511 11.6 0.089 0.089 8.556 8.566
grid_integrate_task_list 125 12.3 8.357 8.523 8.357 8.523
fft_wrap_pw1pw2 1261 12.7 0.010 0.010 8.327 8.336
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 8.076 8.076
mp_alltoall_d11v 2385 14.1 7.184 7.329 7.184 7.329
fft_wrap_pw1pw2_140 511 13.2 0.527 0.531 7.306 7.314
calculate_dm_sparse 125 9.5 0.001 0.001 6.741 6.805
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.636 6.714
fft3d_ps 1261 14.7 2.698 2.715 6.420 6.431
copy_dbcsr_to_fm 185 11.7 0.004 0.004 6.194 6.276
grid_collocate_task_list 125 9.7 6.193 6.227 6.193 6.227
density_rs2pw 125 9.7 0.005 0.005 6.131 6.163
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=296.644000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2654.272727, yerr=162.032595
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 1.260605E+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_ISendRecv 570 19200.
MP_Wait 1302
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 0.019 0.033 84.422 84.423
qs_energies 1 2.0 0.000 0.000 84.004 84.011
ls_scf 1 3.0 0.000 0.000 83.091 83.098
dbcsr_multiply_generic 111 6.7 0.014 0.015 72.252 72.473
multiply_cannon 111 7.7 0.017 0.020 55.730 57.026
multiply_cannon_loop 111 8.7 0.211 0.224 52.282 53.562
ls_scf_main 1 4.0 0.000 0.000 52.048 52.049
density_matrix_trs4 2 5.0 0.002 0.003 46.820 46.869
ls_scf_init_scf 1 4.0 0.000 0.001 28.071 28.072
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.006 27.061
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 24.937 24.950
mp_waitall_1 11316 10.9 21.486 24.743 21.486 24.743
multiply_cannon_multrec 2664 9.7 8.233 9.004 15.551 17.285
multiply_cannon_sync_h2d 2664 9.7 14.251 15.972 14.251 15.972
make_m2s 222 7.7 0.009 0.011 13.022 13.510
make_images 222 8.7 0.099 0.110 13.000 13.489
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.020 11.793
make_images_data 222 9.7 0.004 0.005 7.587 8.176
hybrid_alltoall_any 227 10.6 0.216 1.831 6.512 8.021
dbcsr_mm_accdrv_process 4760 10.4 0.509 0.607 6.932 7.910
multiply_cannon_metrocomm3 2664 9.7 0.009 0.011 5.196 7.567
dbcsr_mm_accdrv_process_sort 4760 11.4 6.223 7.102 6.223 7.102
calculate_norms 4752 9.8 5.613 6.265 5.613 6.265
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.806 4.899
mp_sum_l 807 5.4 2.968 4.800 2.968 4.800
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.032 3.729
mp_irecv_dv 6231 10.9 2.014 3.700 2.014 3.700
make_images_sizes 222 9.7 0.000 0.000 0.689 3.551
mp_alltoall_i44 222 10.7 0.688 3.551 0.688 3.551
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.201 3.534
arnoldi_extremal 4 6.8 0.000 0.000 3.275 3.298
arnoldi_normal_ev 4 7.8 0.001 0.003 3.275 3.298
build_subspace 16 8.4 0.009 0.012 3.182 3.184
ls_scf_post 1 4.0 0.000 0.000 2.972 2.979
ls_scf_store_result 1 5.0 0.000 0.000 2.782 2.834
dbcsr_special_finalize 555 9.7 0.005 0.006 2.391 2.743
dbcsr_merge_single_wm 555 10.7 0.471 0.617 2.383 2.735
make_images_pack 222 9.7 2.209 2.632 2.211 2.634
dbcsr_matrix_vector_mult 304 9.0 0.003 0.010 2.326 2.565
dbcsr_sort_data 658 11.4 2.170 2.481 2.170 2.481
dbcsr_matrix_vector_mult_local 304 10.0 2.057 2.447 2.059 2.449
buffer_matrices_ensure_size 222 8.7 1.786 2.206 1.786 2.206
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.153 2.204
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.720 1.721
rebuild_ks_matrix 3 7.3 0.000 0.000 1.710 1.712
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.710 1.712
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.423000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1141.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 2.105696E+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_ISendRecv 282 57600.
MP_Wait 828
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 0.042 0.103 91.044 91.045
qs_energies 1 2.0 0.000 0.000 90.358 90.394
ls_scf 1 3.0 0.000 0.001 89.007 89.011
dbcsr_multiply_generic 111 6.7 0.015 0.016 75.054 75.397
multiply_cannon 111 7.7 0.028 0.038 53.146 56.741
ls_scf_main 1 4.0 0.000 0.001 54.542 54.548
multiply_cannon_loop 111 8.7 0.116 0.124 49.867 53.088
density_matrix_trs4 2 5.0 0.002 0.003 48.820 49.053
ls_scf_init_scf 1 4.0 0.001 0.003 30.847 30.849
mp_waitall_1 9246 10.9 21.061 30.298 21.061 30.298
ls_scf_init_matrix_S 1 5.0 0.000 0.001 29.668 29.787
multiply_cannon_multrec 1332 9.7 13.088 16.859 22.297 27.262
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.245 27.261
multiply_cannon_metrocomm3 1332 9.7 0.006 0.007 11.639 20.956
make_m2s 222 7.7 0.006 0.007 15.349 15.964
make_images 222 8.7 1.565 1.960 15.319 15.934
dbcsr_mm_accdrv_process 4041 10.4 0.262 0.449 8.814 10.366
make_images_data 222 9.7 0.004 0.005 8.789 9.932
dbcsr_mm_accdrv_process_sort 4041 11.4 8.408 9.917 8.408 9.917
hybrid_alltoall_any 227 10.6 0.523 2.491 8.233 9.244
mp_sum_l 807 5.4 5.311 8.429 5.311 8.429
multiply_cannon_metrocomm4 1221 9.7 0.006 0.008 3.254 7.764
mp_irecv_dv 3311 11.0 3.234 7.706 3.234 7.706
calculate_norms 2376 9.8 6.005 6.894 6.005 6.894
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.986 6.613
multiply_cannon_sync_h2d 1332 9.7 4.837 6.101 4.837 6.101
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.117 5.369
arnoldi_extremal 4 6.8 0.000 0.000 4.641 4.665
arnoldi_normal_ev 4 7.8 0.001 0.005 4.641 4.665
build_subspace 16 8.4 0.014 0.021 4.378 4.381
ls_scf_post 1 4.0 0.000 0.002 3.618 3.623
ls_scf_store_result 1 5.0 0.000 0.000 3.277 3.436
dbcsr_matrix_vector_mult 304 9.0 0.005 0.017 3.125 3.366
dbcsr_matrix_vector_mult_local 304 10.0 2.740 3.221 2.742 3.223
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.542 2.657
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 1.253 2.580
make_images_pack 222 9.7 2.032 2.435 2.034 2.437
mp_allgather_i34 111 8.7 0.983 2.295 0.983 2.295
dbcsr_sort_data 436 11.2 1.846 2.131 1.846 2.131
dbcsr_data_new 4174 10.1 1.604 1.846 1.604 1.846
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.837 1.839
rebuild_ks_matrix 3 7.3 0.000 0.000 1.824 1.826
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.824 1.826
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=91.045000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1744.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 2.713715E+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_ISendRecv 186 57600.
MP_Wait 732
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 0.085 0.161 99.820 99.822
qs_energies 1 2.0 0.000 0.001 99.001 99.008
ls_scf 1 3.0 0.000 0.001 97.585 97.594
dbcsr_multiply_generic 111 6.7 0.017 0.022 81.632 81.884
multiply_cannon 111 7.7 0.044 0.108 57.870 61.731
multiply_cannon_loop 111 8.7 0.100 0.112 54.272 58.224
ls_scf_main 1 4.0 0.000 0.006 57.283 57.287
density_matrix_trs4 2 5.0 0.002 0.003 51.310 51.519
ls_scf_init_scf 1 4.0 0.001 0.003 36.680 36.682
ls_scf_init_matrix_S 1 5.0 0.020 0.055 35.376 35.436
mp_waitall_1 7374 11.0 24.467 34.693 24.467 34.693
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 32.818 32.866
multiply_cannon_multrec 888 9.7 12.588 15.405 21.125 24.530
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.663 23.945
make_m2s 222 7.7 0.006 0.008 16.940 18.158
make_images 222 8.7 1.972 2.291 16.901 18.122
make_images_data 222 9.7 0.004 0.005 9.702 10.807
hybrid_alltoall_any 227 10.6 0.621 2.864 9.397 10.661
dbcsr_mm_accdrv_process 3754 10.4 0.253 0.439 8.048 9.250
mp_sum_l 807 5.4 5.288 8.842 5.288 8.842
dbcsr_mm_accdrv_process_sort 3754 11.4 7.671 8.811 7.671 8.811
multiply_cannon_sync_h2d 888 9.7 6.067 7.969 6.067 7.969
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 3.711 7.220
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.457 7.058
mp_irecv_dv 2335 11.1 2.442 7.006 2.442 7.006
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.952 6.813
compute_matrix_preconditioner 1 6.0 0.000 0.001 6.094 6.106
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.116 5.342
arnoldi_extremal 4 6.8 0.000 0.000 5.184 5.236
arnoldi_normal_ev 4 7.8 0.001 0.005 5.184 5.236
acc_transpose_blocks 888 9.7 0.004 0.005 4.551 5.010
acc_transpose_blocks_kernels 888 10.7 0.012 0.012 4.511 4.970
jit_kernel_transpose 1 13.0 4.500 4.958 4.500 4.958
build_subspace 16 8.4 0.014 0.021 4.856 4.861
calculate_norms 1584 9.8 4.271 4.649 4.271 4.649
mp_allgather_i34 111 8.7 1.434 3.789 1.434 3.789
dbcsr_matrix_vector_mult 304 9.0 0.005 0.016 3.448 3.772
ls_scf_post 1 4.0 0.002 0.007 3.622 3.631
dbcsr_matrix_vector_mult_local 304 10.0 3.025 3.608 3.027 3.610
ls_scf_store_result 1 5.0 0.000 0.000 3.352 3.449
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.777 2.892
make_images_pack 222 9.7 1.833 2.133 1.836 2.135
dbcsr_sort_data 325 11.1 1.903 2.065 1.903 2.065
make_images_sizes 222 9.7 0.000 0.000 0.880 2.020
mp_alltoall_i44 222 10.7 0.879 2.019 0.879 2.019
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=99.822000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2206.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 3.298767E+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_ISendRecv 138 86400.
MP_Wait 600
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 0.036 0.050 96.999 97.001
qs_energies 1 2.0 0.000 0.001 96.287 96.294
ls_scf 1 3.0 0.000 0.001 94.647 94.651
dbcsr_multiply_generic 111 6.7 0.016 0.017 78.551 78.805
ls_scf_main 1 4.0 0.000 0.003 58.869 58.871
multiply_cannon 111 7.7 0.056 0.147 51.613 56.137
density_matrix_trs4 2 5.0 0.002 0.003 52.842 52.962
multiply_cannon_loop 111 8.7 0.115 0.127 46.500 48.848
ls_scf_init_scf 1 4.0 0.001 0.002 32.544 32.545
ls_scf_init_matrix_S 1 5.0 0.000 0.001 31.393 31.467
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.933 28.944
mp_waitall_1 6438 11.0 22.815 28.836 22.815 28.836
multiply_cannon_multrec 1332 9.7 14.230 17.288 22.051 24.603
make_m2s 222 7.7 0.007 0.007 21.215 22.679
make_images 222 8.7 3.145 3.613 21.165 22.631
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.279 17.039
make_images_data 222 9.7 0.004 0.004 11.879 13.573
hybrid_alltoall_any 227 10.6 0.795 3.781 11.193 12.919
dbcsr_mm_accdrv_process 3641 10.4 0.201 0.398 7.467 8.996
dbcsr_mm_accdrv_process_sort 3641 11.4 7.097 8.585 7.097 8.585
mp_sum_l 807 5.4 4.139 7.726 4.139 7.726
multiply_cannon_sync_h2d 1332 9.7 5.498 6.143 5.498 6.143
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.062 6.108
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.193 6.082
mp_irecv_dv 3229 10.9 2.040 6.034 2.040 6.034
arnoldi_extremal 4 6.8 0.000 0.000 5.283 5.299
arnoldi_normal_ev 4 7.8 0.001 0.004 5.283 5.299
build_subspace 16 8.4 0.014 0.021 4.943 4.951
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.481 4.834
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.539 4.709
mp_allgather_i34 111 8.7 2.236 4.668 2.236 4.668
calculate_norms 2376 9.8 4.189 4.514 4.189 4.514
dbcsr_matrix_vector_mult 304 9.0 0.006 0.016 3.604 3.920
dbcsr_matrix_vector_mult_local 304 10.0 3.199 3.693 3.201 3.695
dbcsr_sort_data 658 11.4 3.064 3.439 3.064 3.439
ls_scf_post 1 4.0 0.000 0.001 3.233 3.240
dbcsr_special_finalize 555 9.7 0.006 0.007 2.799 3.206
dbcsr_merge_single_wm 555 10.7 0.532 0.660 2.791 3.198
ls_scf_store_result 1 5.0 0.000 0.000 2.979 3.036
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.904 2.949
dbcsr_data_release 10477 10.7 1.583 2.438 1.583 2.438
dbcsr_finalize 304 7.8 0.049 0.061 1.805 1.993
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.001000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2718.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 4.643697E+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_ISendRecv 90 115200.
MP_Wait 573
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 0.040 0.054 98.989 98.990
qs_energies 1 2.0 0.000 0.000 98.238 98.244
ls_scf 1 3.0 0.000 0.000 96.332 96.337
dbcsr_multiply_generic 111 6.7 0.017 0.018 78.044 78.207
ls_scf_main 1 4.0 0.000 0.000 62.055 62.055
multiply_cannon 111 7.7 0.084 0.176 55.537 60.962
density_matrix_trs4 2 5.0 0.002 0.003 55.036 55.177
multiply_cannon_loop 111 8.7 0.070 0.080 50.911 52.582
mp_waitall_1 5481 11.0 26.571 31.956 26.571 31.956
ls_scf_init_scf 1 4.0 0.000 0.001 30.658 30.663
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.453 29.486
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.270 27.277
multiply_cannon_multrec 444 9.7 14.130 16.448 21.142 24.382
make_m2s 222 7.7 0.004 0.005 17.627 20.219
make_images 222 8.7 3.718 4.420 17.565 20.159
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.349 16.453
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.123 14.825
make_images_data 222 9.7 0.003 0.004 9.890 12.402
hybrid_alltoall_any 227 10.6 0.787 3.744 9.732 12.255
multiply_cannon_sync_h2d 444 9.7 6.521 8.492 6.521 8.492
dbcsr_mm_accdrv_process 3003 10.4 0.188 0.337 6.717 7.858
dbcsr_mm_accdrv_process_sort 3003 11.4 6.402 7.513 6.402 7.513
mp_allgather_i34 111 8.7 2.858 7.039 2.858 7.039
arnoldi_extremal 4 6.8 0.000 0.000 5.846 5.853
arnoldi_normal_ev 4 7.8 0.002 0.005 5.846 5.853
build_subspace 16 8.4 0.015 0.020 5.452 5.460
mp_sum_l 807 5.4 2.915 4.987 2.915 4.987
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.607 4.786
dbcsr_matrix_vector_mult 304 9.0 0.007 0.017 4.185 4.366
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.597 4.319
mp_irecv_dv 1241 11.2 1.577 4.290 1.577 4.290
dbcsr_matrix_vector_mult_local 304 10.0 3.692 4.130 3.694 4.132
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.044 3.764
calculate_norms 792 9.8 3.534 3.670 3.534 3.670
ls_scf_post 1 4.0 0.000 0.000 3.619 3.625
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.405 3.515
make_images_sizes 222 9.7 0.000 0.000 1.096 3.466
mp_alltoall_i44 222 10.7 1.095 3.466 1.095 3.466
ls_scf_store_result 1 5.0 0.000 0.000 3.404 3.443
dbcsr_finalize 304 7.8 0.062 0.078 2.200 2.321
dbcsr_merge_all 275 8.9 0.472 0.526 2.048 2.155
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.989 1.990
dbcsr_data_release 10123 10.8 1.333 1.988 1.333 1.988
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=98.990000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3630.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/e64daf41ce02fb1dd7f72831acc068f4c092e916_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 8.782266E+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_ISendRecv 84 732600.
MP_Wait 309
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 0.078 0.097 107.650 107.651
qs_energies 1 2.0 0.000 0.000 106.222 106.237
ls_scf 1 3.0 0.000 0.000 103.333 103.347
dbcsr_multiply_generic 111 6.7 0.023 0.028 77.536 77.687
ls_scf_main 1 4.0 0.000 0.000 65.707 65.708
density_matrix_trs4 2 5.0 0.002 0.003 56.864 56.937
multiply_cannon 111 7.7 0.183 0.329 49.629 50.826
multiply_cannon_loop 111 8.7 0.068 0.073 46.177 47.459
ls_scf_init_scf 1 4.0 0.001 0.001 33.941 33.942
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.619 32.631
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.801 29.812
mp_waitall_1 4569 11.1 22.015 26.220 22.015 26.220
make_m2s 222 7.7 0.005 0.005 23.931 24.878
make_images 222 8.7 4.595 5.013 23.824 24.769
multiply_cannon_multrec 444 9.7 17.869 18.622 22.454 23.033
hybrid_alltoall_any 227 10.6 1.660 3.626 12.870 15.805
make_images_data 222 9.7 0.003 0.003 13.158 15.544
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.273 11.295
multiply_cannon_sync_h2d 444 9.7 8.850 8.895 8.850 8.895
arnoldi_extremal 4 6.8 0.000 0.000 7.349 7.365
arnoldi_normal_ev 4 7.8 0.003 0.009 7.349 7.365
build_subspace 16 8.4 0.026 0.036 6.776 6.785
dbcsr_matrix_vector_mult 304 9.0 0.009 0.025 5.448 5.592
dbcsr_matrix_vector_mult_local 304 10.0 5.035 5.329 5.037 5.332
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.019 5.271
ls_scf_dm_to_ks 2 5.0 0.000 0.000 4.802 4.896
dbcsr_mm_accdrv_process 1814 10.4 0.184 0.326 4.420 4.555
dbcsr_mm_accdrv_process_sort 1814 11.4 4.121 4.260 4.121 4.260
ls_scf_post 1 4.0 0.000 0.000 3.685 3.699
make_images_sizes 222 9.7 0.000 0.000 1.437 3.663
mp_alltoall_i44 222 10.7 1.436 3.663 1.436 3.663
ls_scf_store_result 1 5.0 0.000 0.000 3.436 3.449
mp_allgather_i34 111 8.7 1.074 3.385 1.074 3.385
calculate_norms 792 9.8 3.239 3.286 3.239 3.286
dbcsr_finalize 304 7.8 0.082 0.089 3.079 3.142
dbcsr_merge_all 275 8.9 0.890 0.916 2.864 2.921
dbcsr_complete_redistribute 5 7.6 1.438 1.478 2.766 2.906
mp_sum_l 807 5.4 1.649 2.896 1.649 2.896
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.859 2.859
dbcsr_data_release 12724 10.6 2.326 2.828 2.326 2.828
matrix_ls_to_qs 2 6.0 0.000 0.000 2.414 2.572
dbcsr_sort_data 325 11.1 2.440 2.521 2.440 2.521
dbcsr_new_transposed 4 7.5 0.243 0.251 2.326 2.355
dbcsr_frobenius_norm 74 6.6 2.054 2.135 2.193 2.240
dbcsr_add_d 103 6.2 0.000 0.000 2.131 2.206
dbcsr_add_anytype 103 7.2 0.860 0.894 2.131 2.205
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.187 2.189
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=107.651000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6972.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: e64daf41ce02fb1dd7f72831acc068f4c092e916
Summary: empty
Status: OK