=== 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: 7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.2, ELPA 2022.11.001.rc2, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.0.0, 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 (13.12.2022)
#
# \
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=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2022.11.001.rc2
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.0.0
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.0
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.3.2
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/01
job id: 43511705
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/02
job id: 43511706
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/03
job id: 43511707
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/04
job id: 43511708
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/05
job id: 43511709
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/06
job id: 43511710
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/07
job id: 43511713
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/08
job id: 43511715
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/09
job id: 43511716
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/10
job id: 43511718
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/11
job id: 43511721
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/12
job id: 43511723
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/13
job id: 43511725
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/14
job id: 43511728
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/15
job id: 43511730
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/16
job id: 43511731
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/17
job id: 43511745
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/18
job id: 43511746
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/19
job id: 43511747
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/20
job id: 43511748
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/21
job id: 43511749
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/22
job id: 43511750
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/23
job id: 43511751
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/24
job id: 43511752
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/25
job id: 43511753
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/26
job id: 43511754
--- 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/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 0.088 0.089 135.538 135.538
farming_run 1 2.0 134.613 134.615 135.402 135.420
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.463718E+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 485 2282278.
MP_Sync 27
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
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 0.010 0.024 115.854 115.855
qs_energies 1 2.0 0.000 0.000 115.571 115.572
mp2_main 1 3.0 0.000 0.000 113.291 113.293
mp2_gpw_main 1 4.0 0.030 0.037 112.307 112.309
mp2_ri_gpw_compute_in 1 5.0 0.188 0.206 93.386 93.571
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.005 55.350 55.535
mp2_eri_3c_integrate_gpw 272 7.0 0.153 0.169 41.647 46.693
get_2c_integrals 1 6.0 0.001 0.014 37.585 37.848
integrate_v_rspace 273 8.0 0.435 0.455 25.043 29.716
pw_transfer 6555 10.6 0.377 0.395 27.430 28.095
fft_wrap_pw1pw2 5465 11.4 0.045 0.048 26.072 26.582
grid_integrate_task_list 273 9.0 20.867 25.976 20.867 25.976
fft_wrap_pw1pw2_100 2178 12.4 1.228 1.473 23.633 24.165
compute_2c_integrals 1 7.0 0.013 0.014 19.890 19.890
compute_2c_integrals_loop_lm 1 8.0 0.024 0.024 19.029 19.525
mp2_eri_2c_integrate_gpw 1 9.0 2.374 2.447 19.005 19.501
rpa_ri_compute_en 1 5.0 0.011 0.013 18.804 18.881
cp_fm_cholesky_decompose 12 8.2 17.769 18.047 17.769 18.047
cholesky_decomp 1 7.0 0.000 0.000 16.537 16.810
fft3d_s 5443 13.4 16.163 16.446 16.185 16.468
ao_to_mo_and_store_B_mult_1 272 7.0 10.860 15.583 10.860 15.583
calculate_wavefunction 272 8.0 5.402 5.472 12.525 13.146
rpa_num_int 1 6.0 0.000 0.003 10.800 10.810
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.660 10.713
calc_mat_Q 8 8.0 0.000 0.000 9.389 9.510
contract_S_to_Q 8 9.0 0.000 0.000 8.811 8.929
calc_potential_gpw 544 9.5 0.005 0.006 8.302 8.680
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.286 8.626
parallel_gemm_fm 14 9.1 0.000 0.000 8.390 8.511
parallel_gemm_fm_cosma 14 10.1 8.390 8.511 8.390 8.511
potential_pw2rs 545 10.0 0.106 0.108 7.700 8.394
collocate_single_gaussian 272 10.0 0.040 0.043 7.518 7.766
create_integ_mat 1 6.0 0.014 0.027 7.711 7.720
array2fm 1 7.0 0.000 0.000 6.719 7.145
pw_scatter_s 2720 13.7 4.439 4.569 4.439 4.569
pw_gather_s 2722 13.2 3.873 4.235 3.873 4.235
array2fm_buffer_send 1 8.0 3.009 3.181 3.009 3.181
pw_poisson_solve 545 10.5 1.123 1.172 2.186 2.393
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=112.308869, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2735.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 0.059 0.083 404.395 404.402
farming_run 1 2.0 402.896 402.906 404.266 404.311
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.219482E+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 1821 23730.
MP_Sync 38
MP_Alltoall 77 5929977.
MP_SendRecv 2876 2171486.
MP_ISendRecv 1034 172620.
MP_Wait 1346
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.050 211.764 211.765
qs_energies 1 2.0 0.000 0.000 211.514 211.529
scf_env_do_scf 1 3.0 0.000 0.000 107.257 107.257
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.784 105.795
rebuild_ks_matrix 4 6.0 0.000 0.000 105.783 105.793
qs_ks_build_kohn_sham_matrix 4 7.0 0.056 0.067 105.783 105.793
hfx_ks_matrix 4 8.0 0.001 0.001 105.388 105.391
integrate_four_center 4 9.0 0.143 0.452 105.387 105.390
mp2_main 1 3.0 0.003 0.026 103.966 103.981
mp2_gpw_main 1 4.0 0.038 0.074 102.461 102.481
integrate_four_center_main 4 10.0 0.106 0.502 96.740 99.396
integrate_four_center_bin 262 11.0 96.634 99.357 96.634 99.357
init_scf_loop 1 4.0 0.000 0.000 92.478 92.478
mp2_ri_gpw_compute_in 1 5.0 0.070 0.113 74.873 76.001
mp2_ri_gpw_compute_in_loop 1 6.0 0.003 0.039 54.545 55.670
mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.163 42.261 47.172
integrate_v_rspace 95 8.0 0.397 0.563 28.641 33.344
pw_transfer 2240 10.6 0.146 0.162 29.938 30.465
fft_wrap_pw1pw2 1868 11.4 0.018 0.020 28.955 29.516
mp2_ri_gpw_compute_en 1 5.0 0.086 0.115 27.381 29.156
grid_integrate_task_list 95 9.0 23.910 28.834 23.910 28.834
fft_wrap_pw1pw2_100 730 12.4 1.263 1.416 26.647 27.059
ao_to_mo_and_store_B_mult_1 91 7.0 10.600 25.868 10.600 25.868
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.856 1.910 25.539 25.550
get_2c_integrals 1 6.0 0.001 0.009 20.208 20.265
compute_2c_integrals 1 7.0 0.011 0.041 19.187 19.197
compute_2c_integrals_loop_lm 1 8.0 0.003 0.007 18.848 19.033
mp2_eri_2c_integrate_gpw 1 9.0 1.738 1.875 18.845 19.032
fft3d_s 1823 13.4 18.410 18.831 18.423 18.844
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.775 14.775
calculate_wavefunction 91 8.0 2.028 2.061 9.751 9.983
potential_pw2rs 186 10.0 0.033 0.035 8.633 9.231
mp2_ri_gpw_compute_en_expansio 172 7.0 0.554 0.575 8.758 9.078
mp2_ri_gpw_compute_en_comm 22 7.0 0.493 0.507 8.532 8.831
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.227 8.620
local_gemm 172 8.0 8.204 8.514 8.204 8.514
calc_potential_gpw 182 9.5 0.002 0.002 7.922 8.164
collocate_single_gaussian 91 10.0 0.017 0.021 7.873 8.126
mp_sendrecv_dm3 2068 8.0 6.590 6.881 6.590 6.881
mp2_ri_gpw_compute_en_ener 172 7.0 6.348 6.448 6.348 6.448
pw_gather_s 912 13.2 4.928 5.431 4.928 5.431
mp_sync 38 10.4 3.306 5.356 3.306 5.356
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=102.436254, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1506.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 451.588096E+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 465423.
MP_SendRecv 22610 5520.
MP_ISendRecv 22610 5520.
MP_Wait 37876
MP_comm_split 50
MP_ISend 20771 42672.
MP_IRecv 20771 42672.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.093 0.200 59.120 59.122
qs_mol_dyn_low 1 2.0 0.003 0.005 58.551 58.558
qs_forces 11 3.9 0.002 0.004 58.487 58.488
qs_energies 11 4.9 0.002 0.009 56.820 56.838
scf_env_do_scf 11 5.9 0.002 0.011 49.442 49.442
scf_env_do_scf_inner_loop 108 6.5 0.003 0.011 46.726 46.727
dbcsr_multiply_generic 2286 12.5 0.094 0.102 34.950 35.403
qs_scf_new_mos 108 7.5 0.000 0.001 34.456 34.707
qs_scf_loop_do_ot 108 8.5 0.000 0.001 34.456 34.706
ot_scf_mini 108 9.5 0.002 0.003 32.714 32.892
velocity_verlet 10 3.0 0.002 0.002 28.851 28.852
multiply_cannon 2286 13.5 0.194 0.207 26.179 27.707
multiply_cannon_loop 2286 14.5 1.489 1.564 25.070 26.588
ot_mini 108 10.5 0.001 0.001 20.335 20.566
qs_ot_get_derivative 108 11.5 0.001 0.001 17.318 17.501
mp_waitall_1 245248 16.5 9.226 15.165 9.226 15.165
multiply_cannon_metrocomm3 54864 15.5 0.066 0.072 5.818 13.152
multiply_cannon_multrec 54864 15.5 4.191 6.558 7.459 10.788
rebuild_ks_matrix 119 8.3 0.000 0.000 9.137 9.251
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.013 9.136 9.251
qs_ks_update_qs_env 119 7.6 0.001 0.001 8.035 8.142
qs_ot_get_p 119 10.4 0.003 0.018 7.642 7.928
mp_sum_l 7207 12.9 5.649 7.334 5.649 7.334
multiply_cannon_sync_h2d 54864 15.5 5.789 7.118 5.789 7.118
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.778 6.288
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.994 6.097
qs_rho_update_rho_low 119 7.7 0.000 0.001 5.718 5.833
calculate_rho_elec 119 8.7 0.011 0.016 5.718 5.832
init_scf_run 11 5.9 0.000 0.001 5.808 5.809
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.808 5.809
sum_up_and_integrate 119 10.3 0.012 0.015 5.461 5.480
integrate_v_rspace 119 11.3 0.002 0.002 5.449 5.469
dbcsr_mm_accdrv_process 76910 16.1 1.168 1.781 3.190 4.459
rs_pw_transfer 974 11.9 0.011 0.012 4.179 4.331
qs_ot_p2m_diag 50 11.0 0.004 0.007 4.123 4.174
density_rs2pw 119 9.7 0.004 0.004 3.591 3.736
multiply_cannon_metrocomm1 54864 15.5 0.051 0.056 2.020 3.390
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.041 3.241
apply_single 119 13.6 0.000 0.000 3.041 3.241
cp_dbcsr_syevd 50 12.0 0.002 0.003 3.225 3.226
calculate_dm_sparse 119 9.5 0.000 0.001 3.011 3.130
make_m2s 4572 13.5 0.055 0.057 2.875 2.949
wfi_extrapolate 11 7.9 0.001 0.001 2.896 2.896
make_images 4572 14.5 0.134 0.140 2.790 2.863
pw_transfer 1439 11.6 0.051 0.056 2.712 2.838
calculate_first_density_matrix 1 7.0 0.000 0.003 2.801 2.804
fft_wrap_pw1pw2 1201 12.6 0.006 0.007 2.637 2.765
ot_diis_step 108 11.5 0.006 0.006 2.750 2.752
potential_pw2rs 119 12.3 0.004 0.004 2.697 2.742
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.723 2.725
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.707 2.710
cp_fm_redistribute_end 50 14.0 2.461 2.678 2.468 2.682
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.584 2.654
init_scf_loop 11 6.9 0.001 0.004 2.636 2.640
jit_kernel_multiply 13 15.8 1.959 2.619 1.959 2.619
cp_fm_diag_elpa_base 50 14.0 0.211 2.547 0.212 2.556
fft3d_ps 1201 14.6 0.362 0.467 2.420 2.544
mp_alltoall_d11v 2130 13.8 2.145 2.306 2.145 2.306
acc_transpose_blocks 54864 15.5 0.230 0.250 1.802 2.288
mp_sum_d 4125 12.0 1.575 2.234 1.575 2.234
fft_wrap_pw1pw2_140 487 13.2 0.076 0.091 2.052 2.180
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.121 2.175
grid_integrate_task_list 119 12.3 2.007 2.117 2.007 2.117
make_images_sizes 4572 15.5 0.004 0.005 1.446 1.764
mp_alltoall_i44 4572 16.5 1.442 1.760 1.442 1.760
mp_waitany 12084 13.8 1.431 1.612 1.431 1.612
mp_alltoall_z22v 1201 16.6 1.507 1.606 1.507 1.606
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.452 1.473
grid_collocate_task_list 119 9.7 1.290 1.374 1.290 1.374
arnoldi_extremal 119 11.4 0.002 0.009 1.153 1.287
arnoldi_normal_ev 119 12.4 0.006 0.035 1.151 1.286
parallel_gemm_fm 81 9.0 0.000 0.000 1.260 1.265
parallel_gemm_fm_cosma 81 10.0 1.260 1.265 1.260 1.265
prepare_preconditioner 11 7.9 0.000 0.000 1.169 1.197
make_preconditioner 11 8.9 0.000 0.002 1.169 1.197
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=59.122000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=430.454545, yerr=0.782030
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 487.583744E+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 1063860.
MP_SendRecv 16779 37093.
MP_ISendRecv 16779 37093.
MP_Wait 23539
MP_comm_split 50
MP_ISend 5720 128509.
MP_IRecv 5720 128509.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.039 43.095 43.097
qs_mol_dyn_low 1 2.0 0.003 0.004 42.756 42.764
qs_forces 11 3.9 0.002 0.004 42.689 42.690
qs_energies 11 4.9 0.002 0.008 40.860 40.864
scf_env_do_scf 11 5.9 0.003 0.023 34.940 34.941
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 31.978 31.979
dbcsr_multiply_generic 2286 12.5 0.100 0.103 23.081 23.538
qs_scf_new_mos 108 7.5 0.001 0.001 22.171 22.411
qs_scf_loop_do_ot 108 8.5 0.001 0.001 22.171 22.411
ot_scf_mini 108 9.5 0.003 0.004 21.182 21.358
velocity_verlet 10 3.0 0.002 0.002 20.149 20.150
multiply_cannon 2286 13.5 0.211 0.219 17.069 19.017
multiply_cannon_loop 2286 14.5 0.900 0.970 15.690 17.552
ot_mini 108 10.5 0.001 0.001 12.902 13.150
mp_waitall_1 200699 16.5 6.639 11.855 6.639 11.855
qs_ot_get_derivative 108 11.5 0.001 0.001 10.255 10.431
multiply_cannon_metrocomm3 27432 15.5 0.068 0.070 4.586 9.879
multiply_cannon_multrec 27432 15.5 1.945 4.350 5.950 8.740
rebuild_ks_matrix 119 8.3 0.000 0.000 7.654 7.849
qs_ks_build_kohn_sham_matrix 119 9.3 0.026 0.071 7.653 7.849
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.728 6.904
dbcsr_mm_accdrv_process 47894 16.0 3.087 5.380 3.935 5.870
qs_ot_get_p 119 10.4 0.001 0.001 5.146 5.371
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.860 4.758
sum_up_and_integrate 119 10.3 0.024 0.027 4.591 4.596
integrate_v_rspace 119 11.3 0.002 0.002 4.566 4.574
mp_sum_l 7207 12.9 2.483 4.464 2.483 4.464
init_scf_run 11 5.9 0.000 0.001 4.449 4.450
scf_env_initial_rho_setup 11 6.9 0.002 0.004 4.449 4.450
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.208 4.269
apply_single 119 13.6 0.000 0.000 3.208 4.269
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.131 4.167
calculate_rho_elec 119 8.7 0.021 0.024 4.131 4.167
make_m2s 4572 13.5 0.054 0.056 3.091 3.388
qs_ot_p2m_diag 50 11.0 0.009 0.013 3.323 3.344
rs_pw_transfer 974 11.9 0.010 0.011 3.261 3.340
make_images 4572 14.5 0.201 0.238 3.001 3.295
init_scf_loop 11 6.9 0.001 0.006 2.908 2.908
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.838 2.838
calculate_first_density_matrix 1 7.0 0.001 0.004 2.759 2.762
multiply_cannon_sync_h2d 27432 15.5 2.134 2.735 2.134 2.735
ot_diis_step 108 11.5 0.010 0.011 2.578 2.579
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.481 2.569
density_rs2pw 119 9.7 0.004 0.004 2.459 2.529
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.387 2.388
cp_fm_redistribute_end 50 14.0 1.964 2.345 1.970 2.347
potential_pw2rs 119 12.3 0.006 0.006 2.294 2.305
cp_fm_diag_elpa_base 50 14.0 0.362 2.211 0.375 2.267
calculate_dm_sparse 119 9.5 0.000 0.001 2.174 2.260
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.208 2.211
pw_transfer 1439 11.6 0.063 0.067 2.017 2.049
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.928 1.973
jit_kernel_multiply 10 16.1 0.795 1.967 0.795 1.967
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 1.928 1.961
grid_integrate_task_list 119 12.3 1.825 1.941 1.825 1.941
prepare_preconditioner 11 7.9 0.000 0.000 1.897 1.936
make_preconditioner 11 8.9 0.007 0.037 1.897 1.936
make_images_data 4572 15.5 0.045 0.051 1.334 1.863
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.776 1.855
hybrid_alltoall_any 4725 16.4 0.051 0.112 1.157 1.702
fft3d_ps 1201 14.6 0.508 0.563 1.636 1.665
mp_allgather_i34 2286 14.5 0.819 1.660 0.819 1.660
mp_alltoall_d11v 2130 13.8 1.474 1.651 1.474 1.651
wfi_extrapolate 11 7.9 0.001 0.001 1.617 1.617
acc_transpose_blocks 27432 15.5 0.108 0.113 1.186 1.548
fft_wrap_pw1pw2_140 487 13.2 0.075 0.081 1.353 1.384
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.325 1.369
grid_collocate_task_list 119 9.7 1.242 1.342 1.242 1.342
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.304 1.314
make_images_sizes 4572 15.5 0.005 0.005 0.920 1.260
mp_alltoall_i44 4572 16.5 0.916 1.256 0.916 1.256
mp_sum_d 4125 12.0 0.804 1.254 0.804 1.254
qs_energies_init_hamiltonians 11 5.9 0.022 0.082 1.135 1.137
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.989 1.007
acc_transpose_blocks_kernels 27432 16.5 0.180 0.270 0.658 0.934
rs_pw_transfer_PW2RS_50 119 14.3 0.584 0.603 0.888 0.923
mp_alltoall_z22v 1201 16.6 0.831 0.901 0.831 0.901
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=43.097000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=463.636364, yerr=1.966664
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 523.100160E+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_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
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 0.165 0.427 37.288 37.292
qs_mol_dyn_low 1 2.0 0.003 0.005 36.319 36.328
qs_forces 11 3.9 0.021 0.063 36.118 36.119
qs_energies 11 4.9 0.002 0.007 34.333 34.356
scf_env_do_scf 11 5.9 0.001 0.004 29.108 29.108
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 25.491 25.492
dbcsr_multiply_generic 2286 12.5 0.094 0.098 17.899 18.141
velocity_verlet 10 3.0 0.001 0.002 17.901 17.907
qs_scf_new_mos 108 7.5 0.001 0.001 16.815 16.833
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.815 16.833
ot_scf_mini 108 9.5 0.004 0.017 15.999 16.014
multiply_cannon 2286 13.5 0.197 0.205 13.654 14.612
multiply_cannon_loop 2286 14.5 0.636 0.665 12.671 13.765
ot_mini 108 10.5 0.001 0.001 9.825 9.853
qs_ot_get_derivative 108 11.5 0.001 0.001 8.222 8.240
multiply_cannon_multrec 18288 15.5 1.929 2.890 6.754 7.066
rebuild_ks_matrix 119 8.3 0.000 0.000 6.903 6.945
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.015 6.902 6.945
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.996 6.034
dbcsr_mm_accdrv_process 38222 16.0 4.107 5.369 4.741 5.576
mp_waitall_1 158411 16.6 3.692 5.112 3.692 5.112
sum_up_and_integrate 119 10.3 0.030 0.031 4.411 4.422
integrate_v_rspace 119 11.3 0.002 0.003 4.380 4.395
init_scf_run 11 5.9 0.000 0.001 3.964 3.965
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.964 3.964
qs_ot_get_p 119 10.4 0.001 0.001 3.732 3.755
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.082 3.711
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.641 3.662
calculate_rho_elec 119 8.7 0.030 0.031 3.640 3.661
init_scf_loop 11 6.9 0.001 0.005 3.595 3.596
rs_pw_transfer 974 11.9 0.009 0.010 2.872 3.034
multiply_cannon_metrocomm3 18288 15.5 0.045 0.046 1.759 2.974
prepare_preconditioner 11 7.9 0.000 0.000 2.674 2.677
make_preconditioner 11 8.9 0.000 0.001 2.674 2.677
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.471 2.575
calculate_first_density_matrix 1 7.0 0.001 0.004 2.470 2.471
make_m2s 4572 13.5 0.046 0.047 2.293 2.459
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.094 2.428
apply_single 119 13.6 0.000 0.000 2.094 2.428
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.372 2.382
make_images 4572 14.5 0.193 0.205 2.206 2.372
density_rs2pw 119 9.7 0.004 0.004 2.171 2.293
mp_sum_l 7207 12.9 1.726 2.230 1.726 2.230
potential_pw2rs 119 12.3 0.007 0.007 2.154 2.177
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.051 2.052
pw_transfer 1439 11.6 0.064 0.066 1.981 2.017
calculate_dm_sparse 119 9.5 0.001 0.002 1.945 1.956
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.890 1.928
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.915 1.923
jit_kernel_multiply 10 16.1 0.582 1.898 0.582 1.898
grid_integrate_task_list 119 12.3 1.796 1.873 1.796 1.873
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.799 1.802
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.691 1.695
cp_fm_redistribute_end 50 14.0 1.256 1.659 1.258 1.660
fft3d_ps 1201 14.6 0.508 0.527 1.578 1.616
cp_fm_diag_elpa_base 50 14.0 0.383 1.555 0.397 1.597
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.551 1.576
ot_diis_step 108 11.5 0.011 0.011 1.561 1.562
multiply_cannon_sync_h2d 18288 15.5 1.339 1.535 1.339 1.535
wfi_extrapolate 11 7.9 0.001 0.001 1.446 1.446
fft_wrap_pw1pw2_140 487 13.2 0.086 0.089 1.390 1.422
grid_collocate_task_list 119 9.7 1.204 1.285 1.204 1.285
cp_fm_cholesky_invert 11 10.9 1.259 1.268 1.259 1.268
make_images_data 4572 15.5 0.045 0.049 1.053 1.265
acc_transpose_blocks 18288 15.5 0.075 0.077 1.241 1.263
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.224 1.233
mp_alltoall_d11v 2130 13.8 1.035 1.167 1.035 1.167
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.121 1.156
hybrid_alltoall_any 4725 16.4 0.055 0.113 0.915 1.092
multiply_cannon_metrocomm1 18288 15.5 0.028 0.029 0.540 1.069
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.009 1.030
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.985 0.995
mp_alltoall_z22v 1201 16.6 0.885 0.987 0.885 0.987
dbcsr_complete_redistribute 329 12.2 0.090 0.097 0.804 0.931
mp_allgather_i34 2286 14.5 0.439 0.916 0.439 0.916
acc_transpose_blocks_kernels 18288 16.5 0.209 0.219 0.801 0.811
make_images_sizes 4572 15.5 0.005 0.005 0.550 0.762
mp_alltoall_i44 4572 16.5 0.545 0.757 0.545 0.757
mp_waitany 9880 13.7 0.607 0.751 0.607 0.751
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=37.292000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.545455, yerr=2.387986
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 549.756928E+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_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
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 0.108 0.181 37.477 37.495
qs_mol_dyn_low 1 2.0 0.005 0.048 36.626 36.633
qs_forces 11 3.9 0.018 0.035 36.564 36.568
qs_energies 11 4.9 0.034 0.115 34.785 34.799
scf_env_do_scf 11 5.9 0.011 0.035 29.078 29.081
scf_env_do_scf_inner_loop 108 6.5 0.003 0.008 25.419 25.420
dbcsr_multiply_generic 2286 12.5 0.099 0.101 18.713 18.836
velocity_verlet 10 3.0 0.002 0.002 18.311 18.313
qs_scf_new_mos 108 7.5 0.001 0.001 16.830 16.889
qs_scf_loop_do_ot 108 8.5 0.001 0.001 16.830 16.888
ot_scf_mini 108 9.5 0.003 0.004 15.879 15.931
multiply_cannon 2286 13.5 0.236 0.281 14.722 15.158
multiply_cannon_loop 2286 14.5 0.936 0.972 13.702 14.062
ot_mini 108 10.5 0.001 0.001 9.797 9.855
multiply_cannon_multrec 27432 15.5 2.417 3.102 8.737 9.149
qs_ot_get_derivative 108 11.5 0.001 0.001 7.963 8.019
dbcsr_mm_accdrv_process 47916 15.9 5.314 7.415 6.225 7.550
rebuild_ks_matrix 119 8.3 0.000 0.000 6.764 6.815
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.016 6.764 6.814
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.020 6.064
init_scf_run 11 5.9 0.000 0.001 4.034 4.043
scf_env_initial_rho_setup 11 6.9 0.025 0.039 4.034 4.043
sum_up_and_integrate 119 10.3 0.035 0.037 3.978 3.986
integrate_v_rspace 119 11.3 0.002 0.003 3.943 3.952
init_scf_loop 11 6.9 0.003 0.013 3.611 3.613
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.515 3.548
calculate_rho_elec 119 8.7 0.040 0.046 3.515 3.548
qs_ot_get_p 119 10.4 0.001 0.001 3.442 3.511
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.898 3.347
mp_waitall_1 137007 16.6 2.242 2.834 2.242 2.834
prepare_preconditioner 11 7.9 0.000 0.000 2.700 2.708
make_preconditioner 11 8.9 0.003 0.014 2.700 2.708
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.303 2.630
calculate_first_density_matrix 1 7.0 0.005 0.028 2.581 2.614
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.117 2.563
apply_single 119 13.6 0.000 0.000 2.116 2.563
rs_pw_transfer 974 11.9 0.009 0.009 2.373 2.520
make_m2s 4572 13.5 0.055 0.057 2.409 2.490
make_images 4572 14.5 0.274 0.334 2.300 2.380
calculate_dm_sparse 119 9.5 0.000 0.000 2.042 2.093
density_rs2pw 119 9.7 0.004 0.004 1.968 2.073
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.051 2.062
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.025 2.053
jit_kernel_multiply 10 15.7 0.850 1.916 0.850 1.916
pw_transfer 1439 11.6 0.063 0.066 1.872 1.906
grid_integrate_task_list 119 12.3 1.821 1.890 1.821 1.890
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.851 1.852
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.782 1.818
mp_sum_l 7207 12.9 1.223 1.805 1.223 1.805
ot_diis_step 108 11.5 0.012 0.012 1.790 1.790
potential_pw2rs 119 12.3 0.008 0.009 1.748 1.759
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.724 1.725
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.643 1.654
fft3d_ps 1201 14.6 0.536 0.585 1.464 1.490
acc_transpose_blocks 27432 15.5 0.112 0.115 1.456 1.487
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.845 1.457
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.453 1.455
fft_wrap_pw1pw2_140 487 13.2 0.084 0.092 1.414 1.453
cp_fm_redistribute_end 50 14.0 0.962 1.426 0.964 1.427
cp_fm_diag_elpa_base 50 14.0 0.441 1.349 0.460 1.387
wfi_extrapolate 11 7.9 0.001 0.001 1.379 1.379
grid_collocate_task_list 119 9.7 1.222 1.290 1.222 1.290
qs_energies_init_hamiltonians 11 5.9 0.003 0.010 1.238 1.257
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.211 1.221
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.142 1.157
cp_fm_upper_to_full 72 13.5 0.805 1.114 0.805 1.114
dbcsr_complete_redistribute 329 12.2 0.127 0.156 0.833 1.107
make_images_data 4572 15.5 0.045 0.049 0.980 1.090
multiply_cannon_sync_h2d 27432 15.5 0.980 1.079 0.980 1.079
hybrid_alltoall_any 4725 16.4 0.061 0.151 0.847 0.996
mp_alltoall_d11v 2130 13.8 0.849 0.925 0.849 0.925
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.795 0.874
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.602 0.869
cp_fm_cholesky_invert 11 10.9 0.853 0.856 0.853 0.856
acc_transpose_blocks_kernels 27432 16.5 0.266 0.274 0.833 0.848
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.837 0.842
mp_alltoall_z22v 1201 16.6 0.810 0.841 0.810 0.841
mp_alltoall_i22 627 13.8 0.466 0.770 0.466 0.770
mp_sum_d 4123 12.0 0.576 0.758 0.576 0.758
qs_env_update_s_mstruct 11 6.9 0.001 0.009 0.705 0.755
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=37.495000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=522.545455, yerr=2.641062
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 608.665600E+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_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
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 0.039 0.058 29.244 29.245
qs_mol_dyn_low 1 2.0 0.003 0.003 28.977 28.984
qs_forces 11 3.9 0.004 0.005 28.897 28.898
qs_energies 11 4.9 0.001 0.001 27.174 27.177
scf_env_do_scf 11 5.9 0.000 0.001 22.262 22.263
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.591 19.592
velocity_verlet 10 3.0 0.009 0.011 14.954 14.958
dbcsr_multiply_generic 2286 12.5 0.091 0.093 12.581 12.662
qs_scf_new_mos 108 7.5 0.001 0.001 11.681 11.706
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.680 11.705
ot_scf_mini 108 9.5 0.002 0.002 10.972 11.003
multiply_cannon 2286 13.5 0.231 0.239 9.715 10.236
multiply_cannon_loop 2286 14.5 0.329 0.342 8.737 8.955
ot_mini 108 10.5 0.001 0.001 5.993 6.032
rebuild_ks_matrix 119 8.3 0.000 0.000 5.995 6.021
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.995 6.021
multiply_cannon_multrec 9144 15.5 1.586 1.822 5.757 5.982
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.359 5.382
qs_ot_get_derivative 108 11.5 0.001 0.001 4.669 4.703
dbcsr_mm_accdrv_process 12550 15.8 3.053 3.601 4.068 4.156
sum_up_and_integrate 119 10.3 0.037 0.041 3.695 3.699
integrate_v_rspace 119 11.3 0.002 0.003 3.658 3.662
init_scf_run 11 5.9 0.000 0.001 3.468 3.468
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.468 3.468
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.427 3.437
calculate_rho_elec 119 8.7 0.059 0.061 3.426 3.436
qs_ot_get_p 119 10.4 0.001 0.001 3.118 3.158
init_scf_loop 11 6.9 0.000 0.000 2.647 2.648
mp_waitall_1 115863 16.7 1.821 2.465 1.821 2.465
calculate_first_density_matrix 1 7.0 0.000 0.000 2.218 2.219
make_m2s 4572 13.5 0.036 0.036 1.972 2.127
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.070 2.072
make_images 4572 14.5 0.270 0.308 1.881 2.034
rs_pw_transfer 974 11.9 0.008 0.008 1.907 1.989
grid_integrate_task_list 119 12.3 1.845 1.927 1.845 1.927
prepare_preconditioner 11 7.9 0.000 0.000 1.887 1.892
make_preconditioner 11 8.9 0.000 0.000 1.887 1.892
density_rs2pw 119 9.7 0.003 0.004 1.817 1.891
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.853 1.853
pw_transfer 1439 11.6 0.064 0.065 1.818 1.826
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.771 1.806
calculate_dm_sparse 119 9.5 0.000 0.000 1.763 1.782
jit_kernel_multiply 10 15.8 0.976 1.777 0.976 1.777
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.727 1.735
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.613 1.631
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.558 1.561
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.507 1.508
cp_fm_redistribute_end 50 14.0 0.752 1.480 0.753 1.481
potential_pw2rs 119 12.3 0.010 0.010 1.469 1.474
cp_fm_diag_elpa_base 50 14.0 0.683 1.406 0.726 1.460
fft3d_ps 1201 14.6 0.543 0.555 1.395 1.403
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.373 1.382
grid_collocate_task_list 119 9.7 1.271 1.377 1.271 1.377
fft_wrap_pw1pw2_140 487 13.2 0.082 0.085 1.365 1.374
ot_diis_step 108 11.5 0.012 0.013 1.308 1.309
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.245 1.261
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.226 1.228
wfi_extrapolate 11 7.9 0.001 0.001 1.201 1.202
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.180 1.202
apply_single 119 13.6 0.000 0.000 1.179 1.201
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.153 1.158
hybrid_alltoall_any 4725 16.4 0.062 0.176 0.889 1.142
make_images_data 4572 15.5 0.039 0.043 0.928 1.131
mp_alltoall_d11v 2130 13.8 0.907 1.035 0.907 1.035
cp_fm_cholesky_invert 11 10.9 0.966 0.970 0.966 0.970
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.869 0.922
acc_transpose_blocks 9144 15.5 0.038 0.039 0.888 0.895
multiply_cannon_sync_h2d 9144 15.5 0.712 0.799 0.712 0.799
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.776 0.786
mp_allgather_i34 2286 14.5 0.313 0.763 0.313 0.763
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.759 0.763
mp_alltoall_z22v 1201 16.6 0.726 0.761 0.726 0.761
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.679 0.721
mp_sum_l 7207 12.9 0.494 0.689 0.494 0.689
multiply_cannon_metrocomm3 9144 15.5 0.019 0.019 0.346 0.685
acc_transpose_blocks_kernels 9144 16.5 0.115 0.118 0.649 0.652
dbcsr_complete_redistribute 329 12.2 0.160 0.170 0.594 0.626
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=29.245000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=573.909091, yerr=8.050461
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 753.713152E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63729.
MP_Allreduce 10074 433.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.127 0.143 45.488 45.489
qs_mol_dyn_low 1 2.0 0.003 0.003 44.537 44.546
qs_forces 11 3.9 0.002 0.002 44.473 44.474
qs_energies 11 4.9 0.001 0.001 42.410 42.414
scf_env_do_scf 11 5.9 0.001 0.001 35.618 35.618
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 27.360 27.362
velocity_verlet 10 3.0 0.002 0.002 24.739 24.745
dbcsr_multiply_generic 2286 12.5 0.107 0.112 19.135 19.421
qs_scf_new_mos 108 7.5 0.001 0.001 17.495 17.586
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.494 17.585
ot_scf_mini 108 9.5 0.002 0.002 16.336 16.434
multiply_cannon 2286 13.5 0.304 0.312 14.445 15.559
multiply_cannon_loop 2286 14.5 0.343 0.351 13.021 14.089
ot_mini 108 10.5 0.001 0.001 9.481 9.599
multiply_cannon_multrec 9144 15.5 3.301 4.717 8.743 8.907
init_scf_loop 11 6.9 0.000 0.000 8.222 8.226
rebuild_ks_matrix 119 8.3 0.000 0.000 7.449 7.596
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.449 7.596
qs_ot_get_derivative 108 11.5 0.001 0.001 7.379 7.473
prepare_preconditioner 11 7.9 0.000 0.000 7.175 7.189
make_preconditioner 11 8.9 0.000 0.000 7.175 7.189
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.679 7.040
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.705 6.840
dbcsr_mm_accdrv_process 12550 15.8 4.294 5.698 5.317 6.630
cp_fm_upper_to_full 72 14.2 3.173 4.586 3.173 4.586
init_scf_run 11 5.9 0.000 0.001 4.494 4.494
scf_env_initial_rho_setup 11 6.9 0.002 0.003 4.494 4.494
mp_waitall_1 94719 16.7 3.270 4.405 3.270 4.405
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.225 4.264
calculate_rho_elec 119 8.7 0.118 0.121 4.224 4.264
sum_up_and_integrate 119 10.3 0.064 0.065 4.064 4.071
integrate_v_rspace 119 11.3 0.003 0.003 4.000 4.007
qs_ot_get_p 119 10.4 0.001 0.001 3.858 4.003
make_m2s 4572 13.5 0.039 0.040 2.887 3.134
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.696 3.111
make_images 4572 14.5 0.355 0.383 2.764 3.011
calculate_first_density_matrix 1 7.0 0.000 0.000 2.984 2.991
dbcsr_complete_redistribute 329 12.2 0.292 0.304 2.147 2.988
multiply_cannon_metrocomm3 9144 15.5 0.020 0.020 1.659 2.640
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.792 2.622
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.252 2.531
apply_single 119 13.6 0.000 0.000 2.252 2.531
calculate_dm_sparse 119 9.5 0.000 0.000 2.453 2.479
pw_transfer 1439 11.6 0.066 0.067 2.405 2.412
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.310 2.316
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.491 2.315
mp_alltoall_i22 627 13.8 1.462 2.308 1.462 2.308
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.288 2.289
density_rs2pw 119 9.7 0.003 0.003 2.242 2.263
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.190 2.191
ot_diis_step 108 11.5 0.014 0.015 2.062 2.063
grid_integrate_task_list 119 12.3 2.031 2.056 2.031 2.056
mp_sum_l 7207 12.9 1.277 2.025 1.277 2.025
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.989 1.989
qs_energies_init_hamiltonians 11 5.9 0.020 0.026 1.987 1.988
rs_pw_transfer 974 11.9 0.009 0.009 1.914 1.953
fft3d_ps 1201 14.6 0.581 0.591 1.944 1.952
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.877 1.928
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.857 1.906
fft_wrap_pw1pw2_140 487 13.2 0.088 0.091 1.857 1.866
make_images_data 4572 15.5 0.043 0.046 1.439 1.772
hybrid_alltoall_any 4725 16.4 0.087 0.149 1.424 1.730
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.659 1.659
cp_fm_diag_elpa_base 50 14.0 1.513 1.567 1.657 1.657
cp_fm_cholesky_invert 11 10.9 1.631 1.634 1.631 1.634
potential_pw2rs 119 12.3 0.014 0.015 1.497 1.503
grid_collocate_task_list 119 9.7 1.468 1.489 1.468 1.489
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.465 1.479
mp_alltoall_d11v 2130 13.8 1.415 1.460 1.415 1.460
wfi_extrapolate 11 7.9 0.001 0.001 1.433 1.434
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.283 1.308
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.282 1.300
mp_alltoall_z22v 1201 16.6 1.226 1.248 1.226 1.248
jit_kernel_multiply 6 15.7 0.995 1.160 0.995 1.160
multiply_cannon_sync_h2d 9144 15.5 1.044 1.049 1.044 1.049
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.988 1.042
acc_transpose_blocks 9144 15.5 0.038 0.038 1.019 1.025
qs_create_task_list 11 7.9 0.049 0.067 1.011 1.023
generate_qs_task_list 11 8.9 0.374 0.393 0.961 0.991
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.961 0.974
mp_allgather_i34 2286 14.5 0.385 0.930 0.385 0.930
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=45.489000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=703.636364, yerr=17.379574
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 500.137984E+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 1383689.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
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 0.058 0.098 95.723 95.729
qs_mol_dyn_low 1 2.0 0.003 0.003 94.691 94.711
qs_forces 11 3.9 0.019 0.033 94.584 94.586
qs_energies 11 4.9 0.001 0.002 90.805 90.842
scf_env_do_scf 11 5.9 0.000 0.001 80.485 80.487
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 74.035 74.036
dbcsr_multiply_generic 2055 12.4 0.107 0.112 56.564 56.841
qs_scf_new_mos 99 7.5 0.000 0.001 54.730 54.842
qs_scf_loop_do_ot 99 8.5 0.000 0.001 54.730 54.842
ot_scf_mini 99 9.5 0.002 0.002 52.087 52.181
velocity_verlet 10 3.0 0.001 0.002 48.094 48.096
multiply_cannon 2055 13.4 0.182 0.192 44.371 45.060
multiply_cannon_loop 2055 14.4 1.541 1.583 42.740 43.339
ot_mini 99 10.5 0.001 0.001 31.157 31.251
qs_ot_get_derivative 99 11.5 0.001 0.001 24.204 24.292
multiply_cannon_multrec 49320 15.4 11.914 12.472 16.960 17.723
rebuild_ks_matrix 110 8.3 0.000 0.001 15.662 15.763
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 15.661 15.762
mp_waitall_1 220248 16.4 14.245 15.262 14.245 15.262
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.371 13.467
qs_ot_get_p 110 10.4 0.001 0.001 11.648 11.807
multiply_cannon_sync_h2d 49320 15.4 9.652 10.253 9.652 10.253
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 8.081 8.620
multiply_cannon_metrocomm3 49320 15.4 0.077 0.081 7.364 8.422
qs_rho_update_rho_low 110 7.6 0.000 0.001 7.999 8.154
calculate_rho_elec 110 8.6 0.020 0.024 7.999 8.154
sum_up_and_integrate 110 10.3 0.036 0.043 7.909 7.926
integrate_v_rspace 110 11.3 0.002 0.003 7.873 7.899
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 7.803 7.851
init_scf_run 11 5.9 0.000 0.001 7.764 7.764
scf_env_initial_rho_setup 11 6.9 0.009 0.025 7.764 7.764
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.258 7.723
apply_single 110 13.6 0.000 0.000 7.258 7.723
qs_ot_p2m_diag 48 11.0 0.012 0.023 7.199 7.254
mp_sum_l 6514 12.8 6.091 6.845 6.091 6.845
ot_diis_step 99 11.5 0.005 0.006 6.662 6.662
init_scf_loop 11 6.9 0.000 0.000 6.417 6.417
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.970 5.974
make_m2s 4110 13.4 0.063 0.067 5.682 5.821
make_images 4110 14.4 0.179 0.192 5.584 5.725
rs_pw_transfer 902 11.9 0.011 0.013 5.278 5.555
multiply_cannon_metrocomm1 49320 15.4 0.059 0.063 3.971 5.237
dbcsr_mm_accdrv_process 87628 16.1 2.067 2.149 4.928 5.222
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.089 5.094
cp_fm_redistribute_end 48 14.0 4.420 5.036 4.426 5.037
density_rs2pw 110 9.6 0.004 0.004 4.838 4.982
cp_fm_diag_elpa_base 48 14.0 0.601 4.839 0.605 4.861
wfi_extrapolate 11 7.9 0.001 0.001 4.657 4.657
pw_transfer 1331 11.6 0.054 0.063 4.167 4.319
calculate_dm_sparse 110 9.5 0.021 0.083 4.196 4.310
prepare_preconditioner 11 7.9 0.000 0.000 4.224 4.236
make_preconditioner 11 8.9 0.000 0.000 4.224 4.236
fft_wrap_pw1pw2 1111 12.6 0.007 0.008 4.079 4.235
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.991 4.032
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.895 3.954
fft3d_ps 1111 14.6 0.754 0.841 3.621 3.761
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.755 3.759
fft_wrap_pw1pw2_140 451 13.1 0.166 0.186 3.362 3.518
mp_alltoall_d11v 2046 13.8 3.036 3.512 3.036 3.512
potential_pw2rs 110 12.3 0.006 0.007 3.383 3.424
grid_integrate_task_list 110 12.3 3.233 3.396 3.233 3.396
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.238 3.289
calculate_first_density_matrix 1 7.0 0.009 0.037 3.012 3.026
make_images_data 4110 15.4 0.042 0.045 2.431 2.729
mp_alltoall_z22v 1111 16.6 2.452 2.656 2.452 2.656
jit_kernel_multiply 13 15.9 2.573 2.597 2.573 2.597
hybrid_alltoall_any 4261 16.3 0.080 0.469 2.157 2.456
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.436 2.449
mp_waitany 14300 13.8 2.075 2.446 2.075 2.446
acc_transpose_blocks 49320 15.4 0.225 0.237 2.183 2.271
make_images_sizes 4110 15.4 0.004 0.004 1.678 2.253
mp_alltoall_i44 4110 16.4 1.674 2.249 1.674 2.249
mp_sum_d 3879 11.9 1.764 2.228 1.764 2.228
grid_collocate_task_list 110 9.6 2.088 2.208 2.088 2.208
cp_fm_cholesky_invert 11 10.9 2.005 2.009 2.005 2.009
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=95.729000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=475.545455, yerr=2.061052
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 585.560064E+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 1995690.
MP_SendRecv 10340 26400.
MP_ISendRecv 10340 26400.
MP_Wait 22352
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 0.106 0.171 79.254 79.255
qs_mol_dyn_low 1 2.0 0.003 0.012 78.225 78.238
qs_forces 11 3.9 0.021 0.034 78.146 78.147
qs_energies 11 4.9 0.002 0.009 74.686 74.702
scf_env_do_scf 11 5.9 0.001 0.002 63.968 63.971
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 54.389 54.389
velocity_verlet 10 3.0 0.001 0.002 40.961 40.962
dbcsr_multiply_generic 2055 12.4 0.114 0.120 40.485 40.905
qs_scf_new_mos 99 7.5 0.001 0.001 35.979 36.085
qs_scf_loop_do_ot 99 8.5 0.001 0.001 35.979 36.084
ot_scf_mini 99 9.5 0.003 0.004 34.239 34.347
multiply_cannon 2055 13.4 0.221 0.243 32.361 33.623
multiply_cannon_loop 2055 14.4 0.924 0.949 30.766 31.976
ot_mini 99 10.5 0.001 0.001 20.011 20.130
rebuild_ks_matrix 110 8.3 0.000 0.001 15.758 15.914
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.015 15.758 15.913
multiply_cannon_multrec 24660 15.4 7.499 9.306 13.674 15.451
qs_ks_update_qs_env 110 7.6 0.001 0.001 14.030 14.166
qs_ot_get_derivative 99 11.5 0.001 0.001 13.826 13.931
mp_waitall_1 176588 16.5 10.326 13.247 10.326 13.247
init_scf_loop 11 6.9 0.001 0.005 9.541 9.541
multiply_cannon_metrocomm3 24660 15.4 0.070 0.073 6.206 9.415
sum_up_and_integrate 110 10.3 0.053 0.059 8.337 8.348
integrate_v_rspace 110 11.3 0.002 0.002 8.284 8.296
multiply_cannon_sync_h2d 24660 15.4 6.937 7.935 6.937 7.935
init_scf_run 11 5.9 0.000 0.001 7.856 7.857
scf_env_initial_rho_setup 11 6.9 0.011 0.014 7.856 7.856
prepare_preconditioner 11 7.9 0.000 0.000 7.446 7.465
make_preconditioner 11 8.9 0.000 0.002 7.446 7.465
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.702 7.443
apply_single 110 13.6 0.000 0.001 6.702 7.443
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.020 7.186
qs_ot_get_p 110 10.4 0.001 0.001 6.984 7.102
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.382 6.400
calculate_rho_elec 110 8.6 0.039 0.047 6.381 6.399
dbcsr_mm_accdrv_process 52282 16.1 4.717 5.391 6.017 6.293
ot_diis_step 99 11.5 0.010 0.010 6.080 6.081
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.904 5.649
make_m2s 4110 13.4 0.058 0.061 5.046 5.440
make_images 4110 14.4 0.401 0.448 4.935 5.329
qs_ot_p2m_diag 48 11.0 0.028 0.044 4.835 4.876
rs_pw_transfer 902 11.9 0.012 0.013 4.455 4.636
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.190 4.193
pw_transfer 1331 11.6 0.066 0.070 3.972 4.146
wfi_extrapolate 11 7.9 0.001 0.001 4.115 4.115
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.865 4.041
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.900 3.957
potential_pw2rs 110 12.3 0.008 0.008 3.899 3.947
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.793 3.796
density_rs2pw 110 9.6 0.004 0.004 3.585 3.742
calculate_first_density_matrix 1 7.0 0.001 0.003 3.636 3.650
cp_fm_cholesky_invert 11 10.9 3.533 3.542 3.533 3.542
fft3d_ps 1111 14.6 1.078 1.272 3.223 3.388
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.335 3.340
mp_sum_l 6514 12.8 2.509 3.311 2.509 3.311
cp_fm_redistribute_end 48 14.0 2.504 3.307 2.508 3.309
grid_integrate_task_list 110 12.3 3.118 3.291 3.118 3.291
make_images_data 4110 15.4 0.047 0.050 2.826 3.269
cp_fm_diag_elpa_base 48 14.0 0.763 3.106 0.794 3.192
hybrid_alltoall_any 4261 16.3 0.101 0.439 2.486 3.180
fft_wrap_pw1pw2_140 451 13.1 0.199 0.217 3.007 3.175
calculate_dm_sparse 110 9.5 0.001 0.001 3.069 3.111
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.993 3.068
mp_alltoall_d11v 2046 13.8 2.636 2.931 2.636 2.931
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.363 2.399
grid_collocate_task_list 110 9.6 2.126 2.287 2.126 2.287
mp_allgather_i34 2055 14.4 0.922 2.189 0.922 2.189
cp_fm_cholesky_decompose 22 10.9 2.147 2.157 2.147 2.157
qs_energies_init_hamiltonians 11 5.9 0.006 0.011 2.085 2.105
dbcsr_complete_redistribute 325 12.2 0.278 0.404 1.808 2.102
parallel_gemm_fm 81 9.0 0.000 0.000 1.986 1.997
parallel_gemm_fm_cosma 81 10.0 1.986 1.997 1.986 1.997
make_basis_sm 11 9.8 0.000 0.001 1.966 1.968
jit_kernel_multiply 10 16.4 0.944 1.943 0.944 1.943
mp_alltoall_z22v 1111 16.6 1.859 1.941 1.859 1.941
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.897 1.914
mp_sum_d 3879 11.9 1.281 1.776 1.281 1.776
multiply_cannon_metrocomm4 22605 15.4 0.072 0.077 0.807 1.675
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.643 1.669
mp_waitany 10164 13.8 1.471 1.642 1.471 1.642
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=79.255000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=554.272727, yerr=7.097899
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 662.646784E+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_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
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 0.031 0.072 64.376 64.377
qs_mol_dyn_low 1 2.0 0.003 0.006 63.985 63.998
qs_forces 11 3.9 0.002 0.005 63.664 63.665
qs_energies 11 4.9 0.004 0.023 60.422 60.430
scf_env_do_scf 11 5.9 0.001 0.001 51.949 51.949
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 42.502 42.502
velocity_verlet 10 3.0 0.011 0.037 34.881 34.890
dbcsr_multiply_generic 2055 12.4 0.109 0.111 30.105 30.493
qs_scf_new_mos 99 7.5 0.001 0.001 27.356 27.452
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.356 27.451
ot_scf_mini 99 9.5 0.002 0.003 26.070 26.179
multiply_cannon 2055 13.4 0.211 0.222 22.868 24.275
multiply_cannon_loop 2055 14.4 0.615 0.628 21.493 22.556
ot_mini 99 10.5 0.001 0.001 14.685 14.804
rebuild_ks_matrix 110 8.3 0.000 0.000 12.662 12.828
qs_ks_build_kohn_sham_matrix 110 9.3 0.016 0.043 12.662 12.827
mp_waitall_1 139946 16.5 8.272 11.537 8.272 11.537
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.194 11.337
multiply_cannon_multrec 16440 15.4 3.719 4.500 9.459 10.144
qs_ot_get_derivative 99 11.5 0.001 0.001 10.031 10.139
init_scf_loop 11 6.9 0.001 0.006 9.403 9.404
multiply_cannon_metrocomm3 16440 15.4 0.044 0.046 5.041 8.034
prepare_preconditioner 11 7.9 0.000 0.000 7.521 7.541
make_preconditioner 11 8.9 0.000 0.003 7.521 7.541
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.819 7.195
sum_up_and_integrate 110 10.3 0.060 0.061 6.757 6.770
integrate_v_rspace 110 11.3 0.002 0.003 6.696 6.709
qs_ot_get_p 110 10.4 0.001 0.001 6.326 6.476
init_scf_run 11 5.9 0.000 0.001 5.874 5.874
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.874 5.874
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.811 5.817
calculate_rho_elec 110 8.6 0.058 0.058 5.810 5.817
dbcsr_mm_accdrv_process 34862 16.1 4.636 5.272 5.597 5.694
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.123 5.579
apply_single 110 13.6 0.000 0.000 5.123 5.578
make_m2s 4110 13.4 0.051 0.053 4.731 5.194
make_images 4110 14.4 0.395 0.517 4.614 5.077
ot_diis_step 99 11.5 0.010 0.011 4.602 4.603
qs_ot_p2m_diag 48 11.0 0.042 0.044 4.406 4.413
multiply_cannon_sync_h2d 16440 15.4 3.656 4.184 3.656 4.184
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.424 4.045
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.015 4.015
grid_integrate_task_list 110 12.3 3.176 3.465 3.176 3.465
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.385 3.390
pw_transfer 1331 11.6 0.065 0.071 3.362 3.374
cp_fm_redistribute_end 48 14.0 2.105 3.339 2.108 3.341
rs_pw_transfer 902 11.9 0.010 0.011 3.068 3.325
make_images_data 4110 15.4 0.043 0.048 2.777 3.312
density_rs2pw 110 9.6 0.004 0.004 3.066 3.284
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.256 3.270
cp_fm_diag_elpa_base 48 14.0 1.160 3.158 1.224 3.269
hybrid_alltoall_any 4261 16.3 0.105 0.374 2.433 3.143
wfi_extrapolate 11 7.9 0.001 0.001 3.137 3.137
cp_fm_cholesky_invert 11 10.9 3.073 3.083 3.073 3.083
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.913 2.915
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.768 2.839
mp_sum_l 6514 12.8 1.911 2.755 1.911 2.755
fft_wrap_pw1pw2_140 451 13.1 0.209 0.214 2.692 2.706
calculate_first_density_matrix 1 7.0 0.001 0.004 2.649 2.651
calculate_dm_sparse 110 9.5 0.001 0.001 2.568 2.606
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.534 2.593
fft3d_ps 1111 14.6 1.068 1.077 2.572 2.585
potential_pw2rs 110 12.3 0.010 0.010 2.496 2.504
multiply_cannon_metrocomm4 14385 15.4 0.045 0.049 0.929 2.466
mp_alltoall_d11v 2046 13.8 2.090 2.453 2.090 2.453
mp_irecv_dv 48980 15.7 0.858 2.338 0.858 2.338
grid_collocate_task_list 110 9.6 2.170 2.336 2.170 2.336
dbcsr_complete_redistribute 325 12.2 0.317 0.331 1.590 2.056
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 2.053 2.055
mp_allgather_i34 2055 14.4 0.700 1.956 0.700 1.956
cp_fm_upper_to_full 70 13.6 1.426 1.899 1.426 1.899
cp_fm_cholesky_decompose 22 10.9 1.824 1.843 1.824 1.843
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.672 1.694
mp_waitany 17072 13.8 1.290 1.587 1.290 1.587
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.068 1.535
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.488 1.504
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.370 1.494
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.386 1.397
rs_gather_matrices 110 12.3 0.138 0.151 0.968 1.335
mp_alltoall_z22v 1111 16.6 1.290 1.313 1.290 1.313
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=64.377000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=625.363636, yerr=8.844852
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 736.083968E+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_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
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 0.033 0.096 67.200 67.202
qs_mol_dyn_low 1 2.0 0.004 0.005 66.760 66.769
qs_forces 11 3.9 0.006 0.032 66.685 66.687
qs_energies 11 4.9 0.003 0.008 63.224 63.233
scf_env_do_scf 11 5.9 0.001 0.001 54.129 54.133
scf_env_do_scf_inner_loop 99 6.5 0.003 0.006 41.921 41.921
velocity_verlet 10 3.0 0.002 0.002 37.719 37.721
dbcsr_multiply_generic 2055 12.4 0.118 0.129 30.352 30.578
qs_scf_new_mos 99 7.5 0.001 0.001 27.205 27.300
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.205 27.300
ot_scf_mini 99 9.5 0.003 0.004 25.549 25.632
multiply_cannon 2055 13.4 0.242 0.262 22.913 24.112
multiply_cannon_loop 2055 14.4 0.884 0.902 21.485 22.126
ot_mini 99 10.5 0.001 0.001 14.540 14.654
multiply_cannon_multrec 24660 15.4 4.154 6.420 12.824 13.894
rebuild_ks_matrix 110 8.3 0.000 0.000 12.104 12.223
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.026 12.104 12.223
init_scf_loop 11 6.9 0.001 0.006 12.167 12.168
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.707 10.812
prepare_preconditioner 11 7.9 0.000 0.000 10.398 10.412
make_preconditioner 11 8.9 0.000 0.001 10.398 10.412
qs_ot_get_derivative 99 11.5 0.001 0.001 10.283 10.374
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.605 10.056
dbcsr_mm_accdrv_process 52304 16.0 6.799 8.008 8.524 9.545
mp_waitall_1 121746 16.5 4.811 6.952 4.811 6.952
sum_up_and_integrate 110 10.3 0.068 0.071 6.463 6.476
integrate_v_rspace 110 11.3 0.002 0.003 6.396 6.410
init_scf_run 11 5.9 0.000 0.001 6.149 6.150
scf_env_initial_rho_setup 11 6.9 0.002 0.005 6.149 6.150
qs_ot_get_p 110 10.4 0.001 0.001 5.958 6.113
make_m2s 4110 13.4 0.061 0.063 5.704 6.017
make_images 4110 14.4 0.577 0.698 5.561 5.871
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.811 5.819
calculate_rho_elec 110 8.6 0.077 0.081 5.810 5.818
cp_fm_upper_to_full 70 13.8 3.265 4.644 3.265 4.644
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.090 4.217
apply_single 110 13.6 0.000 0.000 4.090 4.217
ot_diis_step 99 11.5 0.011 0.012 4.212 4.212
qs_ot_p2m_diag 48 11.0 0.055 0.064 4.074 4.089
dbcsr_complete_redistribute 325 12.2 0.420 0.474 2.749 3.875
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.576 3.577
multiply_cannon_metrocomm3 24660 15.4 0.036 0.038 1.638 3.544
grid_integrate_task_list 110 12.3 3.266 3.474 3.266 3.474
multiply_cannon_sync_h2d 24660 15.4 3.181 3.453 3.181 3.453
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.410 3.450
pw_transfer 1331 11.6 0.065 0.072 3.379 3.408
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.200 3.318
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.273 3.303
make_images_data 4110 15.4 0.047 0.050 2.890 3.267
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.171 3.225
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.200 3.202
hybrid_alltoall_any 4261 16.3 0.119 0.457 2.616 3.193
density_rs2pw 110 9.6 0.004 0.004 2.877 3.092
calculate_dm_sparse 110 9.5 0.001 0.001 3.042 3.078
cp_fm_cholesky_invert 11 10.9 3.064 3.072 3.064 3.072
wfi_extrapolate 11 7.9 0.001 0.001 3.041 3.041
calculate_first_density_matrix 1 7.0 0.001 0.003 3.017 3.020
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.952 2.954
cp_fm_redistribute_end 48 14.0 1.478 2.923 1.479 2.924
cp_fm_diag_elpa_base 48 14.0 1.362 2.778 1.441 2.891
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.784 2.882
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.793 2.829
mp_alltoall_i22 605 13.7 1.664 2.826 1.664 2.826
rs_pw_transfer 902 11.9 0.010 0.011 2.531 2.819
fft_wrap_pw1pw2_140 451 13.1 0.201 0.214 2.750 2.783
fft3d_ps 1111 14.6 1.063 1.095 2.578 2.596
grid_collocate_task_list 110 9.6 2.225 2.401 2.225 2.401
qs_energies_init_hamiltonians 11 5.9 0.001 0.004 2.354 2.361
mp_alltoall_d11v 2046 13.8 2.120 2.353 2.120 2.353
jit_kernel_multiply 10 15.6 1.390 2.222 1.390 2.222
potential_pw2rs 110 12.3 0.012 0.013 2.178 2.191
cp_fm_cholesky_decompose 22 10.9 1.799 1.854 1.799 1.854
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.758 1.788
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.612 1.722
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.646 1.658
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.626 1.645
acc_transpose_blocks 24660 15.4 0.106 0.112 1.518 1.540
mp_sum_l 6514 12.8 1.079 1.495 1.079 1.495
multiply_cannon_metrocomm4 20550 15.4 0.057 0.061 0.876 1.461
mp_waitany 13376 13.8 1.107 1.445 1.107 1.445
qs_env_update_s_mstruct 11 6.9 0.003 0.017 1.282 1.412
mp_allgather_i34 2055 14.4 0.530 1.379 0.530 1.379
mp_alltoall_z22v 1111 16.6 1.299 1.357 1.299 1.357
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=67.202000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=696.363636, yerr=11.088673
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 844.828672E+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_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
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 0.019 0.047 57.028 57.028
qs_mol_dyn_low 1 2.0 0.003 0.003 56.701 56.710
qs_forces 11 3.9 0.002 0.002 56.634 56.635
qs_energies 11 4.9 0.001 0.001 52.964 52.969
scf_env_do_scf 11 5.9 0.000 0.001 44.423 44.423
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 36.257 36.257
velocity_verlet 10 3.0 0.002 0.002 31.736 31.739
dbcsr_multiply_generic 2055 12.4 0.106 0.108 23.749 23.897
qs_scf_new_mos 99 7.5 0.001 0.001 21.641 21.700
qs_scf_loop_do_ot 99 8.5 0.001 0.001 21.640 21.700
ot_scf_mini 99 9.5 0.002 0.002 20.375 20.420
multiply_cannon 2055 13.4 0.247 0.271 17.902 19.352
multiply_cannon_loop 2055 14.4 0.319 0.330 16.475 16.882
rebuild_ks_matrix 110 8.3 0.000 0.000 11.754 11.788
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.754 11.787
ot_mini 99 10.5 0.001 0.001 10.947 10.993
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.458 10.490
multiply_cannon_multrec 8220 15.4 3.234 4.722 7.570 8.765
mp_waitall_1 103326 16.6 6.718 8.417 6.718 8.417
init_scf_loop 11 6.9 0.000 0.000 8.118 8.119
qs_ot_get_derivative 99 11.5 0.001 0.001 7.028 7.069
prepare_preconditioner 11 7.9 0.000 0.000 6.472 6.479
make_preconditioner 11 8.9 0.000 0.000 6.472 6.478
sum_up_and_integrate 110 10.3 0.080 0.081 6.406 6.417
integrate_v_rspace 110 11.3 0.003 0.003 6.326 6.337
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.026 6.110
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.890 5.906
calculate_rho_elec 110 8.6 0.115 0.115 5.890 5.906
qs_ot_get_p 110 10.4 0.001 0.001 5.405 5.457
init_scf_run 11 5.9 0.000 0.001 5.347 5.347
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.347 5.347
dbcsr_mm_accdrv_process 17442 15.9 2.802 3.919 4.206 5.110
make_m2s 4110 13.4 0.040 0.041 4.450 4.689
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 3.184 4.657
make_images 4110 14.4 0.638 0.698 4.318 4.556
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.933 3.937
ot_diis_step 99 11.5 0.012 0.012 3.889 3.890
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.781 3.834
apply_single 110 13.6 0.000 0.000 3.780 3.834
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.605 3.606
grid_integrate_task_list 110 12.3 3.363 3.523 3.363 3.523
pw_transfer 1331 11.6 0.065 0.073 3.446 3.472
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.338 3.367
cp_fm_cholesky_invert 11 10.9 3.214 3.218 3.214 3.218
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.062 3.063
cp_fm_redistribute_end 48 14.0 0.779 3.032 0.783 3.033
multiply_cannon_sync_h2d 8220 15.4 2.893 3.013 2.893 3.013
cp_fm_diag_elpa_base 48 14.0 2.063 2.830 2.241 2.998
make_images_data 4110 15.4 0.039 0.044 2.553 2.987
density_rs2pw 110 9.6 0.004 0.004 2.876 2.957
hybrid_alltoall_any 4261 16.3 0.199 0.864 2.437 2.879
fft_wrap_pw1pw2_140 451 13.1 0.212 0.215 2.843 2.877
wfi_extrapolate 11 7.9 0.001 0.001 2.733 2.733
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.718 2.719
qs_energies_init_hamiltonians 11 5.9 0.006 0.009 2.669 2.670
fft3d_ps 1111 14.6 1.119 1.176 2.597 2.627
calculate_dm_sparse 110 9.5 0.001 0.001 2.492 2.536
calculate_first_density_matrix 1 7.0 0.000 0.000 2.528 2.529
rs_pw_transfer 902 11.9 0.010 0.011 2.364 2.481
grid_collocate_task_list 110 9.6 2.329 2.447 2.329 2.447
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.344 2.360
mp_alltoall_d11v 2046 13.8 1.894 2.075 1.894 2.075
potential_pw2rs 110 12.3 0.015 0.015 2.048 2.060
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.029 2.051
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.976 1.998
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.773 1.994
mp_allgather_i34 2055 14.4 0.613 1.788 0.613 1.788
cp_fm_cholesky_decompose 22 10.9 1.760 1.770 1.760 1.770
jit_kernel_multiply 9 15.5 1.089 1.724 1.089 1.724
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.630 1.633
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.504 1.623
dbcsr_complete_redistribute 325 12.2 0.555 0.576 1.533 1.620
multiply_cannon_metrocomm1 8220 15.4 0.021 0.022 1.055 1.495
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.477 1.490
qs_create_task_list 11 7.9 0.001 0.001 1.229 1.326
generate_qs_task_list 11 8.9 0.379 0.449 1.228 1.325
mp_alltoall_z22v 1111 16.6 1.245 1.268 1.245 1.268
mp_waitany 9240 13.8 1.108 1.232 1.108 1.232
multiply_cannon_metrocomm4 6165 15.4 0.018 0.019 0.485 1.214
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.177 1.210
mp_irecv_dv 24056 15.7 0.460 1.173 0.460 1.173
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=57.028000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=791.181818, yerr=14.788313
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 1.339625E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67104.
MP_Allreduce 9672 819.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.084 0.119 91.801 91.802
qs_mol_dyn_low 1 2.0 0.003 0.004 91.326 91.336
qs_forces 11 3.9 0.017 0.033 91.225 91.226
qs_energies 11 4.9 0.001 0.001 87.089 87.106
scf_env_do_scf 11 5.9 0.001 0.001 76.287 76.288
velocity_verlet 10 3.0 0.012 0.015 58.730 58.739
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 46.261 46.261
dbcsr_multiply_generic 2055 12.4 0.127 0.135 30.746 30.958
init_scf_loop 11 6.9 0.000 0.000 29.946 29.947
qs_scf_new_mos 99 7.5 0.001 0.001 28.856 28.942
qs_scf_loop_do_ot 99 8.5 0.001 0.001 28.855 28.942
prepare_preconditioner 11 7.9 0.000 0.000 27.886 27.897
make_preconditioner 11 8.9 0.000 0.000 27.885 27.897
make_full_inverse_cholesky 11 9.9 0.000 0.000 21.989 27.200
ot_scf_mini 99 9.5 0.002 0.002 27.022 27.094
multiply_cannon 2055 13.4 0.349 0.370 22.718 23.479
multiply_cannon_loop 2055 14.4 0.340 0.343 20.733 21.236
cp_fm_upper_to_full 70 14.2 12.706 18.158 12.706 18.158
ot_mini 99 10.5 0.001 0.001 15.027 15.101
rebuild_ks_matrix 110 8.3 0.001 0.001 13.837 13.924
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 13.837 13.924
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.612 12.693
dbcsr_complete_redistribute 325 12.2 1.030 1.064 7.673 10.900
mp_waitall_1 84994 16.7 9.107 10.465 9.107 10.465
qs_ot_get_derivative 99 11.5 0.001 0.001 9.847 9.924
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.592 9.814
multiply_cannon_multrec 8220 15.4 4.390 4.604 9.624 9.718
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.881 9.073
mp_alltoall_i22 605 13.7 5.402 8.654 5.402 8.654
sum_up_and_integrate 110 10.3 0.150 0.151 7.051 7.065
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.939 6.972
calculate_rho_elec 110 8.6 0.227 0.227 6.939 6.972
integrate_v_rspace 110 11.3 0.003 0.003 6.900 6.914
make_m2s 4110 13.4 0.044 0.045 5.937 6.568
qs_ot_get_p 110 10.4 0.001 0.001 6.428 6.553
cp_fm_cholesky_invert 11 10.9 6.463 6.468 6.463 6.468
make_images 4110 14.4 0.880 0.934 5.747 6.377
init_scf_run 11 5.9 0.000 0.001 6.314 6.314
scf_env_initial_rho_setup 11 6.9 0.009 0.017 6.314 6.314
multiply_cannon_metrocomm3 8220 15.4 0.018 0.019 5.735 6.233
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 5.089 5.533
apply_single 110 13.6 0.000 0.000 5.088 5.533
dbcsr_mm_accdrv_process 11614 15.7 3.243 3.666 5.092 5.361
ot_diis_step 99 11.5 0.015 0.016 5.141 5.143
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.585 4.591
pw_transfer 1331 11.6 0.074 0.074 4.172 4.175
make_images_data 4110 15.4 0.042 0.045 3.301 4.121
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.098 4.099
hybrid_alltoall_any 4261 16.3 0.255 0.551 3.186 4.062
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 4.055 4.059
multiply_cannon_sync_h2d 8220 15.4 3.956 3.961 3.956 3.961
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 3.804 3.822
grid_integrate_task_list 110 12.3 3.673 3.727 3.673 3.727
wfi_extrapolate 11 7.9 0.001 0.001 3.694 3.694
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.081 3.535
fft_wrap_pw1pw2_140 451 13.1 0.215 0.217 3.431 3.435
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.426 3.427
cp_fm_diag_elpa_base 48 14.0 2.891 3.090 3.425 3.425
density_rs2pw 110 9.6 0.004 0.004 3.335 3.363
fft3d_ps 1111 14.6 1.269 1.279 3.266 3.271
calculate_dm_sparse 110 9.5 0.001 0.001 3.234 3.259
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.146 3.148
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.663 2.699
grid_collocate_task_list 110 9.6 2.640 2.678 2.640 2.678
rs_pw_transfer 902 11.9 0.010 0.011 2.533 2.635
mp_alltoall_d11v 2046 13.8 2.533 2.585 2.533 2.585
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.514 2.544
calculate_first_density_matrix 1 7.0 0.000 0.000 2.503 2.511
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.327 2.437
potential_pw2rs 110 12.3 0.021 0.022 2.360 2.375
cp_fm_cholesky_decompose 22 10.9 2.317 2.346 2.317 2.346
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.094 2.184
qs_create_task_list 11 7.9 0.001 0.001 1.935 1.980
generate_qs_task_list 11 8.9 0.739 0.793 1.935 1.980
mp_allgather_i34 2055 14.4 0.766 1.967 0.766 1.967
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.856 1.914
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.835 1.842
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=91.802000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1201.818182, yerr=62.026121
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 631.545856E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175954870160
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4075 56898.
MP_Allreduce 11228 786.
MP_Sync 170
MP_Alltoall 2226 2529110.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
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 0.015 0.031 216.117 216.118
qs_mol_dyn_low 1 2.0 0.003 0.004 215.483 215.496
qs_forces 11 3.9 0.008 0.014 215.388 215.388
qs_energies 11 4.9 0.001 0.001 209.468 209.485
scf_env_do_scf 11 5.9 0.001 0.001 192.060 192.064
scf_env_do_scf_inner_loop 117 6.6 0.003 0.007 169.072 169.075
velocity_verlet 10 3.0 0.008 0.020 130.079 130.080
dbcsr_multiply_generic 2507 12.6 0.178 0.184 127.816 128.534
qs_scf_new_mos 117 7.6 0.001 0.001 127.374 127.696
qs_scf_loop_do_ot 117 8.6 0.001 0.001 127.373 127.695
ot_scf_mini 117 9.6 0.003 0.003 120.641 120.971
multiply_cannon 2507 13.6 0.237 0.245 102.439 104.869
multiply_cannon_loop 2507 14.6 2.085 2.179 99.753 102.301
ot_mini 117 10.6 0.001 0.001 68.676 69.023
qs_ot_get_derivative 117 11.6 0.001 0.001 43.177 43.485
multiply_cannon_multrec 60168 15.6 33.366 34.815 41.627 43.002
rebuild_ks_matrix 128 8.3 0.001 0.001 35.278 35.616
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 35.278 35.616
mp_waitall_1 267128 16.5 30.908 33.879 30.908 33.879
qs_ks_update_qs_env 128 7.6 0.001 0.001 31.595 31.889
qs_ot_get_p 128 10.4 0.001 0.001 29.601 29.920
multiply_cannon_sync_h2d 60168 15.6 27.355 29.149 27.355 29.149
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.703 25.529
apply_single 128 13.6 0.001 0.001 24.703 25.529
ot_diis_step 117 11.6 0.007 0.008 25.055 25.058
init_scf_loop 11 6.9 0.000 0.000 22.910 22.912
qs_ot_p2m_diag 83 11.4 0.077 0.091 21.284 21.360
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.255 20.499
cp_dbcsr_syevd 83 12.4 0.004 0.005 18.651 18.658
prepare_preconditioner 11 7.9 0.000 0.000 18.066 18.115
make_preconditioner 11 8.9 0.000 0.000 18.066 18.115
multiply_cannon_metrocomm3 60168 15.6 0.116 0.121 16.018 18.011
make_full_inverse_cholesky 11 9.9 0.000 0.000 17.257 17.450
make_m2s 5014 13.6 0.106 0.115 15.527 15.925
sum_up_and_integrate 128 10.3 0.089 0.107 15.733 15.751
make_images 5014 14.6 0.403 0.422 15.343 15.750
integrate_v_rspace 128 11.3 0.003 0.004 15.644 15.666
cp_fm_diag_elpa 83 13.4 0.000 0.000 14.760 14.804
cp_fm_redistribute_end 83 14.4 11.650 14.672 11.673 14.689
cp_fm_diag_elpa_base 83 14.4 2.938 14.196 2.969 14.301
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.107 14.197
calculate_rho_elec 128 8.7 0.045 0.063 14.107 14.196
init_scf_run 11 5.9 0.000 0.001 12.713 12.713
scf_env_initial_rho_setup 11 6.9 0.002 0.002 12.712 12.713
cp_fm_cholesky_invert 11 10.9 10.310 10.322 10.310 10.322
mp_sum_l 7870 13.0 8.587 9.978 8.587 9.978
wfi_extrapolate 11 7.9 0.001 0.001 9.083 9.083
multiply_cannon_metrocomm1 60168 15.6 0.088 0.093 6.856 8.792
calculate_dm_sparse 128 9.5 0.001 0.001 8.529 8.647
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.482 8.613
make_images_data 5014 15.6 0.066 0.070 7.518 8.413
pw_transfer 1547 11.6 0.075 0.106 8.036 8.378
dbcsr_mm_accdrv_process 124484 16.2 3.167 3.403 7.824 8.364
fft_wrap_pw1pw2 1291 12.7 0.010 0.013 7.832 8.151
density_rs2pw 128 9.7 0.006 0.007 7.671 8.143
qs_ot_get_orbitals 117 10.6 0.001 0.001 7.910 8.005
rs_pw_transfer 1046 11.9 0.016 0.018 7.244 7.764
hybrid_alltoall_any 5200 16.5 0.287 2.244 6.384 7.555
grid_integrate_task_list 128 12.3 7.021 7.494 7.021 7.494
fft3d_ps 1291 14.7 2.093 2.579 6.628 6.894
cp_dbcsr_sm_fm_multiply 37 9.5 0.003 0.003 6.754 6.763
mp_alltoall_d11v 2415 14.1 5.203 6.480 5.203 6.480
fft_wrap_pw1pw2_140 523 13.2 0.443 0.506 6.133 6.335
potential_pw2rs 128 12.3 0.009 0.010 5.863 5.932
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.801 5.872
cp_fm_cholesky_decompose 22 10.9 5.457 5.472 5.457 5.472
grid_collocate_task_list 128 9.7 4.723 5.066 4.723 5.066
mp_sum_d 4455 12.2 4.033 4.880 4.033 4.880
make_images_sizes 5014 15.6 0.006 0.007 2.445 4.356
mp_alltoall_i44 5014 16.6 2.439 4.350 2.439 4.350
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=216.118000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=595.909091, yerr=8.140299
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430460020736 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613089636352 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228663E+12 0.0% 0.0% 100.0%
flops max/rank 2.199914E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806316384 0.0% 0.0% 100.0%
number of processed stacks 6022464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.2
marketing flops 145.647559E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 833.388544E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2425920
MPI messages size (bytes):
total size 4.132350E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703416E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 71436 2336489472
32768 < size <= 131072 728832 55956209664
131072 < size <= 4194304 1386864 1409906900992
4194304 < size <= 16777216 155760 1473826772352
16777216 < size 68112 1190343475200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4107 56904.
MP_Allreduce 11307 944.
MP_Sync 170
MP_Alltoall 1983 5140744.
MP_SendRecv 12126 47072.
MP_ISendRecv 12126 47072.
MP_Wait 26114
MP_comm_split 83
MP_ISend 11836 212447.
MP_IRecv 11836 212447.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.095 0.253 207.772 207.781
qs_mol_dyn_low 1 2.0 0.004 0.021 206.508 206.521
qs_forces 11 3.9 0.006 0.008 206.251 206.252
qs_energies 11 4.9 0.011 0.034 199.365 199.383
scf_env_do_scf 11 5.9 0.004 0.012 180.754 180.765
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 144.459 144.462
velocity_verlet 10 3.0 0.019 0.044 129.547 129.552
qs_scf_new_mos 118 7.6 0.001 0.001 103.267 103.727
qs_scf_loop_do_ot 118 8.6 0.001 0.001 103.266 103.727
dbcsr_multiply_generic 2527 12.6 0.188 0.195 102.165 103.485
ot_scf_mini 118 9.6 0.015 0.035 98.238 98.678
multiply_cannon 2527 13.6 0.483 0.531 79.591 84.972
multiply_cannon_loop 2527 14.6 1.263 1.298 75.580 78.543
ot_mini 118 10.6 0.001 0.001 53.327 53.777
mp_waitall_1 216438 16.6 27.700 42.787 27.700 42.787
multiply_cannon_multrec 30324 15.6 21.709 26.121 31.387 36.170
init_scf_loop 11 6.9 0.003 0.008 36.145 36.146
rebuild_ks_matrix 129 8.3 0.001 0.001 34.510 34.954
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.020 34.509 34.953
qs_ks_update_qs_env 129 7.6 0.001 0.001 31.157 31.579
prepare_preconditioner 11 7.9 0.000 0.000 31.409 31.470
make_preconditioner 11 8.9 0.000 0.002 31.409 31.470
qs_ot_get_derivative 118 11.6 0.001 0.002 30.599 31.035
multiply_cannon_metrocomm3 30324 15.6 0.095 0.102 17.427 30.913
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.999 30.558
qs_ot_get_p 129 10.4 0.001 0.001 25.969 26.558
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 22.520 23.758
apply_single 129 13.6 0.001 0.001 22.519 23.758
ot_diis_step 118 11.6 0.014 0.015 22.509 22.511
multiply_cannon_sync_h2d 30324 15.6 19.024 21.080 19.024 21.080
qs_ot_p2m_diag 83 11.4 0.187 0.215 20.447 20.497
cp_dbcsr_syevd 83 12.4 0.005 0.006 19.122 19.123
cp_fm_cholesky_invert 11 10.9 18.385 18.401 18.385 18.401
make_m2s 5054 13.6 0.093 0.098 15.814 17.800
make_images 5054 14.6 1.169 1.353 15.600 17.588
cp_fm_diag_elpa 83 13.4 0.000 0.001 15.603 15.614
sum_up_and_integrate 129 10.3 0.117 0.136 15.497 15.525
cp_fm_redistribute_end 83 14.4 9.093 15.472 9.108 15.474
integrate_v_rspace 129 11.3 0.003 0.003 15.380 15.412
cp_fm_diag_elpa_base 83 14.4 6.108 14.872 6.339 15.251
qs_rho_update_rho_low 129 7.7 0.001 0.001 14.097 14.189
calculate_rho_elec 129 8.7 0.088 0.106 14.096 14.189
init_scf_run 11 5.9 0.000 0.001 13.193 13.195
scf_env_initial_rho_setup 11 6.9 0.035 0.039 13.193 13.195
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 12.219 12.556
make_images_data 5054 15.6 0.065 0.071 9.478 11.815
hybrid_alltoall_any 5240 16.5 0.344 1.525 8.028 10.881
multiply_cannon_metrocomm4 27797 15.6 0.095 0.110 3.864 10.563
mp_irecv_dv 70031 16.3 3.670 10.179 3.670 10.179
dbcsr_mm_accdrv_process 62734 16.2 4.597 5.675 9.162 9.840
wfi_extrapolate 11 7.9 0.001 0.001 8.965 8.965
pw_transfer 1559 11.6 0.086 0.099 8.488 8.570
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 8.262 8.342
density_rs2pw 129 9.7 0.006 0.006 7.659 8.146
cp_fm_cholesky_decompose 22 10.9 8.002 8.080 8.002 8.080
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 6.933 7.700
rs_pw_transfer 1054 12.0 0.014 0.016 6.973 7.687
grid_integrate_task_list 129 12.3 7.183 7.605 7.183 7.605
mp_sum_l 7930 13.1 5.062 7.266 5.062 7.266
fft_wrap_pw1pw2_140 527 13.2 0.469 0.515 6.973 7.059
calculate_dm_sparse 129 9.5 0.001 0.002 6.705 6.883
fft3d_ps 1301 14.7 2.799 2.965 6.624 6.668
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.488 6.499
mp_allgather_i34 2527 14.6 2.570 6.044 2.570 6.044
potential_pw2rs 129 12.3 0.014 0.016 5.735 5.759
qs_ot_get_orbitals 118 10.6 0.001 0.001 5.652 5.726
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.401 5.539
mp_alltoall_d11v 2423 14.1 4.844 5.477 4.844 5.477
grid_collocate_task_list 129 9.7 4.951 5.281 4.951 5.281
mp_sum_d 4489 12.2 3.618 5.248 3.618 5.248
dbcsr_dot_sd 1330 12.0 0.778 0.858 3.199 4.596
dbcsr_complete_redistribute 395 12.7 0.805 0.910 3.595 4.509
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=207.781000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=793.636364, yerr=2.900556
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/17/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 2.928533E+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 3984192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1695.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 938.401792E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1042912
MPI messages size (bytes):
total size 2.716210E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.604448E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 281856 36943429632
131072 < size <= 4194304 660064 996105256960
4194304 < size <= 16777216 65632 931530938576
16777216 < size 28672 751619276800
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4075 57335.
MP_Allreduce 11226 986.
MP_Sync 170
MP_Alltoall 1712 9388896.
MP_SendRecv 7936 75008.
MP_ISendRecv 7936 75008.
MP_Wait 21820
MP_comm_split 83
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.074 0.099 183.735 183.738
qs_mol_dyn_low 1 2.0 0.003 0.005 183.066 183.082
qs_forces 11 3.9 0.004 0.009 182.960 182.964
qs_energies 11 4.9 0.006 0.021 176.287 176.298
scf_env_do_scf 11 5.9 0.001 0.001 159.733 159.733
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 122.316 122.317
velocity_verlet 10 3.0 0.015 0.018 117.578 117.580
qs_scf_new_mos 117 7.6 0.001 0.001 85.147 85.484
qs_scf_loop_do_ot 117 8.6 0.001 0.001 85.146 85.483
dbcsr_multiply_generic 2507 12.6 0.191 0.208 83.291 84.480
ot_scf_mini 117 9.6 0.004 0.005 80.891 81.281
multiply_cannon 2507 13.6 0.497 0.527 62.549 67.241
multiply_cannon_loop 2507 14.6 0.857 0.887 59.205 61.953
ot_mini 117 10.6 0.001 0.001 43.801 44.203
init_scf_loop 11 6.9 0.004 0.018 37.299 37.299
mp_waitall_1 170520 16.6 26.183 35.532 26.183 35.532
prepare_preconditioner 11 7.9 0.000 0.000 33.140 33.187
make_preconditioner 11 8.9 0.007 0.031 33.140 33.187
make_full_inverse_cholesky 11 9.9 0.000 0.000 30.739 32.185
rebuild_ks_matrix 128 8.3 0.001 0.001 30.647 31.160
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.029 30.647 31.160
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.615 28.092
multiply_cannon_metrocomm3 20056 15.6 0.061 0.064 15.812 25.156
multiply_cannon_multrec 20056 15.6 13.392 16.397 22.088 25.118
qs_ot_get_derivative 117 11.6 0.001 0.002 23.908 24.303
qs_ot_get_p 128 10.4 0.001 0.001 22.160 22.591
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.986 20.830
apply_single 128 13.6 0.001 0.001 19.986 20.830
ot_diis_step 117 11.6 0.018 0.019 19.767 19.768
qs_ot_p2m_diag 83 11.4 0.265 0.274 17.712 17.725
cp_dbcsr_syevd 83 12.4 0.005 0.005 16.603 16.604
make_m2s 5014 13.6 0.084 0.089 15.370 16.457
make_images 5014 14.6 1.185 1.279 15.134 16.217
cp_fm_cholesky_invert 11 10.9 15.691 15.700 15.691 15.700
multiply_cannon_sync_h2d 20056 15.6 14.324 15.652 14.324 15.652
sum_up_and_integrate 128 10.3 0.133 0.146 14.507 14.529
integrate_v_rspace 128 11.3 0.003 0.004 14.373 14.398
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.335 13.374
calculate_rho_elec 128 8.7 0.132 0.146 13.335 13.373
cp_fm_diag_elpa 83 13.4 0.000 0.001 13.314 13.317
cp_fm_redistribute_end 83 14.4 5.024 13.214 5.044 13.216
cp_fm_diag_elpa_base 83 14.4 7.711 12.564 8.140 13.052
init_scf_run 11 5.9 0.000 0.001 11.185 11.186
scf_env_initial_rho_setup 11 6.9 0.001 0.003 11.185 11.185
make_images_data 5014 15.6 0.061 0.069 9.405 10.843
hybrid_alltoall_any 5200 16.5 0.432 1.976 8.110 9.849
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.471 9.755
multiply_cannon_metrocomm4 17549 15.6 0.061 0.071 3.415 9.253
mp_irecv_dv 50230 16.2 3.293 9.012 3.293 9.012
dbcsr_mm_accdrv_process 41502 16.2 4.381 5.298 8.173 8.373
cp_fm_cholesky_decompose 22 10.9 8.133 8.168 8.133 8.168
pw_transfer 1547 11.6 0.085 0.105 7.894 7.995
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 7.670 7.779
grid_integrate_task_list 128 12.3 7.341 7.762 7.341 7.762
cp_fm_upper_to_full 105 14.5 5.841 7.695 5.841 7.695
wfi_extrapolate 11 7.9 0.001 0.001 7.685 7.686
density_rs2pw 128 9.7 0.006 0.006 6.806 7.225
dbcsr_complete_redistribute 395 12.7 1.241 1.292 4.894 6.734
fft_wrap_pw1pw2_140 523 13.2 0.475 0.525 6.558 6.672
rs_pw_transfer 1046 11.9 0.014 0.014 5.769 6.193
calculate_dm_sparse 128 9.5 0.001 0.003 5.941 6.026
fft3d_ps 1291 14.7 2.692 2.904 5.939 5.993
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.733 5.737
grid_collocate_task_list 128 9.7 5.081 5.451 5.081 5.451
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.550 5.393
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.679 5.346
mp_allgather_i34 2507 14.6 1.868 5.034 1.868 5.034
mp_sum_l 7870 13.0 3.571 4.961 3.571 4.961
potential_pw2rs 128 12.3 0.020 0.023 4.855 4.878
mp_alltoall_d11v 2415 14.1 4.316 4.826 4.316 4.826
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.616 4.770
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.202 4.234
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.374 4.191
mp_alltoall_i22 716 14.1 2.075 4.093 2.075 4.093
qs_energies_init_hamiltonians 11 5.9 0.009 0.022 4.061 4.064
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=183.738000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=884.818182, yerr=12.582948
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528903135232 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 4.353791E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 5977344 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.145086E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1143192
MPI messages size (bytes):
total size 2.023815E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.770320E+06
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 0 0
8192 < size <= 32768 396 8650752
32768 < size <= 131072 319024 36042702848
131072 < size <= 4194304 715736 785529176064
4194304 < size <= 16777216 70320 665379475120
16777216 < size 30720 536870912000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4085 57194.
MP_Allreduce 11251 1067.
MP_Sync 170
MP_Alltoall 1712 12503107.
MP_SendRecv 5888 75008.
MP_ISendRecv 5888 75008.
MP_Wait 22442
MP_comm_split 83
MP_ISend 14952 244818.
MP_IRecv 14952 244818.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.052 0.108 190.428 190.428
qs_mol_dyn_low 1 2.0 0.003 0.013 189.678 189.691
qs_forces 11 3.9 0.004 0.005 189.513 189.522
qs_energies 11 4.9 0.001 0.001 182.464 182.475
scf_env_do_scf 11 5.9 0.001 0.001 165.087 165.098
velocity_verlet 10 3.0 0.009 0.009 126.087 126.092
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 117.350 117.352
qs_scf_new_mos 117 7.6 0.001 0.001 81.121 81.395
qs_scf_loop_do_ot 117 8.6 0.001 0.001 81.121 81.395
dbcsr_multiply_generic 2507 12.6 0.190 0.195 80.166 80.751
ot_scf_mini 117 9.6 0.003 0.003 76.590 76.846
multiply_cannon 2507 13.6 0.554 0.593 54.961 58.040
multiply_cannon_loop 2507 14.6 1.181 1.208 51.268 52.568
init_scf_loop 11 6.9 0.000 0.000 47.597 47.598
prepare_preconditioner 11 7.9 0.000 0.000 43.502 43.530
make_preconditioner 11 8.9 0.000 0.000 43.502 43.529
ot_mini 117 10.6 0.001 0.001 42.737 43.008
make_full_inverse_cholesky 11 9.9 0.000 0.000 37.123 42.077
multiply_cannon_multrec 30084 15.6 14.097 19.040 26.058 30.750
rebuild_ks_matrix 128 8.3 0.001 0.001 29.430 29.629
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.018 29.430 29.629
mp_waitall_1 147882 16.7 18.073 28.256 18.073 28.256
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.617 26.785
qs_ot_get_derivative 117 11.6 0.001 0.002 23.065 23.319
make_m2s 5014 13.6 0.099 0.103 20.814 21.822
make_images 5014 14.6 1.971 2.233 20.503 21.509
qs_ot_get_p 128 10.4 0.001 0.001 19.938 20.241
ot_diis_step 117 11.6 0.017 0.018 19.549 19.551
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 19.092 19.550
apply_single 128 13.6 0.001 0.001 19.091 19.549
cp_fm_cholesky_invert 11 10.9 17.032 17.042 17.032 17.042
cp_fm_upper_to_full 105 14.7 10.964 16.255 10.964 16.255
qs_ot_p2m_diag 83 11.4 0.342 0.389 15.716 15.768
multiply_cannon_metrocomm3 30084 15.6 0.047 0.049 6.674 15.306
cp_dbcsr_syevd 83 12.4 0.005 0.005 14.406 14.408
sum_up_and_integrate 128 10.3 0.140 0.152 14.349 14.372
integrate_v_rspace 128 11.3 0.003 0.004 14.209 14.236
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.448 13.486
calculate_rho_elec 128 8.7 0.175 0.190 13.448 13.485
dbcsr_complete_redistribute 395 12.7 1.564 1.741 9.252 12.996
make_images_data 5014 15.6 0.064 0.067 11.302 12.957
multiply_cannon_sync_h2d 30084 15.6 11.768 12.511 11.768 12.511
dbcsr_mm_accdrv_process 62264 16.2 7.273 8.354 11.538 12.045
hybrid_alltoall_any 5200 16.5 0.523 2.210 10.149 11.942
copy_fm_to_dbcsr 209 11.7 0.002 0.002 7.794 11.542
init_scf_run 11 5.9 0.000 0.001 11.346 11.348
scf_env_initial_rho_setup 11 6.9 0.011 0.012 11.346 11.347
cp_fm_diag_elpa 83 13.4 0.000 0.001 11.195 11.198
cp_fm_redistribute_end 83 14.4 1.931 11.096 1.948 11.102
cp_fm_diag_elpa_base 83 14.4 8.522 10.487 9.117 10.955
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.356 9.951
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.631 9.828
mp_alltoall_i22 716 14.1 5.552 9.283 5.552 9.283
pw_transfer 1547 11.6 0.085 0.101 8.119 8.218
cp_fm_cholesky_decompose 22 10.9 8.014 8.110 8.014 8.110
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 7.894 7.999
grid_integrate_task_list 128 12.3 7.526 7.957 7.526 7.957
wfi_extrapolate 11 7.9 0.001 0.001 7.785 7.785
fft_wrap_pw1pw2_140 523 13.2 0.478 0.485 6.911 7.035
multiply_cannon_metrocomm4 25070 15.6 0.075 0.085 2.765 7.014
density_rs2pw 128 9.7 0.006 0.006 6.526 6.847
mp_irecv_dv 76098 16.2 2.619 6.752 2.619 6.752
calculate_dm_sparse 128 9.5 0.001 0.001 6.195 6.275
fft3d_ps 1291 14.7 2.796 2.892 6.132 6.209
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.726 5.775
grid_collocate_task_list 128 9.7 5.200 5.478 5.200 5.478
mp_alltoall_d11v 2415 14.1 5.092 5.432 5.092 5.432
rs_pw_transfer 1046 11.9 0.013 0.015 4.896 5.219
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.480 4.562
potential_pw2rs 128 12.3 0.022 0.023 4.506 4.519
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.437 4.510
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.492 4.494
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.229 4.291
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=190.428000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1071.363636, yerr=19.648145
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/19/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 5.865088E+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 1960712 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3445.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.510957E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 240672
MPI messages size (bytes):
total size 1.331455E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.532237E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 113904 59718500352
4194304 < size <= 16777216 104976 550376570880
16777216 < size 20208 721350092304
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8931 51.
MP_Alltoall 9654 799394.
MP_ISend 40068 2102572.
MP_IRecv 40068 2101675.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58352.
MP_Allreduce 10977 1175.
MP_Sync 87
MP_Alltoall 1712 18838210.
MP_SendRecv 3840 122880.
MP_ISendRecv 3840 122880.
MP_Wait 16122
MP_ISend 10680 423556.
MP_IRecv 10680 423556.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.083 176.541 176.541
qs_mol_dyn_low 1 2.0 0.003 0.004 175.913 175.938
qs_forces 11 3.9 0.027 0.051 175.783 175.785
qs_energies 11 4.9 0.001 0.001 168.367 168.383
scf_env_do_scf 11 5.9 0.001 0.001 150.777 150.791
velocity_verlet 10 3.0 0.028 0.030 115.022 115.026
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 113.874 113.875
qs_scf_new_mos 117 7.6 0.001 0.001 77.820 77.883
qs_scf_loop_do_ot 117 8.6 0.001 0.001 77.819 77.882
dbcsr_multiply_generic 2507 12.6 0.183 0.188 75.485 75.921
ot_scf_mini 117 9.6 0.003 0.004 73.367 73.423
multiply_cannon 2507 13.6 0.587 0.625 55.139 60.143
multiply_cannon_loop 2507 14.6 0.444 0.458 50.378 51.277
ot_mini 117 10.6 0.001 0.001 40.608 40.667
init_scf_loop 11 6.9 0.000 0.000 36.742 36.744
mp_waitall_1 125778 16.7 26.569 34.147 26.569 34.147
prepare_preconditioner 11 7.9 0.000 0.000 32.770 32.792
make_preconditioner 11 8.9 0.000 0.000 32.770 32.792
make_full_inverse_cholesky 11 9.9 0.000 0.000 30.619 30.884
rebuild_ks_matrix 128 8.3 0.001 0.001 28.899 28.956
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.018 28.898 28.956
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.357 26.408
multiply_cannon_multrec 10028 15.6 10.543 14.884 18.008 20.741
qs_ot_get_derivative 117 11.6 0.001 0.002 20.469 20.527
ot_diis_step 117 11.6 0.020 0.020 20.066 20.067
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.646 19.925
apply_single 128 13.6 0.001 0.001 19.646 19.925
qs_ot_get_p 128 10.4 0.001 0.001 19.681 19.740
multiply_cannon_metrocomm3 10028 15.6 0.023 0.025 12.560 19.485
make_m2s 5014 13.6 0.068 0.074 16.600 19.015
cp_fm_cholesky_invert 11 10.9 18.799 18.805 18.799 18.805
make_images 5014 14.6 2.348 2.684 16.291 18.706
qs_ot_p2m_diag 83 11.4 0.495 0.501 15.749 15.765
cp_dbcsr_syevd 83 12.4 0.005 0.005 14.566 14.568
sum_up_and_integrate 128 10.3 0.181 0.190 14.263 14.317
integrate_v_rspace 128 11.3 0.003 0.004 14.082 14.144
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.429 13.481
calculate_rho_elec 128 8.7 0.258 0.269 13.429 13.481
make_images_data 5014 15.6 0.053 0.061 10.041 12.636
hybrid_alltoall_any 5200 16.5 0.784 3.629 9.800 12.459
multiply_cannon_sync_h2d 10028 15.6 11.604 11.977 11.604 11.977
cp_fm_diag_elpa 83 13.4 0.000 0.000 11.366 11.375
cp_fm_diag_elpa_base 83 14.4 11.103 11.187 11.352 11.361
init_scf_run 11 5.9 0.000 0.001 10.644 10.644
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.644 10.644
cp_fm_cholesky_decompose 22 10.9 8.231 8.353 8.231 8.353
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 8.175 8.221
grid_integrate_task_list 128 12.3 7.736 8.141 7.736 8.141
mp_allgather_i34 2507 14.6 2.964 8.043 2.964 8.043
pw_transfer 1547 11.6 0.084 0.090 8.010 8.037
dbcsr_mm_accdrv_process 20762 16.1 2.759 3.702 7.093 7.847
fft_wrap_pw1pw2 1291 12.7 0.010 0.010 7.790 7.811
multiply_cannon_metrocomm1 10028 15.6 0.029 0.029 4.582 7.515
wfi_extrapolate 11 7.9 0.001 0.001 7.505 7.505
fft_wrap_pw1pw2_140 523 13.2 0.502 0.523 6.789 6.809
density_rs2pw 128 9.7 0.005 0.006 6.184 6.583
calculate_dm_sparse 128 9.5 0.001 0.001 6.115 6.176
fft3d_ps 1291 14.7 2.738 2.835 5.949 5.997
dbcsr_complete_redistribute 395 12.7 2.134 2.199 5.411 5.832
grid_collocate_task_list 128 9.7 5.482 5.794 5.482 5.794
mp_alltoall_d11v 2415 14.1 5.001 5.733 5.001 5.733
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.333 5.340
qs_energies_init_hamiltonians 11 5.9 0.009 0.020 5.283 5.292
rs_pw_transfer 1046 11.9 0.012 0.013 4.332 4.741
potential_pw2rs 128 12.3 0.027 0.027 4.249 4.264
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.137 4.175
multiply_cannon_metrocomm4 7521 15.6 0.024 0.027 1.862 4.154
mp_irecv_dv 28860 15.9 1.824 4.068 1.824 4.068
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.572 3.878
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.527 3.875
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.754 3.784
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.679 3.750
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=176.541000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1412.272727, yerr=51.352774
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 11.786061E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 1980288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3437.1
marketing flops 145.650931E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 3.037155E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101160
MPI messages size (bytes):
total size 1.144970E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.318403E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 44 2883584
131072 < size <= 4194304 45648 35433480192
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592466352
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 58564.
MP_Allreduce 11104 1511.
MP_Sync 88
MP_Alltoall 1724 36993632.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.037 0.067 316.454 316.455
qs_mol_dyn_low 1 2.0 0.005 0.018 315.503 315.515
qs_forces 11 3.9 0.013 0.032 315.402 315.406
qs_energies 11 4.9 0.001 0.001 305.724 305.742
scf_env_do_scf 11 5.9 0.001 0.001 281.674 281.685
velocity_verlet 10 3.0 0.002 0.002 226.032 226.040
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 151.855 151.857
init_scf_loop 11 6.9 0.000 0.000 129.542 129.544
prepare_preconditioner 11 7.9 0.000 0.000 124.509 124.534
make_preconditioner 11 8.9 0.000 0.000 124.509 124.534
make_full_inverse_cholesky 11 9.9 0.000 0.000 100.070 121.342
qs_scf_new_mos 118 7.6 0.001 0.001 106.036 106.182
qs_scf_loop_do_ot 118 8.6 0.001 0.001 106.035 106.181
ot_scf_mini 118 9.6 0.004 0.004 100.887 101.059
dbcsr_multiply_generic 2529 12.6 0.222 0.228 93.580 94.078
cp_fm_upper_to_full 106 14.8 52.545 75.601 52.545 75.601
multiply_cannon 2529 13.6 0.712 0.765 65.452 67.720
multiply_cannon_loop 2529 14.6 0.471 0.482 61.194 62.362
ot_mini 118 10.6 0.001 0.001 51.351 51.546
dbcsr_complete_redistribute 397 12.7 3.993 4.019 30.606 43.381
copy_fm_to_dbcsr 210 11.7 0.002 0.002 26.983 39.740
mp_waitall_1 104580 16.8 35.492 39.503 35.492 39.503
cp_fm_cholesky_invert 11 10.9 37.488 37.497 37.488 37.497
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 24.394 37.069
rebuild_ks_matrix 129 8.3 0.001 0.001 36.614 36.910
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 36.614 36.909
mp_alltoall_i22 720 14.1 21.955 34.786 21.955 34.786
qs_ks_update_qs_env 129 7.6 0.001 0.001 33.576 33.780
qs_ot_get_p 129 10.4 0.001 0.001 32.372 32.611
qs_ot_get_derivative 118 11.6 0.002 0.002 27.711 27.874
qs_ot_p2m_diag 84 11.4 0.890 0.896 27.068 27.103
cp_dbcsr_syevd 84 12.4 0.006 0.006 25.181 25.186
multiply_cannon_metrocomm3 10116 15.6 0.024 0.024 24.034 25.060
make_m2s 5058 13.6 0.080 0.084 22.230 23.569
ot_diis_step 118 11.6 0.022 0.023 23.550 23.551
make_images 5058 14.6 3.832 4.024 21.742 23.087
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 22.129 22.627
apply_single 129 13.6 0.001 0.001 22.128 22.627
cp_fm_diag_elpa 84 13.4 0.000 0.000 21.481 21.482
cp_fm_diag_elpa_base 84 14.4 17.098 18.679 21.474 21.476
multiply_cannon_multrec 10116 15.6 10.719 12.592 18.594 18.683
sum_up_and_integrate 129 10.3 0.324 0.325 17.165 17.270
qs_rho_update_rho_low 129 7.7 0.001 0.001 16.956 17.050
calculate_rho_elec 129 8.7 0.487 0.488 16.955 17.050
integrate_v_rspace 129 11.3 0.004 0.004 16.841 16.947
multiply_cannon_sync_h2d 10116 15.6 15.794 15.805 15.794 15.805
make_images_data 5058 15.6 0.060 0.066 12.268 14.699
hybrid_alltoall_any 5245 16.5 1.309 3.061 12.175 14.448
init_scf_run 11 5.9 0.000 0.001 13.716 13.717
scf_env_initial_rho_setup 11 6.9 0.002 0.003 13.716 13.717
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 11.183 11.288
pw_transfer 1559 11.6 0.092 0.094 9.779 9.797
cp_fm_cholesky_decompose 22 10.9 9.728 9.741 9.728 9.741
wfi_extrapolate 11 7.9 0.001 0.001 9.724 9.725
fft_wrap_pw1pw2 1301 12.7 0.011 0.011 9.542 9.559
dbcsr_mm_accdrv_process 20934 16.1 3.947 5.831 7.633 9.495
mp_alltoall_d11v 2429 14.1 8.968 9.119 8.968 9.119
grid_integrate_task_list 129 12.3 8.611 8.807 8.611 8.807
qs_energies_init_hamiltonians 11 5.9 0.074 0.125 8.287 8.300
fft_wrap_pw1pw2_140 527 13.2 0.544 0.547 8.284 8.299
fft3d_ps 1301 14.7 2.786 2.813 7.569 7.591
calculate_dm_sparse 129 9.5 0.001 0.001 7.356 7.496
density_rs2pw 129 9.7 0.005 0.005 7.303 7.333
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 7.007 7.079
copy_dbcsr_to_fm 187 11.8 0.004 0.004 6.477 6.540
grid_collocate_task_list 129 9.7 6.392 6.426 6.392 6.426
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=316.455000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2735.454545, yerr=161.334415
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 1.261392E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255669.
MP_Allreduce 3059 6274.
MP_Sync 4
MP_Alltoall 54
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.052 0.211 86.148 86.161
qs_energies 1 2.0 0.000 0.000 85.327 85.341
ls_scf 1 3.0 0.000 0.000 84.418 84.431
dbcsr_multiply_generic 111 6.7 0.014 0.015 73.091 73.239
multiply_cannon 111 7.7 0.018 0.021 56.005 57.358
multiply_cannon_loop 111 8.7 0.208 0.219 52.516 53.976
ls_scf_main 1 4.0 0.000 0.000 52.883 52.884
density_matrix_trs4 2 5.0 0.002 0.003 47.194 47.281
ls_scf_init_scf 1 4.0 0.000 0.000 28.476 28.478
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.326 27.387
mp_waitall_1 11031 10.9 23.179 25.720 23.179 25.720
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.105 25.121
multiply_cannon_multrec 2664 9.7 8.094 8.787 15.482 17.125
multiply_cannon_sync_h2d 2664 9.7 13.407 15.448 13.407 15.448
make_m2s 222 7.7 0.008 0.011 13.467 13.968
make_images 222 8.7 0.099 0.109 13.446 13.950
multiply_cannon_metrocomm1 2664 9.7 0.010 0.010 10.012 12.471
make_images_data 222 9.7 0.004 0.005 7.999 8.603
hybrid_alltoall_any 227 10.6 0.214 1.822 6.820 8.578
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.487 8.244
dbcsr_mm_accdrv_process 4760 10.4 0.509 0.629 7.004 7.975
dbcsr_mm_accdrv_process_sort 4760 11.4 6.296 7.209 6.296 7.209
calculate_norms 4752 9.8 5.481 6.099 5.481 6.099
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.108 5.215
mp_sum_l 807 5.4 3.227 4.473 3.227 4.473
make_images_sizes 222 9.7 0.000 0.000 0.861 3.705
mp_alltoall_i44 222 10.7 0.861 3.705 0.861 3.705
arnoldi_extremal 4 6.8 0.000 0.000 3.400 3.419
arnoldi_normal_ev 4 7.8 0.001 0.003 3.400 3.419
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.288 3.366
build_subspace 16 8.4 0.009 0.012 3.280 3.282
multiply_cannon_metrocomm4 2442 9.7 0.012 0.014 2.060 3.106
mp_irecv_dv 6231 10.9 2.042 3.081 2.042 3.081
ls_scf_post 1 4.0 0.000 0.000 3.058 3.072
ls_scf_store_result 1 5.0 0.000 0.000 2.877 2.919
dbcsr_special_finalize 555 9.7 0.005 0.006 2.259 2.734
dbcsr_merge_single_wm 555 10.7 0.459 0.594 2.251 2.727
dbcsr_matrix_vector_mult 304 9.0 0.003 0.010 2.376 2.628
make_images_pack 222 9.7 2.203 2.620 2.205 2.621
dbcsr_matrix_vector_mult_local 304 10.0 2.067 2.479 2.069 2.481
dbcsr_sort_data 658 11.4 2.051 2.458 2.051 2.458
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.372 2.444
buffer_matrices_ensure_size 222 8.7 1.757 2.047 1.757 2.047
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.825 1.826
rebuild_ks_matrix 3 7.3 0.000 0.000 1.814 1.816
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.002 1.814 1.816
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=86.161000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1142.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 2.101662E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 10339.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.077 0.122 92.516 92.521
qs_energies 1 2.0 0.000 0.000 91.770 91.774
ls_scf 1 3.0 0.003 0.021 90.421 90.425
dbcsr_multiply_generic 111 6.7 0.015 0.015 75.843 76.168
multiply_cannon 111 7.7 0.028 0.044 53.654 57.078
ls_scf_main 1 4.0 0.000 0.000 55.508 55.518
multiply_cannon_loop 111 8.7 0.115 0.122 50.356 53.178
density_matrix_trs4 2 5.0 0.002 0.003 49.683 49.869
mp_waitall_1 9105 10.9 21.760 31.401 21.760 31.401
ls_scf_init_scf 1 4.0 0.000 0.003 31.321 31.323
ls_scf_init_matrix_S 1 5.0 0.000 0.002 29.863 29.967
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.423 27.436
multiply_cannon_multrec 1332 9.7 13.054 16.685 22.340 27.100
multiply_cannon_metrocomm3 1332 9.7 0.006 0.008 12.250 21.730
make_m2s 222 7.7 0.006 0.008 15.499 16.115
make_images 222 8.7 1.577 1.966 15.468 16.084
dbcsr_mm_accdrv_process 4041 10.4 0.280 0.434 8.889 10.448
dbcsr_mm_accdrv_process_sort 4041 11.4 8.486 10.023 8.486 10.023
make_images_data 222 9.7 0.004 0.004 8.929 9.826
hybrid_alltoall_any 227 10.6 0.522 2.479 8.302 9.451
mp_sum_l 807 5.4 5.397 8.798 5.397 8.798
multiply_cannon_metrocomm4 1221 9.7 0.006 0.008 3.247 7.767
mp_irecv_dv 3311 11.0 3.227 7.716 3.227 7.716
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.225 7.076
calculate_norms 2376 9.8 5.981 6.690 5.981 6.690
multiply_cannon_sync_h2d 1332 9.7 4.753 5.765 4.753 5.765
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.041 5.240
arnoldi_extremal 4 6.8 0.000 0.000 4.833 4.855
arnoldi_normal_ev 4 7.8 0.001 0.004 4.833 4.855
build_subspace 16 8.4 0.014 0.021 4.575 4.579
ls_scf_post 1 4.0 0.000 0.001 3.589 3.593
dbcsr_matrix_vector_mult 304 9.0 0.005 0.017 3.252 3.451
ls_scf_store_result 1 5.0 0.000 0.000 3.258 3.411
dbcsr_matrix_vector_mult_local 304 10.0 2.746 3.224 2.748 3.225
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.215 3.186
ls_scf_dm_to_ks 2 5.0 0.001 0.027 2.664 2.757
make_images_pack 222 9.7 2.022 2.414 2.024 2.416
mp_allgather_i34 111 8.7 0.997 2.351 0.997 2.351
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.116 2.118
rebuild_ks_matrix 3 7.3 0.000 0.000 2.103 2.105
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.103 2.105
dbcsr_sort_data 436 11.2 1.804 2.045 1.804 2.045
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=92.521000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1748.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 2.696159E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265470.
MP_Allreduce 3058 11181.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.218 0.375 96.993 97.031
qs_energies 1 2.0 0.005 0.023 95.471 95.511
ls_scf 1 3.0 0.004 0.033 93.787 93.798
dbcsr_multiply_generic 111 6.7 0.016 0.022 77.773 78.052
ls_scf_main 1 4.0 0.003 0.020 58.467 58.488
multiply_cannon 111 7.7 0.043 0.105 53.324 57.308
multiply_cannon_loop 111 8.7 0.099 0.105 49.755 53.880
density_matrix_trs4 2 5.0 0.003 0.021 52.466 52.652
mp_waitall_1 7281 11.0 24.586 35.149 24.586 35.149
ls_scf_init_scf 1 4.0 0.008 0.023 31.663 31.706
ls_scf_init_matrix_S 1 5.0 0.002 0.017 30.026 30.111
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.490 27.509
multiply_cannon_multrec 888 9.7 12.557 15.090 21.183 24.309
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.298 24.121
make_m2s 222 7.7 0.006 0.010 17.491 18.697
make_images 222 8.7 1.974 2.298 17.452 18.658
hybrid_alltoall_any 227 10.6 0.620 2.858 9.652 11.220
make_images_data 222 9.7 0.004 0.004 9.981 11.009
dbcsr_mm_accdrv_process 3754 10.4 0.238 0.438 8.143 9.375
mp_sum_l 807 5.4 5.464 9.271 5.464 9.271
dbcsr_mm_accdrv_process_sort 3754 11.4 7.768 8.937 7.768 8.937
multiply_cannon_sync_h2d 888 9.7 6.035 7.560 6.035 7.560
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.191 7.238
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.441 6.959
mp_irecv_dv 2335 11.1 2.426 6.915 2.426 6.915
multiply_cannon_metrocomm1 888 9.7 0.002 0.003 3.929 6.870
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.042 5.248
arnoldi_extremal 4 6.8 0.000 0.000 5.202 5.223
arnoldi_normal_ev 4 7.8 0.001 0.005 5.202 5.223
build_subspace 16 8.4 0.014 0.020 4.882 4.887
calculate_norms 1584 9.8 4.255 4.616 4.255 4.616
mp_allgather_i34 111 8.7 1.410 3.901 1.410 3.901
dbcsr_matrix_vector_mult 304 9.0 0.005 0.016 3.507 3.818
ls_scf_post 1 4.0 0.006 0.027 3.653 3.667
dbcsr_matrix_vector_mult_local 304 10.0 3.032 3.613 3.034 3.615
ls_scf_store_result 1 5.0 0.000 0.000 3.375 3.478
ls_scf_dm_to_ks 2 5.0 0.001 0.016 2.750 2.846
make_images_sizes 222 9.7 0.000 0.000 1.161 2.413
mp_alltoall_i44 222 10.7 1.161 2.413 1.161 2.413
dbcsr_sort_data 325 11.1 1.891 2.165 1.891 2.165
make_images_pack 222 9.7 1.838 2.136 1.840 2.139
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.067 2.108
rebuild_ks_matrix 3 7.3 0.000 0.000 2.048 2.089
qs_ks_build_kohn_sham_matrix 3 8.3 0.015 0.068 2.048 2.089
dbcsr_data_release 9322 10.9 1.318 1.952 1.318 1.952
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=97.031000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2155.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 3.357680E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 13371.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.271 0.395 98.777 98.789
qs_energies 1 2.0 0.001 0.011 97.337 97.364
ls_scf 1 3.0 0.004 0.017 95.572 95.589
dbcsr_multiply_generic 111 6.7 0.016 0.017 78.763 79.046
ls_scf_main 1 4.0 0.000 0.002 59.135 59.140
multiply_cannon 111 7.7 0.046 0.099 51.989 56.579
density_matrix_trs4 2 5.0 0.003 0.018 53.025 53.141
multiply_cannon_loop 111 8.7 0.115 0.125 46.850 49.586
ls_scf_init_scf 1 4.0 0.005 0.023 33.154 33.156
ls_scf_init_matrix_S 1 5.0 0.000 0.002 31.618 31.702
mp_waitall_1 6369 11.0 23.107 29.325 23.107 29.325
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.129 29.141
multiply_cannon_multrec 1332 9.7 14.146 16.880 22.065 24.707
make_m2s 222 7.7 0.006 0.008 21.143 22.600
make_images 222 8.7 3.138 3.592 21.092 22.552
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 9.470 17.367
make_images_data 222 9.7 0.004 0.004 11.846 13.469
hybrid_alltoall_any 227 10.6 0.796 3.825 11.234 13.170
dbcsr_mm_accdrv_process 3641 10.4 0.216 0.403 7.561 9.107
dbcsr_mm_accdrv_process_sort 3641 11.4 7.191 8.689 7.191 8.689
mp_sum_l 807 5.4 4.120 7.855 4.120 7.855
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.132 6.127
multiply_cannon_sync_h2d 1332 9.7 5.497 6.083 5.497 6.083
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.063 5.964
mp_irecv_dv 3229 10.9 2.040 5.884 2.040 5.884
arnoldi_extremal 4 6.8 0.000 0.000 5.441 5.468
arnoldi_normal_ev 4 7.8 0.001 0.004 5.441 5.468
build_subspace 16 8.4 0.014 0.021 5.087 5.093
mp_allgather_i34 111 8.7 2.247 4.770 2.247 4.770
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.589 4.766
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.620 4.698
calculate_norms 2376 9.8 4.188 4.492 4.188 4.492
dbcsr_matrix_vector_mult 304 9.0 0.006 0.017 3.624 3.964
dbcsr_matrix_vector_mult_local 304 10.0 3.208 3.714 3.211 3.715
dbcsr_sort_data 658 11.4 3.096 3.395 3.096 3.395
ls_scf_post 1 4.0 0.006 0.022 3.278 3.299
dbcsr_special_finalize 555 9.7 0.006 0.007 2.840 3.169
dbcsr_merge_single_wm 555 10.7 0.538 0.658 2.831 3.160
ls_scf_store_result 1 5.0 0.000 0.000 2.976 3.067
ls_scf_dm_to_ks 2 5.0 0.001 0.020 2.922 2.989
dbcsr_data_release 10477 10.7 1.584 2.432 1.584 2.432
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.077 2.079
rebuild_ks_matrix 3 7.3 0.000 0.000 2.054 2.056
qs_ks_build_kohn_sham_matrix 3 8.3 0.004 0.016 2.054 2.056
dbcsr_finalize 304 7.8 0.049 0.061 1.806 1.983
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=98.789000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2704.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 4.643148E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265558.
MP_Allreduce 3049 15663.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.086 0.112 99.221 99.228
qs_energies 1 2.0 0.000 0.000 98.204 98.210
ls_scf 1 3.0 0.000 0.000 96.283 96.288
dbcsr_multiply_generic 111 6.7 0.017 0.018 77.858 78.076
ls_scf_main 1 4.0 0.000 0.000 62.042 62.043
multiply_cannon 111 7.7 0.070 0.170 55.275 60.886
density_matrix_trs4 2 5.0 0.002 0.003 54.981 55.080
multiply_cannon_loop 111 8.7 0.069 0.077 50.682 52.517
mp_waitall_1 5436 11.0 26.444 32.423 26.444 32.423
ls_scf_init_scf 1 4.0 0.000 0.000 30.662 30.668
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.461 29.493
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 27.292 27.299
multiply_cannon_multrec 444 9.7 13.925 16.194 21.035 23.654
make_m2s 222 7.7 0.004 0.005 17.695 20.250
make_images 222 8.7 3.723 4.415 17.633 20.191
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.160 16.206
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.135 15.239
make_images_data 222 9.7 0.003 0.004 9.949 12.435
hybrid_alltoall_any 227 10.6 0.790 3.772 9.774 12.319
dbcsr_mm_accdrv_process 3003 10.4 0.186 0.347 6.817 7.941
multiply_cannon_sync_h2d 444 9.7 6.500 7.892 6.500 7.892
dbcsr_mm_accdrv_process_sort 3003 11.4 6.503 7.597 6.503 7.597
mp_allgather_i34 111 8.7 2.813 6.994 2.813 6.994
arnoldi_extremal 4 6.8 0.000 0.000 5.936 5.944
arnoldi_normal_ev 4 7.8 0.001 0.005 5.936 5.944
build_subspace 16 8.4 0.015 0.020 5.532 5.547
mp_sum_l 807 5.4 2.942 5.353 2.942 5.353
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.612 4.752
dbcsr_matrix_vector_mult 304 9.0 0.007 0.016 4.263 4.440
dbcsr_matrix_vector_mult_local 304 10.0 3.739 4.218 3.741 4.220
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.017 4.029
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.673 3.864
mp_irecv_dv 1241 11.2 1.653 3.819 1.653 3.819
calculate_norms 792 9.8 3.540 3.684 3.540 3.684
ls_scf_post 1 4.0 0.000 0.000 3.578 3.584
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.417 3.525
make_images_sizes 222 9.7 0.000 0.000 1.101 3.475
mp_alltoall_i44 222 10.7 1.100 3.475 1.100 3.475
ls_scf_store_result 1 5.0 0.000 0.000 3.356 3.407
dbcsr_finalize 304 7.8 0.062 0.078 2.208 2.310
dbcsr_merge_all 275 8.9 0.473 0.531 2.054 2.142
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.988 1.989
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=99.228000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3662.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3_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 8.781820E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284111.
MP_Allreduce 3043 21950.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.101 0.191 112.205 112.249
qs_energies 1 2.0 0.000 0.000 110.713 110.727
ls_scf 1 3.0 0.000 0.000 107.823 107.836
dbcsr_multiply_generic 111 6.7 0.023 0.026 80.937 81.055
ls_scf_main 1 4.0 0.000 0.000 69.235 69.236
density_matrix_trs4 2 5.0 0.002 0.003 59.849 59.912
multiply_cannon 111 7.7 0.144 0.301 52.277 53.710
multiply_cannon_loop 111 8.7 0.067 0.069 48.968 50.276
ls_scf_init_scf 1 4.0 0.000 0.000 34.724 34.725
ls_scf_init_matrix_S 1 5.0 0.000 0.000 33.364 33.375
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 30.477 30.489
mp_waitall_1 4527 11.1 25.600 28.874 25.600 28.874
make_m2s 222 7.7 0.005 0.005 24.800 25.639
make_images 222 8.7 4.591 4.955 24.694 25.531
multiply_cannon_multrec 444 9.7 17.777 18.645 22.421 23.086
hybrid_alltoall_any 227 10.6 1.661 3.611 13.619 16.628
make_images_data 222 9.7 0.003 0.003 13.936 16.321
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 12.828 13.928
multiply_cannon_sync_h2d 444 9.7 8.844 8.898 8.844 8.898
arnoldi_extremal 4 6.8 0.000 0.000 7.958 7.970
arnoldi_normal_ev 4 7.8 0.003 0.009 7.958 7.970
build_subspace 16 8.4 0.025 0.036 7.399 7.409
dbcsr_matrix_vector_mult 304 9.0 0.009 0.025 5.935 6.098
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.282 5.533
dbcsr_matrix_vector_mult_local 304 10.0 5.136 5.468 5.139 5.470
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.154 5.304
dbcsr_mm_accdrv_process 1814 10.4 0.206 0.318 4.479 4.617
dbcsr_mm_accdrv_process_sort 1814 11.4 4.182 4.320 4.182 4.320
ls_scf_post 1 4.0 0.000 0.000 3.863 3.877
make_images_sizes 222 9.7 0.000 0.000 1.536 3.734
mp_alltoall_i44 222 10.7 1.535 3.734 1.535 3.734
ls_scf_store_result 1 5.0 0.000 0.000 3.571 3.580
calculate_norms 792 9.8 3.247 3.283 3.247 3.283
dbcsr_finalize 304 7.8 0.082 0.089 3.087 3.126
dbcsr_complete_redistribute 5 7.6 1.428 1.472 2.894 3.041
dbcsr_merge_all 275 8.9 0.890 0.915 2.873 2.911
qs_energies_init_hamiltonians 1 3.0 0.008 0.014 2.860 2.860
mp_allgather_i34 111 8.7 0.947 2.845 0.947 2.845
dbcsr_data_release 12724 10.6 2.323 2.830 2.323 2.830
matrix_ls_to_qs 2 6.0 0.000 0.000 2.531 2.695
dbcsr_sort_data 325 11.1 2.443 2.501 2.443 2.501
dbcsr_new_transposed 4 7.5 0.260 0.287 2.328 2.342
mp_sum_l 807 5.4 1.558 2.282 1.558 2.282
mp_alltoall_d11v 48 9.2 2.198 2.251 2.198 2.251
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=112.249000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6851.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 7c2cbadb1dfa2ab43dbf397fe5b72069add6d4c3
Summary: empty
Status: OK