=== 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: 644f5a76d68b492ea6ab77ab84438411b373c60e
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1,
# Cray-FFTW 3.3.8.10, COSMA 2.6.6, ELPA 2023.05.001,
# HDF5 1.14.2, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.2.2, LIBVORI 220621, LIBXSMM 1.17,
# PLUMED 2.9.0, SIRIUS 7.4.3, SPGLIB 1.16.2,
# LIBGRPP 20231215
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Last update: 15.12.2023
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2023.05.001
USE_HDF5 := 1.14.2
USE_LIBGRPP := 20231215
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.9.0
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBGRPP),)
USE_LIBGRPP := $(strip $(USE_LIBGRPP))
LIBGRPP_INC := $(INSTALL_PATH)/libgrpp-main-$(USE_LIBGRPP)/include
LIBGRPP_LIB := $(INSTALL_PATH)/libgrpp-main-$(USE_LIBGRPP)/lib
CFLAGS += -I$(LIBGRPP_INC)
DFLAGS += -D__LIBGRPP
LIBS += $(LIBGRPP_LIB)/liblibgrpp.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
endif
ifneq ($(USE_HDF5),)
USE_HDF5 := $(strip $(USE_HDF5))
HDF5_INC := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/include
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/lib
CFLAGS += -I$(HDF5_INC)
DFLAGS += -D__HDF5
LIBS += $(HDF5_LIB)/libhdf5_fortran.a
LIBS += $(HDF5_LIB)/libhdf5_hl.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/01
job id: 50922243
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/02
job id: 50922244
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/03
job id: 50922245
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/04
job id: 50922246
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/05
job id: 50922247
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/06
job id: 50922248
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/07
job id: 50922249
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/08
job id: 50922250
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/09
job id: 50922251
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/10
job id: 50922252
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/11
job id: 50922253
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/12
job id: 50922254
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/13
job id: 50922256
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/14
job id: 50922257
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/15
job id: 50922258
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/16
job id: 50922261
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/17
job id: 50922262
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/18
job id: 50922263
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/19
job id: 50922264
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/20
job id: 50922265
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/21
job id: 50922266
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/22
job id: 50922267
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/23
job id: 50922268
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/24
job id: 50922269
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/25
job id: 50922270
--- 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/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/26
job id: 50922271
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: 512 H2O (4 NVE MD steps on 64 nodes)
input file: benchmarks/QS/00512_H2O/H2O-512_md.inp
required files: []
output file: result.log
# nodes = 64
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/27
job id: 50922272
--- Point ---
name: 601
plot: h2o_512_md
regex: CP2K
label: (64n/12r/1t)
--- Point ---
name: 602
plot: h2o_512_md_mem
regex: Estimated peak process memory
label: (64n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 172669.
MP_Allreduce 424 8.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.034 133.607 133.607
farming_run 1 2.0 133.184 133.185 133.579 133.583
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.496490E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1134128.
MP_Allreduce 571 1938539.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.005 0.025 114.743 114.743
qs_energies 1 2.0 0.000 0.000 114.532 114.534
mp2_main 1 3.0 0.000 0.000 112.459 112.461
mp2_gpw_main 1 4.0 0.028 0.050 111.470 111.472
mp2_ri_gpw_compute_in 1 5.0 0.172 0.173 92.505 92.923
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.004 54.663 55.081
mp2_eri_3c_integrate_gpw 272 7.0 0.152 0.163 41.064 46.239
get_2c_integrals 1 6.0 0.008 0.009 36.831 37.670
integrate_v_rspace 273 8.0 0.440 0.455 24.667 29.628
pw_transfer 6555 10.6 0.380 0.392 26.833 27.512
fft_wrap_pw1pw2 5465 11.4 0.046 0.050 25.468 26.225
grid_integrate_task_list 273 9.0 20.566 26.003 20.566 26.003
fft_wrap_pw1pw2_100 2178 12.4 0.028 0.031 23.008 23.780
rpa_ri_compute_en 1 5.0 0.019 0.021 18.843 19.063
compute_2c_integrals 1 7.0 0.002 0.002 18.938 18.939
compute_2c_integrals_loop_lm 1 8.0 0.002 0.003 18.560 18.676
mp2_eri_2c_integrate_gpw 1 9.0 2.392 2.431 18.559 18.676
cp_fm_cholesky_decompose 12 8.2 17.815 18.627 17.815 18.627
cholesky_decomp 1 7.0 0.000 0.000 16.733 17.547
fft3d_s 5443 13.4 16.096 16.741 16.117 16.763
ao_to_mo_and_store_B_mult_1 272 7.0 10.771 15.346 10.771 15.346
calculate_wavefunction 272 8.0 5.377 5.517 12.308 12.998
rpa_num_int 1 6.0 0.000 0.001 10.595 10.595
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.586 10.591
calc_mat_Q 8 8.0 0.000 0.000 9.453 9.550
contract_S_to_Q 8 9.0 0.000 0.000 8.876 8.972
calc_potential_gpw 544 9.5 0.004 0.004 8.242 8.658
parallel_gemm_fm 14 9.1 0.000 0.000 8.454 8.549
parallel_gemm_fm_cosma 14 10.1 8.454 8.549 8.454 8.549
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.001 8.056 8.277
potential_pw2rs 545 10.0 0.107 0.109 7.460 8.155
create_integ_mat 1 6.0 0.023 0.028 7.736 7.736
collocate_single_gaussian 272 10.0 0.038 0.041 7.308 7.514
array2fm 1 7.0 0.000 0.000 6.620 7.179
pw_scatter_s 2720 13.7 4.354 4.501 4.354 4.501
pw_gather_s 2722 13.2 3.488 3.863 3.488 3.863
array2fm_buffer_send 1 8.0 2.902 3.193 2.902 3.193
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=111.471325, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2799.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 200775.
MP_Allreduce 424 9.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.037 419.765 419.767
farming_run 1 2.0 419.158 419.162 419.732 419.734
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.241158E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 757 478553.
MP_Allreduce 2021 21391.
MP_Sync 37
MP_Alltoall 77
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 0.011 0.030 217.121 217.122
qs_energies 1 2.0 0.000 0.000 216.910 216.916
scf_env_do_scf 1 3.0 0.000 0.000 114.816 114.817
qs_ks_update_qs_env 5 5.0 0.000 0.000 113.915 113.922
rebuild_ks_matrix 4 6.0 0.000 0.000 113.914 113.921
qs_ks_build_kohn_sham_matrix 4 7.0 0.054 0.061 113.914 113.921
hfx_ks_matrix 4 8.0 0.001 0.001 113.565 113.570
integrate_four_center 4 9.0 0.152 0.476 113.564 113.569
integrate_four_center_main 4 10.0 0.115 0.525 101.945 105.003
integrate_four_center_bin 269 11.0 101.830 104.994 101.830 104.994
mp2_main 1 3.0 0.000 0.000 101.804 101.810
mp2_gpw_main 1 4.0 0.049 0.069 100.947 100.956
init_scf_loop 1 4.0 0.000 0.000 96.821 96.822
mp2_ri_gpw_compute_in 1 5.0 0.064 0.065 74.042 75.343
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 53.859 55.160
mp2_eri_3c_integrate_gpw 91 7.0 0.144 0.162 41.580 46.595
integrate_v_rspace 95 8.0 0.397 0.559 28.059 32.933
pw_transfer 2240 10.6 0.144 0.165 29.540 30.035
fft_wrap_pw1pw2 1868 11.4 0.018 0.020 28.531 29.076
mp2_ri_gpw_compute_en 1 5.0 0.055 0.064 26.758 28.873
ao_to_mo_and_store_B_mult_1 91 7.0 10.604 28.666 10.604 28.666
grid_integrate_task_list 95 9.0 23.477 28.555 23.477 28.555
fft_wrap_pw1pw2_100 730 12.4 0.012 0.014 26.263 26.685
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.841 1.887 24.805 24.817
get_2c_integrals 1 6.0 0.000 0.000 20.095 20.117
compute_2c_integrals 1 7.0 0.002 0.003 19.073 19.077
compute_2c_integrals_loop_lm 1 8.0 0.001 0.001 18.716 18.948
mp2_eri_2c_integrate_gpw 1 9.0 1.733 1.857 18.715 18.947
fft3d_s 1823 13.4 18.474 18.799 18.487 18.812
scf_env_do_scf_inner_loop 4 4.0 0.000 0.001 17.993 17.993
calculate_wavefunction 91 8.0 2.008 2.041 9.616 9.856
mp2_ri_gpw_compute_en_expansio 172 7.0 0.557 0.592 8.656 9.375
potential_pw2rs 186 10.0 0.033 0.035 8.450 9.059
local_gemm 172 8.0 8.099 8.784 8.099 8.784
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.212 8.558
mp2_ri_gpw_compute_en_comm 22 7.0 0.496 0.518 7.913 8.322
calc_potential_gpw 182 9.5 0.002 0.002 7.931 8.125
collocate_single_gaussian 91 10.0 0.016 0.020 7.787 8.026
integrate_four_center_load 4 10.0 0.000 0.000 6.776 6.780
hfx_load_balance 1 11.0 0.000 0.000 6.776 6.780
mp2_ri_gpw_compute_en_ener 172 7.0 6.351 6.427 6.351 6.427
mp_sendrecv_dm3 2068 8.0 5.953 6.355 5.953 6.355
mp_sync 37 10.5 3.571 6.318 3.571 6.318
pw_gather_s 912 13.2 4.465 5.059 4.465 5.059
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=100.942416, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1500.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 453.009408E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62379.
MP_Allreduce 10329 270.
MP_Sync 530
MP_Alltoall 2083
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 0.008 0.028 56.376 56.377
qs_mol_dyn_low 1 2.0 0.003 0.005 56.145 56.153
qs_forces 11 3.9 0.004 0.007 56.081 56.082
qs_energies 11 4.9 0.002 0.006 54.518 54.529
scf_env_do_scf 11 5.9 0.000 0.001 48.528 48.528
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 46.393 46.393
qs_scf_new_mos 108 7.5 0.000 0.001 35.909 36.198
qs_scf_loop_do_ot 108 8.5 0.000 0.001 35.908 36.198
dbcsr_multiply_generic 2286 12.5 0.093 0.096 35.241 35.692
ot_scf_mini 108 9.5 0.002 0.002 34.152 34.321
multiply_cannon 2286 13.5 0.190 0.209 26.843 28.545
multiply_cannon_loop 2286 14.5 1.822 1.957 26.131 27.863
velocity_verlet 10 3.0 0.001 0.002 27.718 27.719
ot_mini 108 10.5 0.001 0.006 20.742 20.965
qs_ot_get_derivative 108 11.5 0.001 0.001 17.610 17.799
mp_waitall_1 245248 16.5 9.272 15.211 9.272 15.211
multiply_cannon_metrocomm3 54864 15.5 0.072 0.077 6.489 13.229
multiply_cannon_multrec 54864 15.5 3.618 5.698 7.593 10.860
qs_ot_get_p 119 10.4 0.001 0.001 8.487 8.791
rebuild_ks_matrix 119 8.3 0.000 0.000 8.072 8.228
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.014 8.072 8.228
mp_sum_l 7287 12.8 6.306 7.836 6.306 7.836
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.112 7.249
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 6.069 6.530
multiply_cannon_sync_h2d 54864 15.5 5.088 6.257 5.088 6.257
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 6.007 6.115
dbcsr_mm_accdrv_process 76910 16.1 1.833 2.906 3.889 5.589
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.497 5.548
init_scf_run 11 5.9 0.000 0.001 4.652 4.652
scf_env_initial_rho_setup 11 6.9 0.000 0.002 4.652 4.652
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.445 4.651
calculate_rho_elec 119 8.7 0.011 0.017 4.444 4.650
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.647 4.647
sum_up_and_integrate 119 10.3 0.001 0.003 4.601 4.607
integrate_v_rspace 119 11.3 0.002 0.003 4.590 4.598
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.414 4.414
cp_fm_redistribute_end 50 14.0 2.248 4.383 2.257 4.387
cp_fm_diag_elpa_base 50 14.0 2.123 4.273 2.127 4.280
multiply_cannon_metrocomm1 54864 15.5 0.055 0.060 2.017 3.493
apply_preconditioner_dbcsr 119 12.6 0.000 0.001 2.953 3.151
apply_single 119 13.6 0.000 0.000 2.952 3.151
calculate_dm_sparse 119 9.5 0.000 0.000 2.987 3.131
ot_diis_step 108 11.5 0.007 0.018 2.832 2.833
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.658 2.735
acc_transpose_blocks 54864 15.5 0.222 0.245 2.201 2.723
jit_kernel_multiply 13 15.8 1.994 2.593 1.994 2.593
density_rs2pw 119 9.7 0.004 0.004 2.353 2.434
calculate_first_density_matrix 1 7.0 0.001 0.010 2.356 2.361
wfi_extrapolate 11 7.9 0.001 0.001 2.217 2.218
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.211 2.213
mp_sum_d 4137 12.0 1.508 2.122 1.508 2.122
init_scf_loop 11 6.9 0.000 0.002 2.114 2.114
grid_integrate_task_list 119 12.3 2.006 2.100 2.006 2.100
potential_pw2rs 119 12.3 0.004 0.004 1.967 1.984
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.911 1.974
pw_transfer 1439 11.6 0.052 0.057 1.861 1.941
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 1.786 1.868
make_m2s 4572 13.5 0.053 0.055 1.779 1.828
make_images 4572 14.5 0.133 0.140 1.697 1.745
mp_alltoall_d11v 2130 13.8 1.478 1.719 1.478 1.719
fft3d_ps 1201 14.6 0.371 0.480 1.499 1.578
transfer_rs2pw 487 10.6 0.005 0.006 1.480 1.559
fft_wrap_pw1pw2_140 487 13.2 0.108 0.115 1.377 1.461
mp_waitany 12084 13.8 1.283 1.452 1.283 1.452
transfer_pw2rs 487 13.2 0.006 0.006 1.416 1.432
acc_transpose_blocks_sync 164592 16.5 1.196 1.431 1.196 1.431
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.405 1.427
grid_collocate_task_list 119 9.7 1.348 1.421 1.348 1.421
dbcsr_dot_sd 1205 11.9 0.048 0.059 0.841 1.223
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=56.377000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=432.181818, yerr=1.113404
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 489.185280E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10306 303.
MP_Sync 54
MP_Alltoall 2060 1394209.
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 0.007 0.023 38.346 38.346
qs_mol_dyn_low 1 2.0 0.003 0.003 38.144 38.152
qs_forces 11 3.9 0.002 0.003 38.085 38.086
qs_energies 11 4.9 0.004 0.005 36.428 36.433
scf_env_do_scf 11 5.9 0.001 0.001 31.338 31.338
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 28.982 28.982
dbcsr_multiply_generic 2286 12.5 0.100 0.103 22.071 22.400
qs_scf_new_mos 108 7.5 0.001 0.001 20.503 20.727
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.503 20.727
ot_scf_mini 108 9.5 0.002 0.003 19.589 19.746
multiply_cannon 2286 13.5 0.209 0.217 17.105 18.593
velocity_verlet 10 3.0 0.001 0.001 18.057 18.058
multiply_cannon_loop 2286 14.5 1.193 1.252 15.963 17.383
ot_mini 108 10.5 0.001 0.001 12.106 12.330
mp_waitall_1 200699 16.5 5.595 10.694 5.595 10.694
qs_ot_get_derivative 108 11.5 0.001 0.001 9.757 9.918
multiply_cannon_metrocomm3 27432 15.5 0.071 0.073 4.149 9.480
multiply_cannon_multrec 27432 15.5 1.826 4.077 6.263 9.135
rebuild_ks_matrix 119 8.3 0.000 0.000 6.675 6.811
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.675 6.811
dbcsr_mm_accdrv_process 47894 16.0 3.661 5.956 4.357 6.466
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.870 5.993
qs_ot_get_p 119 10.4 0.001 0.001 4.573 4.788
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.783 4.591
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.951 4.043
apply_single 119 13.6 0.000 0.000 2.951 4.043
init_scf_run 11 5.9 0.000 0.001 3.906 3.907
scf_env_initial_rho_setup 11 6.9 0.000 0.000 3.906 3.906
mp_sum_l 7287 12.8 1.956 3.824 1.956 3.824
sum_up_and_integrate 119 10.3 0.001 0.001 3.622 3.627
integrate_v_rspace 119 11.3 0.002 0.002 3.608 3.613
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.534 3.562
calculate_rho_elec 119 8.7 0.021 0.024 3.533 3.561
qs_ot_p2m_diag 50 11.0 0.009 0.013 2.954 2.972
make_m2s 4572 13.5 0.051 0.053 2.556 2.886
make_images 4572 14.5 0.206 0.246 2.466 2.800
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.560 2.560
calculate_first_density_matrix 1 7.0 0.000 0.000 2.427 2.429
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.294 2.377
init_scf_loop 11 6.9 0.000 0.000 2.336 2.337
ot_diis_step 108 11.5 0.010 0.011 2.302 2.303
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.253 2.254
cp_fm_redistribute_end 50 14.0 1.142 2.229 1.145 2.232
calculate_dm_sparse 119 9.5 0.000 0.001 2.115 2.191
cp_fm_diag_elpa_base 50 14.0 1.055 2.142 1.083 2.179
multiply_cannon_sync_h2d 27432 15.5 1.674 2.170 1.674 2.170
acc_transpose_blocks 27432 15.5 0.115 0.121 1.647 2.028
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.953 1.955
grid_integrate_task_list 119 12.3 1.837 1.913 1.837 1.913
density_rs2pw 119 9.7 0.004 0.004 1.838 1.901
pw_transfer 1439 11.6 0.065 0.069 1.806 1.835
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.765 1.806
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.714 1.746
jit_kernel_multiply 8 16.3 0.639 1.667 0.639 1.667
make_images_data 4572 15.5 0.048 0.055 1.215 1.517
prepare_preconditioner 11 7.9 0.000 0.000 1.475 1.502
make_preconditioner 11 8.9 0.000 0.000 1.475 1.501
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.382 1.439
wfi_extrapolate 11 7.9 0.001 0.001 1.422 1.422
fft_wrap_pw1pw2_140 487 13.2 0.127 0.132 1.375 1.407
hybrid_alltoall_any 4725 16.4 0.053 0.115 1.083 1.404
potential_pw2rs 119 12.3 0.006 0.006 1.380 1.388
fft3d_ps 1201 14.6 0.519 0.576 1.345 1.371
grid_collocate_task_list 119 9.7 1.286 1.348 1.286 1.348
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.307 1.316
mp_alltoall_d11v 2130 13.8 1.174 1.296 1.174 1.296
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.194 1.240
transfer_rs2pw 487 10.6 0.005 0.005 1.023 1.097
mp_sum_d 4137 12.0 0.522 0.997 0.522 0.997
mp_allgather_i34 2286 14.5 0.565 0.951 0.565 0.951
acc_transpose_blocks_kernels 27432 16.5 0.188 0.277 0.693 0.942
acc_transpose_blocks_sync 82296 16.5 0.814 0.940 0.814 0.940
qs_energies_init_hamiltonians 11 5.9 0.000 0.000 0.925 0.925
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.861 0.874
transfer_pw2rs 487 13.2 0.004 0.006 0.860 0.865
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.701 0.768
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=38.346000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.545455, yerr=1.724879
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 522.473472E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63490.
MP_Allreduce 10155 305.
MP_Sync 54
MP_Alltoall 1821 1607811.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.028 31.933 31.934
qs_mol_dyn_low 1 2.0 0.003 0.003 31.742 31.750
qs_forces 11 3.9 0.002 0.003 31.685 31.685
qs_energies 11 4.9 0.001 0.002 30.145 30.147
scf_env_do_scf 11 5.9 0.001 0.001 25.390 25.390
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 23.005 23.006
dbcsr_multiply_generic 2286 12.5 0.095 0.097 17.097 17.166
qs_scf_new_mos 108 7.5 0.001 0.001 15.375 15.389
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.375 15.388
velocity_verlet 10 3.0 0.001 0.001 15.012 15.013
ot_scf_mini 108 9.5 0.002 0.002 14.616 14.629
multiply_cannon 2286 13.5 0.192 0.197 13.939 14.562
multiply_cannon_loop 2286 14.5 0.863 0.921 13.173 13.804
ot_mini 108 10.5 0.001 0.001 9.072 9.087
qs_ot_get_derivative 108 11.5 0.001 0.001 7.586 7.600
multiply_cannon_multrec 18288 15.5 1.878 2.825 7.111 7.420
rebuild_ks_matrix 119 8.3 0.000 0.000 5.945 5.960
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.014 5.944 5.960
dbcsr_mm_accdrv_process 38222 16.0 5.077 5.805 5.140 5.870
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.234 5.249
mp_waitall_1 158411 16.6 2.662 3.769 2.662 3.769
init_scf_run 11 5.9 0.000 0.001 3.586 3.586
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.585 3.586
sum_up_and_integrate 119 10.3 0.001 0.001 3.516 3.522
integrate_v_rspace 119 11.3 0.002 0.003 3.503 3.511
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.936 3.477
qs_ot_get_p 119 10.4 0.001 0.001 3.298 3.320
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.232 3.239
calculate_rho_elec 119 8.7 0.031 0.031 3.232 3.239
multiply_cannon_metrocomm3 18288 15.5 0.048 0.049 1.519 2.487
calculate_first_density_matrix 1 7.0 0.000 0.000 2.372 2.372
init_scf_loop 11 6.9 0.000 0.000 2.369 2.370
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.944 2.343
apply_single 119 13.6 0.000 0.000 1.943 2.343
qs_ot_p2m_diag 50 11.0 0.012 0.012 2.203 2.210
calculate_dm_sparse 119 9.5 0.000 0.001 1.970 1.977
density_rs2pw 119 9.7 0.004 0.004 1.797 1.946
make_m2s 4572 13.5 0.043 0.045 1.784 1.927
grid_integrate_task_list 119 12.3 1.803 1.916 1.803 1.916
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.907 1.907
pw_transfer 1439 11.6 0.065 0.068 1.844 1.853
make_images 4572 14.5 0.192 0.205 1.698 1.840
acc_transpose_blocks 18288 15.5 0.079 0.081 1.687 1.800
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.751 1.762
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.736 1.737
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.708 1.712
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.661 1.670
cp_fm_diag_elpa_base 50 14.0 1.638 1.651 1.659 1.669
prepare_preconditioner 11 7.9 0.000 0.000 1.628 1.630
make_preconditioner 11 8.9 0.000 0.000 1.628 1.630
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.592 1.599
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.492 1.574
ot_diis_step 108 11.5 0.011 0.011 1.469 1.469
mp_sum_l 7287 12.8 0.982 1.447 0.982 1.447
fft_wrap_pw1pw2_140 487 13.2 0.179 0.183 1.420 1.430
grid_collocate_task_list 119 9.7 1.233 1.361 1.233 1.361
potential_pw2rs 119 12.3 0.007 0.008 1.320 1.325
fft3d_ps 1201 14.6 0.545 0.566 1.306 1.317
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.306 1.310
multiply_cannon_sync_h2d 18288 15.5 1.047 1.211 1.047 1.211
wfi_extrapolate 11 7.9 0.001 0.001 1.162 1.162
transfer_rs2pw 487 10.6 0.005 0.005 0.946 1.102
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.945 0.962
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 0.934 0.935
make_images_data 4572 15.5 0.047 0.051 0.765 0.930
acc_transpose_blocks_kernels 18288 16.5 0.217 0.225 0.851 0.881
hybrid_alltoall_any 4725 16.4 0.058 0.118 0.663 0.856
acc_transpose_blocks_sync 54864 16.5 0.738 0.818 0.738 0.818
mp_alltoall_d11v 2130 13.8 0.657 0.811 0.657 0.811
transfer_pw2rs 487 13.2 0.004 0.004 0.793 0.795
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.779 0.781
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.668 0.740
cp_fm_cholesky_invert 11 10.9 0.687 0.690 0.687 0.690
mp_alltoall_z22v 1201 16.6 0.612 0.681 0.612 0.681
transfer_rs2pw_140 130 11.5 0.115 0.117 0.500 0.657
jit_kernel_transpose 5 15.6 0.634 0.657 0.634 0.657
mp_waitany 9880 13.7 0.511 0.640 0.511 0.640
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=31.934000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.727273, yerr=2.048987
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 559.980544E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63489.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.029 36.862 36.863
qs_mol_dyn_low 1 2.0 0.003 0.003 36.677 36.686
qs_forces 11 3.9 0.002 0.002 36.618 36.619
qs_energies 11 4.9 0.002 0.002 34.896 34.902
scf_env_do_scf 11 5.9 0.001 0.001 29.790 29.791
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 26.331 26.331
dbcsr_multiply_generic 2286 12.5 0.102 0.105 19.979 20.108
velocity_verlet 10 3.0 0.001 0.001 18.792 18.793
qs_scf_new_mos 108 7.5 0.001 0.001 18.094 18.150
qs_scf_loop_do_ot 108 8.5 0.001 0.001 18.094 18.150
ot_scf_mini 108 9.5 0.002 0.003 17.057 17.124
multiply_cannon 2286 13.5 0.218 0.226 16.173 17.024
multiply_cannon_loop 2286 14.5 1.488 1.603 15.194 15.872
ot_mini 108 10.5 0.001 0.001 10.384 10.445
multiply_cannon_multrec 27432 15.5 2.444 3.140 9.081 9.378
qs_ot_get_derivative 108 11.5 0.001 0.001 8.521 8.588
dbcsr_mm_accdrv_process 47916 15.9 6.171 7.677 6.538 7.881
rebuild_ks_matrix 119 8.3 0.000 0.000 6.420 6.471
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.420 6.471
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.689 5.735
qs_ot_get_p 119 10.4 0.001 0.001 3.803 3.892
init_scf_run 11 5.9 0.000 0.001 3.758 3.758
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.758 3.758
sum_up_and_integrate 119 10.3 0.001 0.001 3.561 3.570
integrate_v_rspace 119 11.3 0.002 0.003 3.550 3.559
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.241 3.556
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.461 3.494
calculate_rho_elec 119 8.7 0.040 0.046 3.460 3.493
init_scf_loop 11 6.9 0.000 0.000 3.439 3.439
prepare_preconditioner 11 7.9 0.000 0.000 2.580 2.588
make_preconditioner 11 8.9 0.000 0.000 2.580 2.588
acc_transpose_blocks 27432 15.5 0.119 0.122 2.335 2.541
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.186 2.514
make_m2s 4572 13.5 0.054 0.057 2.373 2.479
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.357 2.390
make_images 4572 14.5 0.274 0.333 2.265 2.368
mp_waitall_1 137007 16.6 1.896 2.362 1.896 2.362
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.334 2.345
calculate_dm_sparse 119 9.5 0.000 0.001 2.249 2.308
calculate_first_density_matrix 1 7.0 0.000 0.000 2.284 2.285
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.030 2.228
apply_single 119 13.6 0.000 0.000 2.030 2.228
pw_transfer 1439 11.6 0.065 0.069 2.052 2.089
density_rs2pw 119 9.7 0.003 0.004 1.896 2.005
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.960 2.000
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.923 1.923
grid_integrate_task_list 119 12.3 1.837 1.902 1.837 1.902
ot_diis_step 108 11.5 0.012 0.012 1.822 1.823
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.821 1.822
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.654 1.663
cp_fm_diag_elpa_base 50 14.0 1.617 1.635 1.651 1.661
fft_wrap_pw1pw2_140 487 13.2 0.227 0.233 1.598 1.639
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.621 1.637
acc_transpose_blocks_sync 82296 16.5 1.359 1.566 1.359 1.566
fft3d_ps 1201 14.6 0.590 0.651 1.449 1.474
mp_sum_l 7287 12.8 1.009 1.452 1.009 1.452
wfi_extrapolate 11 7.9 0.001 0.001 1.420 1.420
potential_pw2rs 119 12.3 0.008 0.009 1.360 1.364
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.340 1.350
multiply_cannon_metrocomm3 27432 15.5 0.039 0.041 0.849 1.324
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.302 1.323
grid_collocate_task_list 119 9.7 1.249 1.320 1.249 1.320
cp_fm_upper_to_full 72 14.2 0.830 1.179 0.830 1.179
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.081 1.082
dbcsr_complete_redistribute 329 12.2 0.117 0.144 0.794 1.073
jit_kernel_multiply 6 16.4 0.302 1.068 0.302 1.068
transfer_rs2pw 487 10.6 0.004 0.005 0.937 1.048
make_images_data 4572 15.5 0.049 0.054 0.880 1.012
hybrid_alltoall_any 4725 16.4 0.066 0.154 0.752 0.945
mp_alltoall_d11v 2130 13.8 0.783 0.900 0.783 0.900
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.801 0.876
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.587 0.859
acc_transpose_blocks_kernels 27432 16.5 0.270 0.278 0.828 0.844
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.822 0.826
transfer_pw2rs 487 13.2 0.004 0.004 0.774 0.777
cp_fm_cholesky_invert 11 10.9 0.773 0.776 0.773 0.776
mp_alltoall_z22v 1201 16.6 0.741 0.769 0.741 0.769
mp_alltoall_i22 627 13.8 0.447 0.754 0.447 0.754
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=36.863000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=531.363636, yerr=2.772355
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 634.769408E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63488.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.030 28.421 28.422
qs_mol_dyn_low 1 2.0 0.003 0.003 28.257 28.265
qs_forces 11 3.9 0.002 0.003 28.199 28.200
qs_energies 11 4.9 0.001 0.001 26.499 26.502
scf_env_do_scf 11 5.9 0.000 0.001 21.759 21.759
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.343 19.344
velocity_verlet 10 3.0 0.001 0.002 14.518 14.520
dbcsr_multiply_generic 2286 12.5 0.093 0.096 13.180 13.270
qs_scf_new_mos 108 7.5 0.001 0.001 11.732 11.759
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.732 11.758
multiply_cannon 2286 13.5 0.223 0.228 10.665 11.168
ot_scf_mini 108 9.5 0.002 0.002 11.033 11.057
multiply_cannon_loop 2286 14.5 0.645 0.667 9.759 9.964
ot_mini 108 10.5 0.001 0.001 6.431 6.459
multiply_cannon_multrec 9144 15.5 1.679 2.033 6.063 6.451
rebuild_ks_matrix 119 8.3 0.000 0.000 5.747 5.772
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.747 5.772
qs_ot_get_derivative 108 11.5 0.001 0.001 5.154 5.179
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.112 5.133
dbcsr_mm_accdrv_process 12550 15.8 3.651 4.303 4.272 4.367
sum_up_and_integrate 119 10.3 0.001 0.001 3.404 3.408
integrate_v_rspace 119 11.3 0.003 0.003 3.394 3.398
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.333 3.340
calculate_rho_elec 119 8.7 0.060 0.061 3.333 3.339
init_scf_run 11 5.9 0.000 0.001 3.307 3.307
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.307 3.307
qs_ot_get_p 119 10.4 0.001 0.001 2.736 2.769
init_scf_loop 11 6.9 0.000 0.000 2.397 2.398
calculate_first_density_matrix 1 7.0 0.000 0.000 2.182 2.184
pw_transfer 1439 11.6 0.065 0.068 2.087 2.098
mp_waitall_1 115863 16.7 1.458 2.046 1.458 2.046
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.993 2.006
grid_integrate_task_list 119 12.3 1.868 1.922 1.868 1.922
make_m2s 4572 13.5 0.033 0.034 1.741 1.900
calculate_dm_sparse 119 9.5 0.000 0.000 1.857 1.875
make_images 4572 14.5 0.269 0.301 1.652 1.810
density_rs2pw 119 9.7 0.003 0.003 1.723 1.796
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.781 1.783
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.683 1.705
prepare_preconditioner 11 7.9 0.000 0.000 1.678 1.683
make_preconditioner 11 8.9 0.000 0.000 1.678 1.683
fft_wrap_pw1pw2_140 487 13.2 0.325 0.332 1.662 1.674
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.573 1.598
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.542 1.542
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.528 1.529
acc_transpose_blocks 9144 15.5 0.043 0.044 1.435 1.484
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.375 1.383
grid_collocate_task_list 119 9.7 1.296 1.366 1.296 1.366
fft3d_ps 1201 14.6 0.650 0.664 1.345 1.355
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.313 1.327
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.267 1.275
cp_fm_diag_elpa_base 50 14.0 1.240 1.258 1.266 1.274
ot_diis_step 108 11.5 0.013 0.013 1.266 1.266
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.253 1.258
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.219 1.257
apply_single 119 13.6 0.000 0.000 1.219 1.257
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.226 1.227
potential_pw2rs 119 12.3 0.010 0.011 1.221 1.223
wfi_extrapolate 11 7.9 0.001 0.001 1.074 1.074
jit_kernel_multiply 6 15.4 0.581 1.065 0.581 1.065
hybrid_alltoall_any 4725 16.4 0.065 0.179 0.747 0.981
make_images_data 4572 15.5 0.043 0.047 0.751 0.931
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.875 0.928
mp_alltoall_d11v 2130 13.8 0.754 0.880 0.754 0.880
cp_fm_cholesky_invert 11 10.9 0.834 0.836 0.834 0.836
multiply_cannon_metrocomm3 9144 15.5 0.020 0.021 0.442 0.824
transfer_rs2pw 487 10.6 0.004 0.004 0.743 0.820
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.805 0.811
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.755 0.758
acc_transpose_blocks_sync 27432 16.5 0.722 0.757 0.722 0.757
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.672 0.719
acc_transpose_blocks_kernels 9144 16.5 0.118 0.120 0.655 0.668
multiply_cannon_metrocomm1 9144 15.5 0.023 0.023 0.314 0.659
transfer_pw2rs 487 13.2 0.003 0.004 0.631 0.633
mp_allgather_i34 2286 14.5 0.223 0.630 0.223 0.630
mp_alltoall_z22v 1201 16.6 0.568 0.603 0.568 0.603
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=28.422000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=603.909091, yerr=5.230236
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 769.425408E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63723.
MP_Allreduce 10154 429.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.030 43.182 43.183
qs_mol_dyn_low 1 2.0 0.003 0.003 42.985 42.993
qs_forces 11 3.9 0.002 0.002 42.921 42.922
qs_energies 11 4.9 0.001 0.002 40.884 40.889
scf_env_do_scf 11 5.9 0.001 0.001 35.161 35.161
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 27.406 27.408
velocity_verlet 10 3.0 0.001 0.002 24.247 24.253
dbcsr_multiply_generic 2286 12.5 0.102 0.103 19.098 19.271
qs_scf_new_mos 108 7.5 0.001 0.001 17.558 17.659
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.557 17.658
ot_scf_mini 108 9.5 0.002 0.002 16.396 16.499
multiply_cannon 2286 13.5 0.297 0.304 15.081 15.943
multiply_cannon_loop 2286 14.5 0.873 0.892 13.814 14.710
ot_mini 108 10.5 0.001 0.001 9.933 10.049
multiply_cannon_multrec 9144 15.5 3.411 4.713 8.794 8.884
qs_ot_get_derivative 108 11.5 0.001 0.001 7.890 7.993
init_scf_loop 11 6.9 0.000 0.000 7.727 7.731
rebuild_ks_matrix 119 8.3 0.000 0.000 7.380 7.523
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.379 7.522
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.648 6.778
prepare_preconditioner 11 7.9 0.000 0.000 6.740 6.753
make_preconditioner 11 8.9 0.000 0.000 6.740 6.753
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.297 6.636
dbcsr_mm_accdrv_process 12550 15.8 4.585 6.485 5.245 6.549
cp_fm_upper_to_full 72 14.2 3.164 4.557 3.164 4.557
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.359 4.419
calculate_rho_elec 119 8.7 0.117 0.120 4.358 4.419
sum_up_and_integrate 119 10.3 0.001 0.001 3.956 3.961
integrate_v_rspace 119 11.3 0.003 0.003 3.946 3.951
qs_ot_get_p 119 10.4 0.001 0.001 3.532 3.669
init_scf_run 11 5.9 0.000 0.001 3.665 3.665
scf_env_initial_rho_setup 11 6.9 0.000 0.001 3.665 3.665
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.862 3.271
mp_waitall_1 94719 16.7 2.322 3.257 2.322 3.257
pw_transfer 1439 11.6 0.069 0.069 3.077 3.083
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.978 2.985
dbcsr_complete_redistribute 329 12.2 0.282 0.286 1.975 2.820
fft_wrap_pw1pw2_140 487 13.2 0.650 0.657 2.531 2.541
make_m2s 4572 13.5 0.037 0.037 2.396 2.539
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.203 2.520
apply_single 119 13.6 0.000 0.000 2.202 2.520
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.674 2.498
make_images 4572 14.5 0.356 0.387 2.276 2.417
calculate_dm_sparse 119 9.5 0.000 0.000 2.292 2.350
density_rs2pw 119 9.7 0.003 0.003 2.292 2.328
mp_alltoall_i22 627 13.8 1.420 2.256 1.420 2.256
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.437 2.255
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.356 2.189
calculate_first_density_matrix 1 7.0 0.000 0.000 2.118 2.182
grid_integrate_task_list 119 12.3 2.096 2.166 2.096 2.166
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.091 2.143
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.111 2.113
ot_diis_step 108 11.5 0.014 0.014 2.017 2.017
fft3d_ps 1201 14.6 0.869 0.888 1.899 1.903
acc_transpose_blocks 9144 15.5 0.045 0.046 1.821 1.852
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.808 1.809
qs_energies_init_hamiltonians 11 5.9 0.000 0.000 1.777 1.779
mp_sum_l 7287 12.8 1.075 1.761 1.075 1.761
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.751 1.752
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.556 1.596
grid_collocate_task_list 119 9.7 1.533 1.577 1.533 1.577
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.451 1.451
cp_fm_diag_elpa_base 50 14.0 1.304 1.359 1.449 1.449
wfi_extrapolate 11 7.9 0.001 0.001 1.422 1.422
potential_pw2rs 119 12.3 0.014 0.014 1.418 1.419
cp_fm_cholesky_invert 11 10.9 1.383 1.387 1.383 1.387
mp_alltoall_d11v 2130 13.8 1.291 1.376 1.291 1.376
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.294 1.355
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.316 1.334
hybrid_alltoall_any 4725 16.4 0.091 0.151 1.081 1.301
make_images_data 4572 15.5 0.047 0.049 1.007 1.197
acc_transpose_blocks_sync 27432 16.5 1.115 1.141 1.115 1.141
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.108 1.121
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.996 1.049
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.942 0.955
qs_create_task_list 11 7.9 0.000 0.001 0.927 0.937
generate_qs_task_list 11 8.9 0.369 0.388 0.927 0.937
mp_alltoall_z22v 1201 16.6 0.894 0.919 0.894 0.919
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=43.183000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=723.090909, yerr=12.929561
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 503.042048E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66212.
MP_Allreduce 9776 488.
MP_Sync 52
MP_Alltoall 1938 2166702.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.024 82.897 82.898
qs_mol_dyn_low 1 2.0 0.003 0.003 82.652 82.661
qs_forces 11 3.9 0.003 0.003 82.585 82.586
qs_energies 11 4.9 0.001 0.002 79.684 79.696
scf_env_do_scf 11 5.9 0.000 0.001 70.917 70.919
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 65.523 65.524
dbcsr_multiply_generic 2055 12.4 0.106 0.110 52.064 52.308
qs_scf_new_mos 99 7.5 0.000 0.001 48.507 48.599
qs_scf_loop_do_ot 99 8.5 0.000 0.001 48.506 48.599
ot_scf_mini 99 9.5 0.002 0.002 46.052 46.154
velocity_verlet 10 3.0 0.001 0.001 44.334 44.335
multiply_cannon 2055 13.4 0.179 0.186 43.009 43.873
multiply_cannon_loop 2055 14.4 1.787 1.848 42.072 42.891
ot_mini 99 10.5 0.001 0.001 27.532 27.637
qs_ot_get_derivative 99 11.5 0.001 0.001 20.630 20.728
multiply_cannon_multrec 49320 15.4 11.331 11.979 17.320 17.948
rebuild_ks_matrix 110 8.3 0.000 0.000 14.450 14.575
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.011 14.450 14.574
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.664 12.776
mp_waitall_1 220248 16.4 11.400 12.603 11.400 12.603
multiply_cannon_sync_h2d 49320 15.4 9.530 10.033 9.530 10.033
qs_ot_get_p 110 10.4 0.001 0.001 9.872 9.984
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 8.160 8.700
multiply_cannon_metrocomm3 49320 15.4 0.081 0.084 6.984 8.148
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.228 7.753
apply_single 110 13.6 0.000 0.001 7.228 7.752
sum_up_and_integrate 110 10.3 0.002 0.003 6.948 6.961
integrate_v_rspace 110 11.3 0.002 0.003 6.922 6.942
init_scf_run 11 5.9 0.000 0.001 6.701 6.701
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.701 6.701
ot_diis_step 99 11.5 0.006 0.006 6.627 6.627
qs_ot_p2m_diag 48 11.0 0.012 0.019 6.540 6.578
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.342 6.498
calculate_rho_elec 110 8.6 0.020 0.024 6.342 6.497
dbcsr_mm_accdrv_process 87628 16.1 3.050 3.171 5.859 6.138
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.675 5.675
mp_sum_l 6594 12.7 4.619 5.671 4.619 5.671
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.492 5.556
init_scf_loop 11 6.9 0.000 0.000 5.363 5.363
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.077 5.104
cp_fm_diag_elpa_base 48 14.0 5.063 5.089 5.074 5.102
make_m2s 4110 13.4 0.060 0.067 3.972 4.082
wfi_extrapolate 11 7.9 0.001 0.001 4.059 4.059
make_images 4110 14.4 0.178 0.193 3.876 3.992
multiply_cannon_metrocomm1 49320 15.4 0.063 0.065 2.611 3.961
calculate_dm_sparse 110 9.5 0.001 0.001 3.818 3.931
density_rs2pw 110 9.6 0.004 0.004 3.332 3.537
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.532 3.536
grid_integrate_task_list 110 12.3 3.253 3.419 3.253 3.419
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.334 3.408
prepare_preconditioner 11 7.9 0.000 0.000 3.212 3.236
make_preconditioner 11 8.9 0.000 0.000 3.212 3.236
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.189 3.231
pw_transfer 1331 11.6 0.055 0.065 3.106 3.182
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.017 3.097
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.021 3.062
fft_wrap_pw1pw2_140 451 13.1 0.309 0.335 2.556 2.652
acc_transpose_blocks 49320 15.4 0.211 0.216 2.529 2.593
potential_pw2rs 110 12.3 0.005 0.006 2.555 2.567
mp_alltoall_d11v 2046 13.8 2.086 2.563 2.086 2.563
jit_kernel_multiply 13 15.9 2.531 2.545 2.531 2.545
calculate_first_density_matrix 1 7.0 0.000 0.000 2.541 2.544
fft3d_ps 1111 14.6 0.799 0.889 2.309 2.363
grid_collocate_task_list 110 9.6 2.160 2.292 2.160 2.292
mp_waitany 14300 13.8 1.834 2.088 1.834 2.088
transfer_rs2pw 451 10.6 0.005 0.006 1.874 2.067
mp_sum_d 3891 11.9 1.497 1.970 1.497 1.970
make_images_data 4110 15.4 0.044 0.048 1.772 1.937
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.916 1.929
hybrid_alltoall_any 4261 16.3 0.084 0.482 1.542 1.813
cp_fm_cholesky_invert 11 10.9 1.755 1.758 1.755 1.758
transfer_pw2rs 451 13.1 0.006 0.007 1.709 1.717
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.642 1.668
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=82.898000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=477.000000, yerr=2.828427
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 592.195584E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66430.
MP_Allreduce 9775 566.
MP_Sync 52
MP_Alltoall 1717 3462318.
MP_SendRecv 10340 26400.
MP_ISendRecv 10340 26400.
MP_Wait 22352
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.028 68.187 68.188
qs_mol_dyn_low 1 2.0 0.003 0.003 67.952 67.962
qs_forces 11 3.9 0.003 0.003 67.886 67.887
qs_energies 11 4.9 0.002 0.004 64.572 64.575
scf_env_do_scf 11 5.9 0.001 0.001 56.157 56.160
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 48.717 48.717
dbcsr_multiply_generic 2055 12.4 0.118 0.122 37.765 37.925
velocity_verlet 10 3.0 0.001 0.002 35.805 35.806
qs_scf_new_mos 99 7.5 0.001 0.001 32.975 33.107
qs_scf_loop_do_ot 99 8.5 0.001 0.001 32.974 33.106
multiply_cannon 2055 13.4 0.223 0.242 31.198 32.546
ot_scf_mini 99 9.5 0.003 0.003 31.305 31.445
multiply_cannon_loop 2055 14.4 1.160 1.190 29.933 30.949
ot_mini 99 10.5 0.001 0.001 18.305 18.444
multiply_cannon_multrec 24660 15.4 6.954 8.801 14.030 15.601
rebuild_ks_matrix 110 8.3 0.000 0.000 13.445 13.543
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.017 13.445 13.542
qs_ot_get_derivative 99 11.5 0.001 0.001 12.536 12.674
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.824 11.906
mp_waitall_1 176588 16.5 7.425 10.195 7.425 10.195
multiply_cannon_metrocomm3 24660 15.4 0.073 0.075 5.092 7.855
multiply_cannon_sync_h2d 24660 15.4 6.335 7.474 6.335 7.474
init_scf_loop 11 6.9 0.000 0.001 7.404 7.405
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.438 7.219
apply_single 110 13.6 0.000 0.001 6.438 7.219
dbcsr_mm_accdrv_process 52282 16.1 5.582 6.495 6.907 7.215
qs_ot_get_p 110 10.4 0.001 0.001 6.353 6.502
sum_up_and_integrate 110 10.3 0.002 0.004 6.312 6.325
integrate_v_rspace 110 11.3 0.002 0.002 6.285 6.299
init_scf_run 11 5.9 0.000 0.001 6.038 6.038
scf_env_initial_rho_setup 11 6.9 0.000 0.001 6.037 6.038
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.846 5.859
calculate_rho_elec 110 8.6 0.039 0.047 5.845 5.859
ot_diis_step 99 11.5 0.010 0.010 5.722 5.722
prepare_preconditioner 11 7.9 0.000 0.000 5.409 5.431
make_preconditioner 11 8.9 0.000 0.001 5.409 5.431
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.717 5.417
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.009 5.165
make_m2s 4110 13.4 0.056 0.059 4.194 4.664
make_images 4110 14.4 0.408 0.460 4.083 4.550
qs_ot_p2m_diag 48 11.0 0.029 0.044 4.475 4.495
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.013 4.013
pw_transfer 1331 11.6 0.066 0.071 3.521 3.656
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.414 3.552
wfi_extrapolate 11 7.9 0.001 0.001 3.495 3.495
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.458 3.472
cp_fm_diag_elpa_base 48 14.0 3.412 3.429 3.455 3.469
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.267 3.333
density_rs2pw 110 9.6 0.004 0.004 3.083 3.330
grid_integrate_task_list 110 12.3 3.158 3.295 3.158 3.295
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.222 3.223
fft_wrap_pw1pw2_140 451 13.1 0.357 0.376 2.934 3.076
calculate_dm_sparse 110 9.5 0.001 0.001 2.972 3.002
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.889 2.913
make_images_data 4110 15.4 0.049 0.053 2.274 2.737
hybrid_alltoall_any 4261 16.3 0.106 0.450 1.990 2.725
fft3d_ps 1111 14.6 1.115 1.344 2.506 2.654
cp_fm_cholesky_invert 11 10.9 2.524 2.530 2.524 2.530
calculate_first_density_matrix 1 7.0 0.000 0.000 2.446 2.449
mp_sum_l 6594 12.7 1.737 2.424 1.737 2.424
grid_collocate_task_list 110 9.6 2.174 2.309 2.174 2.309
potential_pw2rs 110 12.3 0.008 0.009 2.196 2.205
jit_kernel_multiply 11 16.4 0.958 2.130 0.958 2.130
acc_transpose_blocks 24660 15.4 0.116 0.119 1.951 1.984
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.960 1.979
mp_alltoall_d11v 2046 13.8 1.723 1.942 1.723 1.942
qs_energies_init_hamiltonians 11 5.9 0.000 0.002 1.865 1.865
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.777 1.789
multiply_cannon_metrocomm4 22605 15.4 0.075 0.079 0.779 1.710
mp_allgather_i34 2055 14.4 0.573 1.593 0.573 1.593
mp_irecv_dv 57340 16.2 0.654 1.592 0.654 1.592
transfer_rs2pw 451 10.6 0.006 0.007 1.399 1.586
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.538 1.551
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.391 1.493
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=68.188000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=560.272727, yerr=7.020631
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 666.009600E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66421.
MP_Allreduce 9774 562.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.008 0.027 59.541 59.542
qs_mol_dyn_low 1 2.0 0.003 0.003 59.289 59.300
qs_forces 11 3.9 0.003 0.003 59.215 59.216
qs_energies 11 4.9 0.001 0.002 55.996 55.999
scf_env_do_scf 11 5.9 0.000 0.001 48.211 48.211
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 39.676 39.676
velocity_verlet 10 3.0 0.001 0.001 32.560 32.562
dbcsr_multiply_generic 2055 12.4 0.109 0.113 28.673 28.954
qs_scf_new_mos 99 7.5 0.001 0.001 25.120 25.238
qs_scf_loop_do_ot 99 8.5 0.001 0.001 25.120 25.237
ot_scf_mini 99 9.5 0.002 0.003 23.872 23.999
multiply_cannon 2055 13.4 0.212 0.219 22.463 23.828
multiply_cannon_loop 2055 14.4 0.818 0.855 21.289 22.291
ot_mini 99 10.5 0.001 0.001 13.692 13.820
rebuild_ks_matrix 110 8.3 0.000 0.000 12.146 12.277
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.145 12.277
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.639 10.758
multiply_cannon_multrec 16440 15.4 3.550 4.537 9.734 10.615
mp_waitall_1 139946 16.5 7.017 9.644 7.017 9.644
qs_ot_get_derivative 99 11.5 0.001 0.001 9.205 9.336
init_scf_loop 11 6.9 0.000 0.000 8.502 8.502
multiply_cannon_metrocomm3 16440 15.4 0.045 0.046 4.478 7.080
prepare_preconditioner 11 7.9 0.000 0.000 6.760 6.776
make_preconditioner 11 8.9 0.000 0.000 6.760 6.776
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.050 6.427
sum_up_and_integrate 110 10.3 0.001 0.002 6.235 6.250
integrate_v_rspace 110 11.3 0.003 0.003 6.209 6.225
dbcsr_mm_accdrv_process 34862 16.1 5.324 5.705 6.031 6.220
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.689 5.698
calculate_rho_elec 110 8.6 0.059 0.059 5.688 5.697
qs_ot_get_p 110 10.4 0.001 0.001 5.338 5.508
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.997 5.479
apply_single 110 13.6 0.000 0.000 4.997 5.479
init_scf_run 11 5.9 0.000 0.001 5.399 5.400
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.399 5.399
make_m2s 4110 13.4 0.049 0.051 4.146 4.536
ot_diis_step 99 11.5 0.011 0.011 4.458 4.458
make_images 4110 14.4 0.400 0.523 4.030 4.417
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.148 3.790
multiply_cannon_sync_h2d 16440 15.4 3.195 3.750 3.195 3.750
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.739 3.742
pw_transfer 1331 11.6 0.065 0.073 3.595 3.606
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.488 3.502
grid_integrate_task_list 110 12.3 3.192 3.388 3.192 3.388
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.371 3.371
density_rs2pw 110 9.6 0.004 0.004 2.923 3.114
fft_wrap_pw1pw2_140 451 13.1 0.461 0.470 3.033 3.054
wfi_extrapolate 11 7.9 0.001 0.001 2.987 2.987
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.840 2.850
cp_fm_diag_elpa_base 48 14.0 2.775 2.804 2.839 2.848
make_images_data 4110 15.4 0.047 0.051 2.361 2.839
hybrid_alltoall_any 4261 16.3 0.108 0.385 2.112 2.800
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.758 2.760
cp_fm_cholesky_invert 11 10.9 2.594 2.600 2.594 2.600
calculate_dm_sparse 110 9.5 0.001 0.001 2.520 2.555
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.412 2.484
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.416 2.475
multiply_cannon_metrocomm4 14385 15.4 0.048 0.052 0.902 2.466
fft3d_ps 1111 14.6 1.119 1.130 2.404 2.419
grid_collocate_task_list 110 9.6 2.222 2.394 2.222 2.394
mp_irecv_dv 48980 15.7 0.827 2.329 0.827 2.329
calculate_first_density_matrix 1 7.0 0.000 0.000 2.323 2.325
mp_alltoall_d11v 2046 13.8 1.808 2.186 1.808 2.186
potential_pw2rs 110 12.3 0.011 0.011 2.092 2.099
mp_sum_l 6594 12.7 1.437 2.031 1.437 2.031
qs_energies_init_hamiltonians 11 5.9 0.000 0.000 1.945 1.946
dbcsr_complete_redistribute 325 12.2 0.322 0.356 1.474 1.939
cp_fm_upper_to_full 70 14.2 1.410 1.790 1.410 1.790
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.699 1.714
acc_transpose_blocks 16440 15.4 0.079 0.083 1.570 1.633
mp_allgather_i34 2055 14.4 0.485 1.597 0.485 1.597
cp_fm_cholesky_decompose 22 10.9 1.565 1.590 1.565 1.590
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.016 1.469
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.456 1.468
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.368 1.465
transfer_rs2pw 451 10.6 0.005 0.006 1.230 1.430
mp_waitany 17072 13.8 1.116 1.337 1.116 1.337
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.273 1.281
rs_gather_matrices 110 12.3 0.137 0.150 0.873 1.233
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=59.542000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=631.545455, yerr=8.183838
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 735.457280E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66419.
MP_Allreduce 9774 603.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.028 65.284 65.285
qs_mol_dyn_low 1 2.0 0.003 0.003 65.065 65.074
qs_forces 11 3.9 0.003 0.003 65.001 65.002
qs_energies 11 4.9 0.001 0.002 61.612 61.615
scf_env_do_scf 11 5.9 0.001 0.001 53.447 53.450
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 41.835 41.835
velocity_verlet 10 3.0 0.001 0.001 37.206 37.208
dbcsr_multiply_generic 2055 12.4 0.117 0.121 30.612 30.782
qs_scf_new_mos 99 7.5 0.001 0.001 27.267 27.368
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.266 27.367
ot_scf_mini 99 9.5 0.002 0.003 25.588 25.686
multiply_cannon 2055 13.4 0.237 0.256 23.555 24.799
multiply_cannon_loop 2055 14.4 1.434 1.492 22.119 22.672
ot_mini 99 10.5 0.001 0.001 14.738 14.856
multiply_cannon_multrec 24660 15.4 4.104 6.750 13.018 14.213
rebuild_ks_matrix 110 8.3 0.000 0.000 11.953 12.052
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.015 11.952 12.052
init_scf_loop 11 6.9 0.000 0.000 11.574 11.575
qs_ot_get_derivative 99 11.5 0.001 0.001 10.553 10.647
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.556 10.642
prepare_preconditioner 11 7.9 0.000 0.000 9.844 9.859
make_preconditioner 11 8.9 0.000 0.000 9.844 9.859
dbcsr_mm_accdrv_process 52304 16.0 7.722 9.125 8.757 9.743
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.047 9.544
sum_up_and_integrate 110 10.3 0.001 0.002 6.177 6.191
integrate_v_rspace 110 11.3 0.003 0.003 6.151 6.164
mp_waitall_1 121746 16.5 4.181 6.076 4.181 6.076
make_m2s 4110 13.4 0.059 0.061 5.452 5.849
qs_ot_get_p 110 10.4 0.001 0.001 5.685 5.832
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.807 5.816
calculate_rho_elec 110 8.6 0.077 0.081 5.807 5.815
make_images 4110 14.4 0.578 0.698 5.311 5.704
init_scf_run 11 5.9 0.000 0.001 5.432 5.433
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.432 5.433
cp_fm_upper_to_full 70 14.2 3.408 4.956 3.408 4.956
ot_diis_step 99 11.5 0.011 0.011 4.148 4.148
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.060 4.145
apply_single 110 13.6 0.000 0.000 4.060 4.145
qs_ot_p2m_diag 48 11.0 0.055 0.064 3.844 3.860
dbcsr_complete_redistribute 325 12.2 0.417 0.455 2.649 3.803
pw_transfer 1331 11.6 0.065 0.074 3.726 3.758
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.619 3.654
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.519 3.564
grid_integrate_task_list 110 12.3 3.282 3.469 3.282 3.469
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.319 3.372
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.285 3.285
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.127 3.273
fft_wrap_pw1pw2_140 451 13.1 0.531 0.545 3.129 3.168
hybrid_alltoall_any 4261 16.3 0.123 0.459 2.288 3.159
multiply_cannon_metrocomm3 24660 15.4 0.037 0.038 1.333 3.113
make_images_data 4110 15.4 0.049 0.054 2.742 3.111
density_rs2pw 110 9.6 0.004 0.004 2.917 3.036
calculate_dm_sparse 110 9.5 0.001 0.001 2.976 3.016
wfi_extrapolate 11 7.9 0.001 0.001 2.987 2.987
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.788 2.922
mp_alltoall_i22 605 13.7 1.690 2.887 1.690 2.887
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.780 2.781
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.764 2.773
cp_fm_diag_elpa_base 48 14.0 2.617 2.673 2.761 2.770
cp_fm_cholesky_invert 11 10.9 2.590 2.598 2.590 2.598
acc_transpose_blocks 24660 15.4 0.112 0.114 2.423 2.552
multiply_cannon_sync_h2d 24660 15.4 2.376 2.498 2.376 2.498
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.429 2.461
fft3d_ps 1111 14.6 1.152 1.192 2.445 2.458
grid_collocate_task_list 110 9.6 2.265 2.377 2.265 2.377
calculate_first_density_matrix 1 7.0 0.000 0.000 2.351 2.354
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 2.240 2.241
potential_pw2rs 110 12.3 0.013 0.013 1.996 1.999
mp_alltoall_d11v 2046 13.8 1.744 1.917 1.744 1.917
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.791 1.825
mp_allgather_i34 2055 14.4 0.524 1.726 0.524 1.726
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.617 1.713
cp_fm_cholesky_decompose 22 10.9 1.617 1.661 1.617 1.661
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.635 1.647
multiply_cannon_metrocomm4 20550 15.4 0.060 0.062 0.843 1.563
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.538 1.555
mp_sum_l 6594 12.7 0.913 1.550 0.913 1.550
acc_transpose_blocks_sync 73980 16.4 1.421 1.541 1.421 1.541
mp_irecv_dv 62702 16.1 0.740 1.480 0.740 1.480
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=65.285000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=694.818182, yerr=8.663593
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 863.412224E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66417.
MP_Allreduce 9774 644.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.032 54.828 54.829
qs_mol_dyn_low 1 2.0 0.003 0.003 54.555 54.564
qs_forces 11 3.9 0.003 0.003 54.491 54.492
qs_energies 11 4.9 0.001 0.001 50.848 50.854
scf_env_do_scf 11 5.9 0.000 0.001 42.625 42.625
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 34.967 34.968
velocity_verlet 10 3.0 0.001 0.001 31.075 31.078
dbcsr_multiply_generic 2055 12.4 0.105 0.108 23.073 23.276
qs_scf_new_mos 99 7.5 0.001 0.001 20.409 20.459
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.408 20.458
ot_scf_mini 99 9.5 0.002 0.002 19.160 19.176
multiply_cannon 2055 13.4 0.238 0.249 17.588 19.066
multiply_cannon_loop 2055 14.4 0.605 0.629 16.316 16.708
rebuild_ks_matrix 110 8.3 0.000 0.000 11.624 11.635
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.624 11.635
ot_mini 99 10.5 0.001 0.001 10.625 10.635
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.333 10.344
multiply_cannon_multrec 8220 15.4 3.160 4.339 7.591 8.469
init_scf_loop 11 6.9 0.000 0.000 7.611 7.612
mp_waitall_1 103326 16.6 5.724 7.197 5.724 7.197
qs_ot_get_derivative 99 11.5 0.001 0.001 6.948 6.964
sum_up_and_integrate 110 10.3 0.001 0.002 6.183 6.194
integrate_v_rspace 110 11.3 0.003 0.003 6.157 6.168
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.994 6.010
calculate_rho_elec 110 8.6 0.113 0.113 5.994 6.009
prepare_preconditioner 11 7.9 0.000 0.000 5.973 5.976
make_preconditioner 11 8.9 0.000 0.000 5.973 5.976
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.552 5.625
dbcsr_mm_accdrv_process 17442 15.9 3.037 4.143 4.292 5.230
init_scf_run 11 5.9 0.000 0.001 5.097 5.097
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.097 5.097
qs_ot_get_p 110 10.4 0.001 0.001 4.595 4.610
make_m2s 4110 13.4 0.037 0.039 4.168 4.453
multiply_cannon_metrocomm3 8220 15.4 0.018 0.019 2.973 4.326
make_images 4110 14.4 0.649 0.713 4.038 4.324
pw_transfer 1331 11.6 0.066 0.070 4.053 4.060
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.945 3.956
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.727 3.773
apply_single 110 13.6 0.000 0.000 3.726 3.773
ot_diis_step 99 11.5 0.012 0.012 3.655 3.655
grid_integrate_task_list 110 12.3 3.380 3.497 3.380 3.497
fft_wrap_pw1pw2_140 451 13.1 0.718 0.731 3.429 3.444
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.253 3.257
density_rs2pw 110 9.6 0.004 0.004 3.001 3.092
cp_dbcsr_syevd 48 12.0 0.003 0.003 2.914 2.915
cp_fm_cholesky_invert 11 10.9 2.825 2.829 2.825 2.829
hybrid_alltoall_any 4261 16.3 0.198 0.839 2.284 2.797
wfi_extrapolate 11 7.9 0.001 0.001 2.748 2.748
qs_energies_init_hamiltonians 11 5.9 0.000 0.000 2.662 2.663
make_images_data 4110 15.4 0.041 0.047 2.240 2.623
calculate_dm_sparse 110 9.5 0.001 0.001 2.495 2.533
grid_collocate_task_list 110 9.6 2.370 2.527 2.370 2.527
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.516 2.517
fft3d_ps 1111 14.6 1.290 1.299 2.489 2.503
multiply_cannon_sync_h2d 8220 15.4 2.381 2.448 2.381 2.448
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.367 2.374
cp_fm_diag_elpa_base 48 14.0 2.312 2.334 2.365 2.373
calculate_first_density_matrix 1 7.0 0.000 0.000 2.251 2.253
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.143 2.167
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.018 2.041
potential_pw2rs 110 12.3 0.015 0.016 1.993 1.994
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.789 1.992
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.888 1.901
mp_alltoall_d11v 2046 13.8 1.697 1.890 1.697 1.890
cp_fm_cholesky_decompose 22 10.9 1.667 1.682 1.667 1.682
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.619 1.622
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.491 1.602
mp_allgather_i34 2055 14.4 0.439 1.535 0.439 1.535
dbcsr_complete_redistribute 325 12.2 0.555 0.579 1.437 1.534
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.450 1.461
acc_transpose_blocks 8220 15.4 0.041 0.041 1.384 1.420
jit_kernel_multiply 8 15.7 0.935 1.355 0.935 1.355
qs_create_task_list 11 7.9 0.000 0.001 1.206 1.301
generate_qs_task_list 11 8.9 0.374 0.442 1.206 1.301
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.226 1.254
transfer_rs2pw 451 10.6 0.005 0.005 1.096 1.218
mp_waitany 9240 13.8 1.044 1.156 1.044 1.156
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=54.829000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=807.090909, yerr=12.702573
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 1.381372E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67098.
MP_Allreduce 9752 812.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.013 0.032 87.278 87.279
qs_mol_dyn_low 1 2.0 0.003 0.003 86.993 87.002
qs_forces 11 3.9 0.003 0.003 86.927 86.927
qs_energies 11 4.9 0.002 0.002 82.740 82.742
scf_env_do_scf 11 5.9 0.000 0.001 72.824 72.825
velocity_verlet 10 3.0 0.001 0.002 55.770 55.777
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 44.586 44.587
dbcsr_multiply_generic 2055 12.4 0.121 0.125 29.360 29.585
init_scf_loop 11 6.9 0.000 0.000 28.169 28.171
qs_scf_new_mos 99 7.5 0.001 0.001 26.704 26.739
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.704 26.738
prepare_preconditioner 11 7.9 0.000 0.000 26.152 26.159
make_preconditioner 11 8.9 0.000 0.000 26.152 26.159
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.391 25.633
ot_scf_mini 99 9.5 0.002 0.002 24.911 24.936
multiply_cannon 2055 13.4 0.331 0.355 22.376 23.018
multiply_cannon_loop 2055 14.4 0.824 0.838 20.606 21.006
cp_fm_upper_to_full 70 14.2 12.768 18.246 12.768 18.246
ot_mini 99 10.5 0.001 0.001 14.168 14.196
rebuild_ks_matrix 110 8.3 0.000 0.001 13.958 13.996
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.014 13.957 13.995
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.623 12.658
dbcsr_complete_redistribute 325 12.2 1.011 1.035 7.294 10.487
multiply_cannon_multrec 8220 15.4 4.082 4.215 9.754 10.123
qs_ot_get_derivative 99 11.5 0.001 0.001 9.616 9.643
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.294 9.494
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.746 8.914
mp_alltoall_i22 605 13.7 5.370 8.585 5.370 8.585
mp_waitall_1 84994 16.7 7.397 8.298 7.397 8.298
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.472 7.514
calculate_rho_elec 110 8.6 0.222 0.223 7.472 7.514
sum_up_and_integrate 110 10.3 0.002 0.002 6.870 6.887
integrate_v_rspace 110 11.3 0.004 0.004 6.842 6.859
pw_transfer 1331 11.6 0.075 0.075 5.794 5.799
make_m2s 4110 13.4 0.042 0.043 5.298 5.785
dbcsr_mm_accdrv_process 11614 15.7 3.845 4.092 5.522 5.769
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 5.677 5.682
init_scf_run 11 5.9 0.000 0.001 5.648 5.648
scf_env_initial_rho_setup 11 6.9 0.000 0.001 5.648 5.648
make_images 4110 14.4 0.886 0.944 5.109 5.597
qs_ot_get_p 110 10.4 0.001 0.001 5.475 5.510
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.706 5.175
apply_single 110 13.6 0.000 0.000 4.706 5.175
cp_fm_cholesky_invert 11 10.9 5.152 5.156 5.152 5.156
multiply_cannon_metrocomm3 8220 15.4 0.019 0.019 4.834 5.098
fft_wrap_pw1pw2_140 451 13.1 1.366 1.371 5.000 5.005
ot_diis_step 99 11.5 0.015 0.015 4.536 4.536
qs_ot_p2m_diag 48 11.0 0.151 0.156 3.939 3.946
density_rs2pw 110 9.6 0.004 0.004 3.929 3.943
grid_integrate_task_list 110 12.3 3.685 3.733 3.685 3.733
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 3.665 3.666
hybrid_alltoall_any 4261 16.3 0.264 0.563 2.764 3.493
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.973 3.428
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.424 3.425
fft3d_ps 1111 14.6 1.886 1.891 3.381 3.388
make_images_data 4110 15.4 0.046 0.049 2.745 3.350
wfi_extrapolate 11 7.9 0.001 0.001 3.318 3.318
calculate_dm_sparse 110 9.5 0.001 0.001 3.155 3.179
multiply_cannon_sync_h2d 8220 15.4 3.126 3.147 3.126 3.147
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.849 2.849
cp_fm_diag_elpa_base 48 14.0 2.316 2.507 2.847 2.847
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.839 2.841
grid_collocate_task_list 110 9.6 2.673 2.694 2.673 2.694
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.625 2.642
potential_pw2rs 110 12.3 0.021 0.022 2.406 2.408
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.352 2.362
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.180 2.242
calculate_first_density_matrix 1 7.0 0.000 0.000 2.213 2.214
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.095 2.167
mp_alltoall_d11v 2046 13.8 1.993 2.032 1.993 2.032
cp_fm_cholesky_decompose 22 10.9 1.953 1.973 1.953 1.973
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.931 1.935
qs_create_task_list 11 7.9 0.000 0.001 1.881 1.928
generate_qs_task_list 11 8.9 0.734 0.785 1.880 1.928
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.720 1.751
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=87.279000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1249.090909, yerr=49.157362
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528903135232 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 631.996416E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175955383376
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 57505.
MP_Allreduce 11129 795.
MP_Sync 87
MP_Alltoall 2226 3090018.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.028 206.737 206.738
qs_mol_dyn_low 1 2.0 0.003 0.003 206.307 206.321
qs_forces 11 3.9 0.005 0.005 206.223 206.224
qs_energies 11 4.9 0.001 0.002 200.585 200.624
scf_env_do_scf 11 5.9 0.001 0.001 184.295 184.299
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 164.088 164.090
dbcsr_multiply_generic 2507 12.6 0.176 0.183 126.336 126.670
qs_scf_new_mos 117 7.6 0.001 0.001 124.961 125.209
qs_scf_loop_do_ot 117 8.6 0.001 0.001 124.960 125.208
velocity_verlet 10 3.0 0.001 0.002 123.317 123.318
ot_scf_mini 117 9.6 0.003 0.003 118.198 118.471
multiply_cannon 2507 13.6 0.236 0.245 101.921 103.260
multiply_cannon_loop 2507 14.6 2.409 2.459 99.798 100.982
ot_mini 117 10.6 0.001 0.001 66.763 66.990
multiply_cannon_multrec 60168 15.6 31.742 33.500 41.532 43.355
qs_ot_get_derivative 117 11.6 0.001 0.001 41.930 42.163
rebuild_ks_matrix 128 8.3 0.001 0.001 33.343 33.530
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 33.342 33.529
mp_waitall_1 267128 16.5 29.294 32.781 29.294 32.781
qs_ks_update_qs_env 128 7.6 0.001 0.001 29.937 30.131
qs_ot_get_p 128 10.4 0.001 0.001 29.050 29.263
multiply_cannon_sync_h2d 60168 15.6 26.249 28.169 26.249 28.169
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.367 24.980
apply_single 128 13.6 0.001 0.001 24.366 24.980
ot_diis_step 117 11.6 0.007 0.008 24.547 24.548
qs_ot_p2m_diag 83 11.4 0.079 0.091 22.460 22.521
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.259 20.448
init_scf_loop 11 6.9 0.000 0.000 20.134 20.136
cp_dbcsr_syevd 83 12.4 0.005 0.005 19.593 19.593
multiply_cannon_metrocomm3 60168 15.6 0.117 0.123 16.501 18.250
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.578 16.611
cp_fm_diag_elpa_base 83 14.4 16.511 16.550 16.574 16.608
prepare_preconditioner 11 7.9 0.000 0.000 15.595 15.642
make_preconditioner 11 8.9 0.000 0.000 15.595 15.642
make_full_inverse_cholesky 11 9.9 0.000 0.000 14.845 15.009
make_m2s 5014 13.6 0.105 0.114 13.762 14.147
make_images 5014 14.6 0.397 0.417 13.578 13.967
sum_up_and_integrate 128 10.3 0.002 0.005 13.843 13.856
integrate_v_rspace 128 11.3 0.003 0.004 13.784 13.802
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.024 13.205
calculate_rho_elec 128 8.7 0.045 0.064 13.023 13.204
init_scf_run 11 5.9 0.000 0.001 12.147 12.147
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.147 12.147
mp_sum_l 7950 12.9 9.393 10.485 9.393 10.485
dbcsr_mm_accdrv_process 124484 16.2 4.727 4.875 9.357 9.857
wfi_extrapolate 11 7.9 0.001 0.001 8.991 8.991
cp_fm_cholesky_invert 11 10.9 8.871 8.879 8.871 8.879
calculate_dm_sparse 128 9.5 0.001 0.001 8.581 8.664
qs_ot_get_orbitals 117 10.6 0.001 0.001 8.258 8.385
multiply_cannon_metrocomm1 60168 15.6 0.091 0.096 6.418 8.355
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.159 8.312
pw_transfer 1547 11.6 0.074 0.086 7.514 7.689
make_images_data 5014 15.6 0.067 0.073 6.735 7.629
grid_integrate_task_list 128 12.3 7.076 7.504 7.076 7.504
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 7.312 7.489
density_rs2pw 128 9.7 0.006 0.007 6.637 7.357
hybrid_alltoall_any 5200 16.5 0.293 2.244 5.902 7.344
fft_wrap_pw1pw2_140 523 13.2 0.856 0.897 6.439 6.612
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.592 6.601
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.745 5.852
fft3d_ps 1291 14.7 2.195 2.859 5.320 5.658
mp_alltoall_d11v 2415 14.1 4.270 5.454 4.270 5.454
grid_collocate_task_list 128 9.7 4.837 5.308 4.837 5.308
cp_fm_cholesky_decompose 22 10.9 4.566 4.579 4.566 4.579
potential_pw2rs 128 12.3 0.009 0.010 4.366 4.383
mp_sum_d 4481 12.1 3.472 4.180 3.472 4.180
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=206.738000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=599.000000, yerr=6.862282
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410023282688 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444707676160 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796579E+12 0.0% 0.0% 100.0%
flops max/rank 2.166472E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500928 0.0% 0.0% 100.0%
number of processed stacks 5925696 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1131.6
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 840.192000E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2385600
MPI messages size (bytes):
total size 4.069300E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.705776E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70188 2295595008
32768 < size <= 131072 716032 54973693952
131072 < size <= 4194304 1363760 1386318135296
4194304 < size <= 16777216 153648 1453843137296
16777216 < size 67056 1171888537600
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4003 58199.
MP_Allreduce 11086 960.
MP_Sync 86
MP_Alltoall 1955 4586287.
MP_SendRecv 11938 47072.
MP_ISendRecv 11938 47072.
MP_Wait 25718
MP_ISend 11660 212488.
MP_IRecv 11660 212488.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.027 185.964 185.965
qs_mol_dyn_low 1 2.0 0.003 0.003 185.588 185.602
qs_forces 11 3.9 0.004 0.004 185.507 185.508
qs_energies 11 4.9 0.001 0.002 178.780 178.791
scf_env_do_scf 11 5.9 0.001 0.001 162.666 162.676
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 130.352 130.354
velocity_verlet 10 3.0 0.001 0.001 116.727 116.729
dbcsr_multiply_generic 2485 12.5 0.185 0.191 96.372 97.615
qs_scf_new_mos 116 7.6 0.001 0.001 92.431 92.972
qs_scf_loop_do_ot 116 8.6 0.001 0.001 92.430 92.971
ot_scf_mini 116 9.6 0.004 0.004 87.673 88.353
multiply_cannon 2485 13.5 0.498 0.560 76.679 81.141
multiply_cannon_loop 2485 14.5 1.552 1.626 73.329 76.182
ot_mini 116 10.6 0.001 0.001 49.555 50.151
mp_waitall_1 212858 16.6 23.553 39.122 23.553 39.122
multiply_cannon_multrec 29820 15.5 20.868 25.917 31.426 36.680
rebuild_ks_matrix 127 8.3 0.001 0.001 31.868 32.522
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.019 31.868 32.521
init_scf_loop 11 6.9 0.000 0.000 32.228 32.229
qs_ks_update_qs_env 127 7.6 0.001 0.001 28.629 29.222
multiply_cannon_metrocomm3 29820 15.5 0.098 0.104 15.478 28.927
qs_ot_get_derivative 116 11.6 0.001 0.002 27.862 28.522
prepare_preconditioner 11 7.9 0.000 0.000 27.934 28.026
make_preconditioner 11 8.9 0.000 0.000 27.934 28.026
make_full_inverse_cholesky 11 9.9 0.000 0.000 26.629 27.200
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 21.872 22.978
apply_single 127 13.6 0.001 0.001 21.872 22.978
ot_diis_step 116 11.6 0.014 0.015 21.519 21.521
qs_ot_get_p 127 10.4 0.001 0.001 20.296 21.075
multiply_cannon_sync_h2d 29820 15.5 17.900 20.182 17.900 20.182
cp_fm_cholesky_invert 11 10.9 16.262 16.274 16.262 16.274
make_m2s 4970 13.5 0.086 0.092 13.853 15.546
qs_ot_p2m_diag 82 11.4 0.186 0.214 15.481 15.515
make_images 4970 14.5 1.160 1.362 13.639 15.331
cp_dbcsr_syevd 82 12.4 0.005 0.006 14.304 14.304
sum_up_and_integrate 127 10.3 0.002 0.004 13.735 13.764
integrate_v_rspace 127 11.3 0.003 0.003 13.675 13.705
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.109 13.145
calculate_rho_elec 127 8.7 0.086 0.103 13.108 13.144
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 10.967 11.457
init_scf_run 11 5.9 0.000 0.001 11.324 11.326
scf_env_initial_rho_setup 11 6.9 0.000 0.001 11.324 11.326
cp_fm_diag_elpa 82 13.4 0.000 0.000 11.221 11.253
cp_fm_diag_elpa_base 82 14.4 10.974 11.083 11.216 11.243
multiply_cannon_metrocomm4 27335 15.5 0.100 0.114 3.788 10.936
dbcsr_mm_accdrv_process 61726 16.2 5.487 6.172 10.016 10.592
mp_irecv_dv 68888 16.3 3.590 10.545 3.590 10.545
make_images_data 4970 15.5 0.067 0.078 8.143 10.310
hybrid_alltoall_any 5155 16.4 0.346 1.530 6.972 9.604
pw_transfer 1535 11.6 0.084 0.098 8.390 8.456
wfi_extrapolate 11 7.9 0.001 0.001 8.289 8.289
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 8.168 8.232
grid_integrate_task_list 127 12.3 7.106 7.429 7.106 7.429
density_rs2pw 127 9.7 0.006 0.006 6.882 7.381
fft_wrap_pw1pw2_140 519 13.2 0.926 0.954 7.261 7.328
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.249 7.101
cp_fm_cholesky_decompose 22 10.9 6.812 6.889 6.812 6.889
calculate_dm_sparse 127 9.5 0.001 0.001 6.409 6.549
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.052 6.061
mp_sum_l 7884 12.9 4.059 5.876 4.059 5.876
fft3d_ps 1281 14.7 2.800 2.979 5.745 5.797
qs_ot_get_orbitals 116 10.6 0.001 0.001 5.294 5.360
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.182 5.336
grid_collocate_task_list 127 9.7 4.967 5.312 4.967 5.312
mp_allgather_i34 2485 14.5 1.878 4.910 1.878 4.910
mp_alltoall_d11v 2401 14.1 4.083 4.856 4.083 4.856
potential_pw2rs 127 12.3 0.015 0.017 4.307 4.318
mp_sum_d 4454 12.1 2.550 3.967 2.550 3.967
dbcsr_complete_redistribute 393 12.7 0.760 0.835 3.042 3.880
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=185.965000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=800.818182, yerr=2.917601
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022121472 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444702699520 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796573E+12 0.0% 0.0% 100.0%
flops max/rank 2.906045E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499744 0.0% 0.0% 100.0%
number of processed stacks 3951168 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1697.1
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 939.302912E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1033760
MPI messages size (bytes):
total size 2.695213E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.607194E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 279168 36591108096
131072 < size <= 4194304 654272 987691483136
4194304 < size <= 16777216 65184 925172769472
16777216 < size 28448 745747251200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4005 58164.
MP_Allreduce 11090 999.
MP_Sync 86
MP_Alltoall 1700 9383497.
MP_SendRecv 7874 75008.
MP_ISendRecv 7874 75008.
MP_Wait 21654
MP_ISend 11660 275234.
MP_IRecv 11660 275234.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.028 173.029 173.030
qs_mol_dyn_low 1 2.0 0.003 0.003 172.649 172.663
qs_forces 11 3.9 0.004 0.004 172.265 172.266
qs_energies 11 4.9 0.001 0.002 165.743 165.753
scf_env_do_scf 11 5.9 0.004 0.009 150.342 150.351
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 115.778 115.779
velocity_verlet 10 3.0 0.001 0.002 110.826 110.839
dbcsr_multiply_generic 2485 12.5 0.181 0.184 81.270 82.505
qs_scf_new_mos 116 7.6 0.001 0.001 79.156 79.495
qs_scf_loop_do_ot 116 8.6 0.001 0.001 79.155 79.494
ot_scf_mini 116 9.6 0.003 0.004 74.940 75.306
multiply_cannon 2485 13.5 0.507 0.529 61.762 66.763
multiply_cannon_loop 2485 14.5 1.123 1.188 58.603 61.697
ot_mini 116 10.6 0.001 0.001 42.197 42.577
init_scf_loop 11 6.9 0.000 0.000 34.463 34.467
mp_waitall_1 169034 16.6 24.065 33.728 24.065 33.728
prepare_preconditioner 11 7.9 0.000 0.000 30.483 30.525
make_preconditioner 11 8.9 0.000 0.000 30.483 30.525
rebuild_ks_matrix 127 8.3 0.001 0.001 29.862 30.357
qs_ks_build_kohn_sham_matrix 127 9.3 0.015 0.017 29.861 30.356
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.132 29.600
qs_ks_update_qs_env 127 7.6 0.001 0.001 26.881 27.332
multiply_cannon_multrec 19880 15.5 13.050 15.836 22.365 25.203
multiply_cannon_metrocomm3 19880 15.5 0.061 0.064 14.837 24.517
qs_ot_get_derivative 116 11.6 0.001 0.002 22.884 23.263
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 19.409 20.404
apply_single 127 13.6 0.001 0.001 19.409 20.404
ot_diis_step 116 11.6 0.018 0.029 19.206 19.206
qs_ot_get_p 127 10.4 0.001 0.001 18.222 18.650
make_m2s 4970 13.5 0.077 0.082 14.365 15.849
make_images 4970 14.5 1.141 1.234 14.132 15.613
multiply_cannon_sync_h2d 19880 15.5 13.541 15.205 13.541 15.205
cp_fm_cholesky_invert 11 10.9 14.077 14.086 14.077 14.086
qs_ot_p2m_diag 82 11.4 0.262 0.269 13.874 13.880
sum_up_and_integrate 127 10.3 0.002 0.003 13.760 13.786
integrate_v_rspace 127 11.3 0.003 0.003 13.701 13.725
qs_rho_update_rho_low 127 7.7 0.001 0.001 13.417 13.455
calculate_rho_elec 127 8.7 0.128 0.142 13.417 13.454
cp_dbcsr_syevd 82 12.4 0.005 0.005 12.839 12.839
make_images_data 4970 15.5 0.064 0.074 8.593 10.515
init_scf_run 11 5.9 0.000 0.001 10.378 10.378
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.378 10.378
hybrid_alltoall_any 5155 16.4 0.443 2.007 7.556 10.067
cp_fm_diag_elpa 82 13.4 0.000 0.000 9.721 9.739
cp_fm_diag_elpa_base 82 14.4 9.316 9.452 9.718 9.735
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.016 9.267
multiply_cannon_metrocomm4 17395 15.5 0.065 0.074 3.410 9.256
mp_irecv_dv 49801 16.2 3.283 8.995 3.283 8.995
dbcsr_mm_accdrv_process 41158 16.2 5.639 5.950 8.777 8.958
pw_transfer 1535 11.6 0.084 0.103 8.664 8.764
fft_wrap_pw1pw2 1281 12.7 0.010 0.011 8.442 8.549
grid_integrate_task_list 127 12.3 7.275 7.711 7.275 7.711
fft_wrap_pw1pw2_140 519 13.2 1.052 1.087 7.519 7.631
wfi_extrapolate 11 7.9 0.001 0.001 7.406 7.406
cp_fm_upper_to_full 104 14.8 5.780 7.303 5.780 7.303
cp_fm_cholesky_decompose 22 10.9 7.208 7.238 7.208 7.238
density_rs2pw 127 9.7 0.005 0.006 6.795 7.195
dbcsr_complete_redistribute 393 12.7 1.162 1.199 4.522 6.276
calculate_dm_sparse 127 9.5 0.001 0.001 5.808 5.918
fft3d_ps 1281 14.7 2.800 3.045 5.777 5.884
grid_collocate_task_list 127 9.7 5.116 5.468 5.116 5.468
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.431 5.440
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.591 5.282
copy_fm_to_dbcsr 208 11.6 0.001 0.002 3.388 5.138
mp_sum_l 7884 12.9 3.320 4.945 3.320 4.945
mp_alltoall_d11v 2401 14.1 4.237 4.699 4.237 4.699
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.535 4.691
mp_allgather_i34 2485 14.5 1.655 4.579 1.655 4.579
potential_pw2rs 127 12.3 0.020 0.022 4.252 4.260
transfer_fm_to_dbcsr 11 9.9 0.019 0.025 2.332 4.055
mp_alltoall_i22 712 14.1 2.011 3.925 2.011 3.925
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.873 3.907
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 3.739 3.739
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.473 3.516
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=173.030000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=887.636364, yerr=6.718668
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 4.353788E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5977344 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.136693E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1143192
MPI messages size (bytes):
total size 2.023815E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.770319E+06
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 0 0
8192 < size <= 32768 396 8650752
32768 < size <= 131072 319024 36042702848
131072 < size <= 4194304 715736 785529176064
4194304 < size <= 16777216 70320 665379241840
16777216 < size 30720 536870912000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58348.
MP_Allreduce 11057 1083.
MP_Sync 87
MP_Alltoall 1712 12503084.
MP_SendRecv 5888 75008.
MP_ISendRecv 5888 75008.
MP_Wait 22442
MP_ISend 14952 244818.
MP_IRecv 14952 244818.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.031 186.813 186.815
qs_mol_dyn_low 1 2.0 0.003 0.003 186.465 186.479
qs_forces 11 3.9 0.004 0.004 186.339 186.348
qs_energies 11 4.9 0.002 0.002 179.246 179.256
scf_env_do_scf 11 5.9 0.001 0.001 162.526 162.542
velocity_verlet 10 3.0 0.001 0.001 124.241 124.245
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 116.269 116.270
dbcsr_multiply_generic 2507 12.6 0.191 0.196 80.494 81.281
qs_scf_new_mos 117 7.6 0.001 0.001 79.989 80.338
qs_scf_loop_do_ot 117 8.6 0.001 0.001 79.988 80.337
ot_scf_mini 117 9.6 0.003 0.003 75.434 75.764
multiply_cannon 2507 13.6 0.555 0.590 55.776 58.682
multiply_cannon_loop 2507 14.6 1.869 1.924 52.189 53.958
init_scf_loop 11 6.9 0.000 0.000 46.137 46.137
ot_mini 117 10.6 0.001 0.001 42.841 43.154
prepare_preconditioner 11 7.9 0.000 0.000 42.026 42.051
make_preconditioner 11 8.9 0.000 0.000 42.026 42.051
make_full_inverse_cholesky 11 9.9 0.012 0.022 35.566 40.645
multiply_cannon_multrec 30084 15.6 13.590 19.148 26.511 31.736
rebuild_ks_matrix 128 8.3 0.001 0.001 29.340 29.653
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.019 29.339 29.653
mp_waitall_1 147882 16.7 17.101 27.025 17.101 27.025
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.484 26.772
qs_ot_get_derivative 117 11.6 0.001 0.002 23.260 23.585
make_m2s 5014 13.6 0.093 0.097 20.267 21.162
make_images 5014 14.6 1.954 2.303 19.957 20.853
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 18.940 19.493
apply_single 128 13.6 0.001 0.001 18.939 19.493
ot_diis_step 117 11.6 0.018 0.019 19.447 19.449
qs_ot_get_p 128 10.4 0.001 0.001 18.709 19.065
cp_fm_upper_to_full 105 14.8 11.664 17.139 11.664 17.139
cp_fm_cholesky_invert 11 10.9 15.774 15.783 15.774 15.783
multiply_cannon_metrocomm3 30084 15.6 0.048 0.051 6.284 14.842
qs_ot_p2m_diag 83 11.4 0.343 0.390 14.542 14.592
sum_up_and_integrate 128 10.3 0.002 0.003 13.936 13.955
integrate_v_rspace 128 11.3 0.003 0.003 13.876 13.899
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.726 13.773
calculate_rho_elec 128 8.7 0.170 0.185 13.726 13.772
cp_dbcsr_syevd 83 12.4 0.005 0.006 13.129 13.132
dbcsr_mm_accdrv_process 62264 16.2 8.404 9.232 12.491 13.037
dbcsr_complete_redistribute 395 12.7 1.492 1.628 9.191 12.989
make_images_data 5014 15.6 0.066 0.073 10.913 12.282
multiply_cannon_sync_h2d 30084 15.6 10.477 11.712 10.477 11.712
copy_fm_to_dbcsr 209 11.7 0.001 0.002 7.848 11.649
hybrid_alltoall_any 5200 16.5 0.528 2.203 9.639 11.157
init_scf_run 11 5.9 0.000 0.001 10.757 10.759
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.757 10.759
transfer_fm_to_dbcsr 11 9.9 0.001 0.006 6.437 10.162
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.759 9.989
cp_fm_diag_elpa 83 13.4 0.000 0.000 9.956 9.970
cp_fm_diag_elpa_base 83 14.4 8.985 9.312 9.951 9.963
mp_alltoall_i22 716 14.1 5.679 9.633 5.679 9.633
pw_transfer 1547 11.6 0.085 0.102 9.158 9.240
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.932 9.021
fft_wrap_pw1pw2_140 523 13.2 1.247 1.290 7.982 8.085
grid_integrate_task_list 128 12.3 7.514 7.886 7.514 7.886
wfi_extrapolate 11 7.9 0.001 0.001 7.757 7.757
multiply_cannon_metrocomm4 25070 15.6 0.080 0.091 2.823 7.690
cp_fm_cholesky_decompose 22 10.9 7.426 7.504 7.426 7.504
mp_irecv_dv 76098 16.2 2.671 7.407 2.671 7.407
density_rs2pw 128 9.7 0.005 0.006 6.766 7.226
calculate_dm_sparse 128 9.5 0.001 0.001 6.249 6.352
fft3d_ps 1291 14.7 3.000 3.078 5.982 6.053
grid_collocate_task_list 128 9.7 5.281 5.808 5.281 5.808
mp_alltoall_d11v 2415 14.1 4.950 5.598 4.950 5.598
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.480 5.533
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.475 4.588
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.389 4.475
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 4.456 4.457
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.207 4.265
potential_pw2rs 128 12.3 0.023 0.024 4.206 4.216
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=186.815000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1077.636364, yerr=16.255196
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/19/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 5.910120E+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 1976928 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3442.9
marketing flops 145.650931E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 1.536524E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 242784
MPI messages size (bytes):
total size 1.341806E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.526748E+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 115008 60297314304
4194304 < size <= 16777216 105840 554906419200
16777216 < size 20352 726592466352
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 9010 51.
MP_Alltoall 9724 794507.
MP_ISend 40420 2100460.
MP_IRecv 40420 2099564.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4043 57624.
MP_Allreduce 11184 1163.
MP_Sync 88
MP_Alltoall 1724 18848034.
MP_SendRecv 3870 122880.
MP_ISendRecv 3870 122880.
MP_Wait 16244
MP_ISend 10760 423501.
MP_IRecv 10760 423501.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.030 168.889 168.890
qs_mol_dyn_low 1 2.0 0.003 0.003 168.517 168.530
qs_forces 11 3.9 0.004 0.004 168.422 168.425
qs_energies 11 4.9 0.002 0.002 161.016 161.022
scf_env_do_scf 11 5.9 0.001 0.001 143.819 143.832
velocity_verlet 10 3.0 0.001 0.001 111.579 111.584
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 108.517 108.519
dbcsr_multiply_generic 2529 12.6 0.184 0.191 72.075 72.484
qs_scf_new_mos 118 7.6 0.001 0.001 71.208 71.297
qs_scf_loop_do_ot 118 8.6 0.001 0.001 71.208 71.296
ot_scf_mini 118 9.6 0.003 0.004 66.762 66.825
multiply_cannon 2529 13.6 0.568 0.590 53.537 56.836
multiply_cannon_loop 2529 14.6 0.820 0.864 50.604 51.376
ot_mini 118 10.6 0.001 0.001 37.464 37.538
init_scf_loop 11 6.9 0.000 0.000 35.154 35.156
prepare_preconditioner 11 7.9 0.000 0.000 31.309 31.337
make_preconditioner 11 8.9 0.000 0.000 31.309 31.337
mp_waitall_1 126876 16.7 24.595 30.908 24.595 30.908
make_full_inverse_cholesky 11 9.9 0.016 0.028 29.231 29.509
rebuild_ks_matrix 129 8.3 0.001 0.001 29.366 29.466
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.018 29.365 29.466
qs_ks_update_qs_env 129 7.6 0.001 0.001 26.796 26.887
multiply_cannon_multrec 10116 15.6 10.479 14.651 18.128 21.607
qs_ot_get_derivative 118 11.6 0.002 0.002 20.363 20.426
multiply_cannon_metrocomm3 10116 15.6 0.024 0.026 12.691 19.976
cp_fm_cholesky_invert 11 10.9 17.821 17.828 17.821 17.828
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 17.070 17.219
apply_single 129 13.6 0.001 0.001 17.070 17.218
ot_diis_step 118 11.6 0.020 0.021 17.030 17.030
qs_ot_get_p 129 10.4 0.001 0.001 16.508 16.578
make_m2s 5058 13.6 0.065 0.071 14.671 15.501
make_images 5058 14.6 2.182 2.591 14.361 15.185
qs_rho_update_rho_low 129 7.7 0.001 0.001 14.283 14.314
calculate_rho_elec 129 8.7 0.255 0.266 14.282 14.313
sum_up_and_integrate 129 10.3 0.002 0.002 14.213 14.263
integrate_v_rspace 129 11.3 0.004 0.004 14.152 14.205
qs_ot_p2m_diag 84 11.4 0.502 0.507 12.789 12.807
cp_dbcsr_syevd 84 12.4 0.005 0.006 11.618 11.619
multiply_cannon_sync_h2d 10116 15.6 10.990 11.371 10.990 11.371
init_scf_run 11 5.9 0.000 0.001 10.354 10.354
scf_env_initial_rho_setup 11 6.9 0.000 0.001 10.353 10.354
pw_transfer 1559 11.6 0.086 0.096 9.908 9.955
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 9.684 9.738
make_images_data 5058 15.6 0.056 0.068 8.442 9.718
hybrid_alltoall_any 5245 16.5 0.847 3.804 8.317 9.700
cp_fm_diag_elpa 84 13.4 0.000 0.000 8.748 8.760
cp_fm_diag_elpa_base 84 14.4 8.510 8.589 8.745 8.756
fft_wrap_pw1pw2_140 527 13.2 1.693 1.730 8.605 8.665
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 8.030 8.082
grid_integrate_task_list 129 12.3 7.847 8.063 7.847 8.063
dbcsr_mm_accdrv_process 20934 16.1 3.262 4.246 7.294 7.996
cp_fm_cholesky_decompose 22 10.9 7.830 7.944 7.830 7.944
wfi_extrapolate 11 7.9 0.001 0.001 7.399 7.399
density_rs2pw 129 9.7 0.005 0.005 6.937 7.235
multiply_cannon_metrocomm1 10116 15.6 0.030 0.030 4.260 6.957
calculate_dm_sparse 129 9.5 0.001 0.001 6.106 6.186
fft3d_ps 1301 14.7 3.163 3.270 6.104 6.153
grid_collocate_task_list 129 9.7 5.617 5.838 5.617 5.838
dbcsr_complete_redistribute 397 12.7 2.102 2.165 5.167 5.496
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 5.267 5.268
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.150 5.186
mp_alltoall_d11v 2429 14.1 4.509 4.975 4.509 4.975
mp_allgather_i34 2529 14.6 1.094 4.534 1.094 4.534
potential_pw2rs 129 12.3 0.027 0.028 4.323 4.326
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.060 4.119
multiply_cannon_metrocomm4 7587 15.6 0.026 0.028 1.736 4.084
mp_irecv_dv 29102 15.9 1.697 4.008 1.697 4.008
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.595 3.880
copy_fm_to_dbcsr 210 11.7 0.002 0.002 3.341 3.651
qs_ot_get_orbitals 118 10.6 0.001 0.001 3.599 3.624
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.553 3.619
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.478 3.488
copy_dbcsr_to_fm 187 11.8 0.004 0.004 3.411 3.488
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=168.890000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1447.454545, yerr=33.117871
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410020794368 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1947767668736 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957875093504 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544039424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714619674624 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444697722880 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019193401344 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019193401344 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624869085184 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.820223E+12 0.0% 0.0% 100.0%
flops max/rank 11.609361E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705860672 0.0% 0.0% 100.0%
number of processed stacks 1951856 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3435.6
marketing flops 143.531365E+12
-------------------------------------------------------------------------------
# multiplications 2496
max memory usage/rank 3.087827E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 99840
MPI messages size (bytes):
total size 1.127791E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.295984E+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 45208 35114713088
4194304 < size <= 16777216 43984 376564613120
16777216 < size 10032 716108831376
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3945 59961.
MP_Allreduce 10940 1507.
MP_Sync 86
MP_Alltoall 1700 36954490.
MP_SendRecv 1778 218624.
MP_ISendRecv 1778 218624.
MP_Wait 9728
MP_ISend 6360 1080477.
MP_IRecv 6360 1080477.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.036 288.119 288.121
qs_mol_dyn_low 1 2.0 0.003 0.003 287.555 287.567
qs_forces 11 3.9 0.005 0.005 287.457 287.462
qs_energies 11 4.9 0.002 0.002 278.620 278.632
scf_env_do_scf 11 5.9 0.001 0.001 255.861 255.877
velocity_verlet 10 3.0 0.002 0.002 208.303 208.311
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 132.275 132.277
init_scf_loop 11 6.9 0.000 0.000 123.316 123.321
prepare_preconditioner 11 7.9 0.000 0.000 118.460 118.480
make_preconditioner 11 8.9 0.000 0.000 118.460 118.480
make_full_inverse_cholesky 11 9.9 0.037 0.039 95.009 115.608
qs_scf_new_mos 116 7.6 0.001 0.001 88.013 88.061
qs_scf_loop_do_ot 116 8.6 0.001 0.001 88.012 88.061
ot_scf_mini 116 9.6 0.004 0.004 83.185 83.191
dbcsr_multiply_generic 2496 12.6 0.213 0.223 81.355 81.809
cp_fm_upper_to_full 104 14.8 53.077 75.418 53.077 75.418
multiply_cannon 2496 13.6 0.681 0.729 58.696 60.250
multiply_cannon_loop 2496 14.6 1.047 1.065 54.684 55.921
ot_mini 116 10.6 0.001 0.001 43.440 43.446
dbcsr_complete_redistribute 393 12.7 3.984 4.053 29.315 42.158
copy_fm_to_dbcsr 208 11.6 0.001 0.002 25.866 38.772
transfer_fm_to_dbcsr 11 9.9 0.031 0.032 23.408 36.257
rebuild_ks_matrix 127 8.3 0.001 0.001 34.330 34.333
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.018 34.329 34.332
mp_alltoall_i22 712 14.1 21.174 34.002 21.174 34.002
cp_fm_cholesky_invert 11 10.9 32.959 32.965 32.959 32.965
mp_waitall_1 103208 16.8 27.408 31.919 27.408 31.919
qs_ks_update_qs_env 127 7.6 0.001 0.001 31.848 31.863
qs_ot_get_p 127 10.4 0.001 0.001 24.748 24.781
qs_ot_get_derivative 116 11.6 0.002 0.002 24.298 24.308
qs_ot_p2m_diag 82 11.4 0.868 0.874 20.764 20.791
multiply_cannon_metrocomm3 9984 15.6 0.025 0.025 18.508 19.732
make_m2s 4992 13.6 0.073 0.076 18.217 19.709
make_images 4992 14.6 3.041 3.229 17.740 19.231
ot_diis_step 116 11.6 0.022 0.022 19.119 19.120
cp_dbcsr_syevd 82 12.4 0.006 0.006 19.006 19.008
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 18.484 18.565
apply_single 127 13.6 0.001 0.001 18.484 18.565
multiply_cannon_multrec 9984 15.6 10.265 12.095 18.265 18.353
qs_rho_update_rho_low 127 7.7 0.001 0.001 17.112 17.125
calculate_rho_elec 127 8.7 0.474 0.475 17.111 17.124
sum_up_and_integrate 127 10.3 0.002 0.002 16.239 16.333
integrate_v_rspace 127 11.3 0.004 0.004 16.178 16.271
cp_fm_diag_elpa 82 13.4 0.000 0.000 15.905 15.905
cp_fm_diag_elpa_base 82 14.4 11.535 13.016 15.900 15.900
multiply_cannon_sync_h2d 9984 15.6 14.183 14.217 14.183 14.217
pw_transfer 1535 11.6 0.094 0.095 12.966 12.979
hybrid_alltoall_any 5177 16.4 1.301 3.039 10.526 12.842
fft_wrap_pw1pw2 1281 12.7 0.011 0.012 12.729 12.735
init_scf_run 11 5.9 0.000 0.001 12.460 12.462
scf_env_initial_rho_setup 11 6.9 0.000 0.001 12.460 12.461
make_images_data 4992 15.6 0.064 0.070 10.230 12.391
fft_wrap_pw1pw2_140 519 13.2 3.203 3.254 11.345 11.362
dbcsr_mm_accdrv_process 20634 16.0 4.222 5.979 7.751 9.647
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 9.524 9.530
wfi_extrapolate 11 7.9 0.001 0.001 9.004 9.004
cp_fm_cholesky_decompose 22 10.9 8.793 8.819 8.793 8.819
grid_integrate_task_list 127 12.3 8.500 8.703 8.500 8.703
density_rs2pw 127 9.7 0.005 0.005 8.198 8.247
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 8.124 8.126
fft3d_ps 1281 14.7 3.934 3.955 7.221 7.281
mp_alltoall_d11v 2401 14.1 6.834 6.978 6.834 6.978
calculate_dm_sparse 127 9.5 0.001 0.001 6.887 6.942
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.496 6.544
grid_collocate_task_list 127 9.7 6.367 6.447 6.367 6.447
copy_dbcsr_to_fm 185 11.7 0.004 0.004 6.066 6.152
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=288.121000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2728.818182, yerr=176.940061
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 1.260851E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255646.
MP_Allreduce 3139 6114.
MP_Sync 4
MP_Alltoall 54
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 0.011 0.034 88.828 88.829
qs_energies 1 2.0 0.000 0.000 88.381 88.388
ls_scf 1 3.0 0.000 0.000 87.491 87.499
dbcsr_multiply_generic 111 6.7 0.014 0.015 76.517 76.713
multiply_cannon 111 7.7 0.017 0.020 59.774 60.704
multiply_cannon_loop 111 8.7 0.231 0.245 56.359 57.498
ls_scf_main 1 4.0 0.000 0.000 52.379 52.380
density_matrix_trs4 2 5.0 0.002 0.003 46.843 46.926
ls_scf_init_scf 1 4.0 0.000 0.000 32.066 32.067
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.944 30.994
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.845 28.871
mp_waitall_1 11031 10.9 22.702 26.564 22.702 26.564
multiply_cannon_multrec 2664 9.7 8.152 8.982 15.665 17.565
multiply_cannon_sync_h2d 2664 9.7 13.587 14.979 13.587 14.979
make_m2s 222 7.7 0.008 0.011 13.142 13.581
make_images 222 8.7 0.098 0.108 13.120 13.560
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.540 12.559
multiply_cannon_metrocomm3 2664 9.7 0.009 0.011 5.736 10.787
make_images_data 222 9.7 0.004 0.005 7.762 8.383
dbcsr_mm_accdrv_process 4760 10.4 0.587 0.699 7.129 8.197
hybrid_alltoall_any 227 10.6 0.215 1.830 6.628 7.925
dbcsr_mm_accdrv_process_sort 4760 11.4 6.343 7.326 6.343 7.326
calculate_norms 4752 9.8 5.541 6.215 5.541 6.215
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.028 5.186
compute_matrix_preconditioner 1 6.0 0.000 0.001 4.856 4.860
mp_sum_l 887 5.1 3.062 4.430 3.062 4.430
acc_transpose_blocks 2664 9.7 0.018 0.020 3.902 4.247
acc_transpose_blocks_kernels 2664 10.7 0.035 0.039 3.761 4.102
jit_kernel_transpose 1 13.0 3.726 4.067 3.726 4.067
make_images_sizes 222 9.7 0.000 0.000 0.735 3.541
mp_alltoall_i44 222 10.7 0.735 3.541 0.735 3.541
arnoldi_extremal 4 6.8 0.000 0.000 3.241 3.270
arnoldi_normal_ev 4 7.8 0.001 0.003 3.241 3.270
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.269 3.213
build_subspace 16 8.4 0.009 0.013 3.146 3.147
ls_scf_post 1 4.0 0.000 0.000 3.046 3.054
ls_scf_store_result 1 5.0 0.000 0.000 2.874 2.909
multiply_cannon_metrocomm4 2442 9.7 0.012 0.014 2.048 2.907
mp_irecv_dv 6231 10.9 2.032 2.899 2.032 2.899
dbcsr_special_finalize 555 9.7 0.005 0.006 2.293 2.785
dbcsr_merge_single_wm 555 10.7 0.461 0.577 2.286 2.777
make_images_pack 222 9.7 2.205 2.626 2.207 2.628
dbcsr_matrix_vector_mult 304 9.0 0.006 0.014 2.334 2.571
dbcsr_sort_data 658 11.4 2.084 2.507 2.084 2.507
dbcsr_matrix_vector_mult_local 304 10.0 2.073 2.465 2.075 2.467
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.313 2.402
buffer_matrices_ensure_size 222 8.7 1.770 2.114 1.770 2.114
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=88.829000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1132.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 2.149179E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 10075.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.040 89.416 89.418
qs_energies 1 2.0 0.000 0.001 88.954 88.960
ls_scf 1 3.0 0.000 0.002 87.619 87.624
dbcsr_multiply_generic 111 6.7 0.015 0.015 73.773 74.092
multiply_cannon 111 7.7 0.027 0.039 52.635 56.646
ls_scf_main 1 4.0 0.000 0.013 54.184 54.189
multiply_cannon_loop 111 8.7 0.136 0.149 49.933 53.050
density_matrix_trs4 2 5.0 0.002 0.003 48.460 48.669
ls_scf_init_scf 1 4.0 0.000 0.002 29.683 29.685
mp_waitall_1 9105 10.9 20.575 29.515 20.575 29.515
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.507 28.586
multiply_cannon_multrec 1332 9.7 13.255 17.330 22.676 27.875
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.128 26.138
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.470 20.485
make_m2s 222 7.7 0.006 0.008 14.855 15.422
make_images 222 8.7 1.369 1.694 14.825 15.392
dbcsr_mm_accdrv_process 4041 10.4 0.340 0.533 9.020 10.554
dbcsr_mm_accdrv_process_sort 4041 11.4 8.523 10.017 8.523 10.017
make_images_data 222 9.7 0.004 0.005 8.594 9.483
hybrid_alltoall_any 227 10.6 0.541 2.584 7.966 9.179
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.218 7.895
mp_sum_l 887 5.1 4.809 7.841 4.809 7.841
mp_irecv_dv 3311 11.0 3.197 7.838 3.197 7.838
calculate_norms 2376 9.8 6.064 6.773 6.064 6.773
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.646 6.173
multiply_cannon_sync_h2d 1332 9.7 4.853 6.109 4.853 6.109
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.977 5.251
arnoldi_extremal 4 6.8 0.000 0.000 4.583 4.597
arnoldi_normal_ev 4 7.8 0.001 0.005 4.583 4.597
build_subspace 16 8.4 0.014 0.021 4.337 4.340
ls_scf_post 1 4.0 0.000 0.001 3.751 3.757
ls_scf_store_result 1 5.0 0.000 0.000 3.431 3.578
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.119 3.353
dbcsr_matrix_vector_mult_local 304 10.0 2.740 3.223 2.742 3.225
mp_allgather_i34 111 8.7 0.688 2.843 0.688 2.843
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.159 2.795
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.632 2.754
dbcsr_data_new 4174 10.1 2.119 2.417 2.119 2.417
make_images_pack 222 9.7 1.821 2.134 1.823 2.136
dbcsr_sort_data 436 11.2 1.817 2.036 1.817 2.036
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.850 1.852
rebuild_ks_matrix 3 7.3 0.000 0.000 1.837 1.839
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.004 1.837 1.839
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=89.418000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1762.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 2.932204E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265448.
MP_Allreduce 3138 10896.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.022 0.042 92.820 92.821
qs_energies 1 2.0 0.000 0.000 92.218 92.224
ls_scf 1 3.0 0.000 0.000 90.797 90.802
dbcsr_multiply_generic 111 6.7 0.016 0.019 75.560 75.901
ls_scf_main 1 4.0 0.000 0.000 56.798 56.802
multiply_cannon 111 7.7 0.032 0.090 52.362 56.619
multiply_cannon_loop 111 8.7 0.117 0.131 49.602 53.593
density_matrix_trs4 2 5.0 0.002 0.003 50.901 51.077
mp_waitall_1 7281 11.0 23.739 33.399 23.739 33.399
ls_scf_init_scf 1 4.0 0.000 0.000 30.360 30.362
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.145 29.251
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.758 26.770
multiply_cannon_multrec 888 9.7 12.615 15.402 21.315 24.698
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.054 22.905
make_m2s 222 7.7 0.006 0.007 16.346 17.072
make_images 222 8.7 1.585 1.847 16.308 17.032
make_images_data 222 9.7 0.004 0.005 9.523 10.561
hybrid_alltoall_any 227 10.6 0.641 2.959 9.058 10.165
dbcsr_mm_accdrv_process 3754 10.4 0.308 0.525 8.225 9.518
dbcsr_mm_accdrv_process_sort 3754 11.4 7.773 8.994 7.773 8.994
mp_sum_l 887 5.1 5.144 8.836 5.144 8.836
multiply_cannon_sync_h2d 888 9.7 6.090 7.904 6.090 7.904
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.498 7.225
mp_irecv_dv 2335 11.1 2.481 7.170 2.481 7.170
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.971 7.127
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.808 6.888
arnoldi_extremal 4 6.8 0.000 0.000 5.049 5.065
arnoldi_normal_ev 4 7.8 0.001 0.005 5.049 5.065
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.782 5.037
calculate_norms 1584 9.8 4.355 4.865 4.355 4.865
build_subspace 16 8.4 0.014 0.020 4.751 4.757
mp_allgather_i34 111 8.7 0.865 3.795 0.865 3.795
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.417 3.755
ls_scf_post 1 4.0 0.000 0.000 3.639 3.647
dbcsr_matrix_vector_mult_local 304 10.0 3.021 3.602 3.024 3.604
ls_scf_store_result 1 5.0 0.000 0.000 3.383 3.466
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.859 2.964
dbcsr_data_new 4116 9.9 2.115 2.478 2.115 2.478
make_images_sizes 222 9.7 0.000 0.000 0.990 2.284
mp_alltoall_i44 222 10.7 0.990 2.283 0.990 2.283
dbcsr_sort_data 325 11.1 1.876 2.183 1.876 2.183
make_images_pack 222 9.7 1.622 1.889 1.625 1.892
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.884 1.886
dbcsr_finalize 304 7.8 0.026 0.032 1.612 1.875
rebuild_ks_matrix 3 7.3 0.000 0.000 1.866 1.868
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 1.866 1.868
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=92.821000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2245.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 3.361698E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 13030.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.044 97.477 97.478
qs_energies 1 2.0 0.000 0.000 96.848 96.854
ls_scf 1 3.0 0.000 0.000 95.180 95.186
dbcsr_multiply_generic 111 6.7 0.016 0.018 78.919 79.217
ls_scf_main 1 4.0 0.000 0.000 59.220 59.221
multiply_cannon 111 7.7 0.040 0.070 51.891 56.235
density_matrix_trs4 2 5.0 0.002 0.003 53.079 53.246
multiply_cannon_loop 111 8.7 0.153 0.170 46.880 50.565
ls_scf_init_scf 1 4.0 0.000 0.000 32.719 32.721
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.499 31.575
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.991 29.002
mp_waitall_1 6369 11.0 22.773 28.859 22.773 28.859
multiply_cannon_multrec 1332 9.7 14.210 17.351 22.261 25.351
make_m2s 222 7.7 0.006 0.007 21.173 22.540
make_images 222 8.7 3.138 3.603 21.123 22.493
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 9.205 17.466
make_images_data 222 9.7 0.004 0.004 11.783 13.350
hybrid_alltoall_any 227 10.6 0.799 3.760 11.170 13.024
dbcsr_mm_accdrv_process 3641 10.4 0.294 0.486 7.685 9.196
dbcsr_mm_accdrv_process_sort 3641 11.4 7.229 8.712 7.229 8.712
mp_sum_l 887 5.1 4.264 8.456 4.264 8.456
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.347 6.716
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.085 5.974
multiply_cannon_sync_h2d 1332 9.7 5.430 5.954 5.430 5.954
mp_irecv_dv 3229 10.9 2.060 5.927 2.060 5.927
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.618 5.771
arnoldi_extremal 4 6.8 0.000 0.000 5.234 5.246
arnoldi_normal_ev 4 7.8 0.008 0.011 5.234 5.246
build_subspace 16 8.4 0.014 0.021 4.896 4.905
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.590 4.780
calculate_norms 2376 9.8 4.180 4.529 4.180 4.529
mp_allgather_i34 111 8.7 2.106 4.386 2.106 4.386
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.596 3.891
dbcsr_matrix_vector_mult_local 304 10.0 3.193 3.687 3.196 3.689
dbcsr_sort_data 658 11.4 3.140 3.470 3.140 3.470
dbcsr_special_finalize 555 9.7 0.006 0.007 2.896 3.258
dbcsr_merge_single_wm 555 10.7 0.545 0.672 2.888 3.250
ls_scf_post 1 4.0 0.000 0.000 3.241 3.247
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.027 3.093
ls_scf_store_result 1 5.0 0.000 0.000 2.971 3.046
dbcsr_data_release 10477 10.7 1.585 2.394 1.585 2.394
dbcsr_finalize 304 7.8 0.049 0.061 1.803 1.996
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.478000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2753.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 4.761076E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265536.
MP_Allreduce 3129 15263.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.034 0.051 91.852 91.853
qs_energies 1 2.0 0.000 0.000 91.060 91.066
ls_scf 1 3.0 0.000 0.000 89.073 89.078
dbcsr_multiply_generic 111 6.7 0.017 0.018 70.709 70.891
ls_scf_main 1 4.0 0.000 0.000 56.519 56.520
multiply_cannon 111 7.7 0.085 0.143 52.880 56.247
multiply_cannon_loop 111 8.7 0.090 0.110 50.308 51.560
density_matrix_trs4 2 5.0 0.002 0.003 49.634 49.675
ls_scf_init_scf 1 4.0 0.000 0.000 29.167 29.168
mp_waitall_1 5436 11.0 24.566 28.405 24.566 28.405
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.904 27.946
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.821 25.830
multiply_cannon_multrec 444 9.7 13.609 16.634 20.805 22.270
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 10.980 15.668
make_m2s 222 7.7 0.005 0.005 13.389 14.335
make_images 222 8.7 2.038 2.474 13.322 14.266
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.245 12.712
hybrid_alltoall_any 227 10.6 0.801 3.808 8.063 9.588
make_images_data 222 9.7 0.003 0.004 8.157 9.570
dbcsr_mm_accdrv_process 3003 10.4 0.353 0.401 6.885 8.029
multiply_cannon_sync_h2d 444 9.7 6.755 7.661 6.755 7.661
dbcsr_mm_accdrv_process_sort 3003 11.4 6.521 7.632 6.521 7.632
arnoldi_extremal 4 6.8 0.000 0.000 5.797 5.809
arnoldi_normal_ev 4 7.8 0.001 0.005 5.797 5.809
build_subspace 16 8.4 0.015 0.020 5.385 5.396
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.325 4.513
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.145 4.363
dbcsr_matrix_vector_mult_local 304 10.0 3.700 4.175 3.702 4.177
mp_sum_l 887 5.1 2.432 4.113 2.432 4.113
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.554 4.002
mp_irecv_dv 1241 11.2 1.542 3.973 1.542 3.973
calculate_norms 792 9.8 3.632 3.738 3.632 3.738
mp_allgather_i34 111 8.7 1.146 3.699 1.146 3.699
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.513 3.563
ls_scf_post 1 4.0 0.000 0.000 3.387 3.392
ls_scf_store_result 1 5.0 0.000 0.000 3.173 3.215
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.699 3.118
make_images_sizes 222 9.7 0.000 0.000 0.844 3.099
mp_alltoall_i44 222 10.7 0.843 3.099 0.843 3.099
dbcsr_finalize 304 7.8 0.062 0.076 2.197 2.292
dbcsr_data_new 4608 9.7 1.798 2.277 1.798 2.277
dbcsr_merge_all 275 8.9 0.478 0.522 2.055 2.134
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.033 2.034
rebuild_ks_matrix 3 7.3 0.000 0.000 2.001 2.002
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.001 2.002
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 1.971 1.971
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=91.853000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3750.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_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 8.880923E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284089.
MP_Allreduce 3123 21388.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.070 0.088 106.695 106.696
qs_energies 1 2.0 0.000 0.000 105.385 105.395
ls_scf 1 3.0 0.000 0.000 102.392 102.402
dbcsr_multiply_generic 111 6.7 0.024 0.027 75.967 76.129
ls_scf_main 1 4.0 0.000 0.000 63.771 63.772
density_matrix_trs4 2 5.0 0.002 0.003 54.702 54.758
multiply_cannon 111 7.7 0.185 0.311 49.847 52.210
multiply_cannon_loop 111 8.7 0.098 0.099 46.758 47.309
ls_scf_init_scf 1 4.0 0.000 0.000 34.848 34.848
ls_scf_init_matrix_S 1 5.0 0.000 0.000 33.278 33.304
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 30.579 30.585
mp_waitall_1 4527 11.1 21.951 25.601 21.951 25.601
make_m2s 222 7.7 0.005 0.005 22.535 23.715
make_images 222 8.7 3.573 3.862 22.427 23.608
multiply_cannon_multrec 444 9.7 17.784 18.459 22.498 22.956
hybrid_alltoall_any 227 10.6 1.654 3.633 12.754 15.585
make_images_data 222 9.7 0.004 0.004 12.948 15.175
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.381 11.765
multiply_cannon_sync_h2d 444 9.7 8.778 8.845 8.778 8.845
arnoldi_extremal 4 6.8 0.000 0.000 7.401 7.410
arnoldi_normal_ev 4 7.8 0.003 0.009 7.401 7.410
build_subspace 16 8.4 0.026 0.036 6.854 6.866
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.485 5.640
dbcsr_matrix_vector_mult_local 304 10.0 5.057 5.370 5.060 5.373
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.187 5.276
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.868 5.113
dbcsr_mm_accdrv_process 1814 10.4 0.290 0.358 4.535 4.670
dbcsr_mm_accdrv_process_sort 1814 11.4 4.198 4.338 4.198 4.338
mp_allgather_i34 111 8.7 1.205 3.855 1.205 3.855
ls_scf_post 1 4.0 0.000 0.000 3.773 3.783
make_images_sizes 222 9.7 0.000 0.000 1.463 3.564
mp_alltoall_i44 222 10.7 1.462 3.564 1.462 3.564
ls_scf_store_result 1 5.0 0.000 0.000 3.494 3.529
calculate_norms 792 9.8 3.243 3.323 3.243 3.323
dbcsr_finalize 304 7.8 0.082 0.089 3.070 3.156
qs_energies_init_hamiltonians 1 3.0 0.000 0.000 2.963 2.963
dbcsr_merge_all 275 8.9 0.883 0.919 2.855 2.936
dbcsr_complete_redistribute 5 7.6 1.447 1.492 2.745 2.857
matrix_ls_to_qs 2 6.0 0.000 0.000 2.391 2.522
dbcsr_sort_data 325 11.1 2.440 2.495 2.440 2.495
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.466 2.468
rebuild_ks_matrix 3 7.3 0.000 0.000 2.400 2.402
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.400 2.402
dbcsr_new_transposed 4 7.5 0.241 0.250 2.282 2.290
dbcsr_frobenius_norm 74 6.6 2.057 2.141 2.204 2.241
dbcsr_data_new 6591 9.6 1.818 2.224 1.818 2.224
dbcsr_add_d 103 6.2 0.000 0.000 2.124 2.201
dbcsr_add_anytype 103 7.2 0.859 0.891 2.123 2.200
dbcsr_data_release 12724 10.6 1.967 2.186 1.967 2.186
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=106.696000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6962.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/644f5a76d68b492ea6ab77ab84438411b373c60e_performance_tests/27/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 7009386627072 0.0% 0.0% 100.0%
flops 9 x 9 x 32 7335108845568 0.0% 0.0% 100.0%
flops 9 x 22 x 32 9866241589248 0.0% 0.0% 100.0%
flops 22 x 9 x 32 9884108906496 0.0% 0.0% 100.0%
flops 22 x 22 x 32 13354440523776 0.0% 0.0% 100.0%
flops 32 x 32 x 9 20607185977344 0.0% 0.0% 100.0%
flops 32 x 32 x 22 25186560638976 0.0% 0.0% 100.0%
flops 9 x 32 x 32 28458319085568 0.0% 0.0% 100.0%
flops 22 x 32 x 32 34782389993472 0.0% 0.0% 100.0%
flops 9 x 32 x 9 42881542373376 0.0% 0.0% 100.0%
flops 22 x 32 x 9 55680402235392 0.0% 0.0% 100.0%
flops 9 x 32 x 22 55680402235392 0.0% 0.0% 100.0%
flops 22 x 32 x 22 72328573419520 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 383.054662E+12 0.0% 0.0% 100.0%
flops max/rank 733.641090E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 26899403712 0.0% 0.0% 100.0%
number of processed stacks 118860288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 226.3
marketing flops 780.439111E+12
-------------------------------------------------------------------------------
# multiplications 1445
max memory usage/rank 589.729792E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 78072.
MP_Allreduce 13232 2081.
MP_Sync 1064
MP_Alltoall 2588 4806963.
MP_SendRecv 168740 11136.
MP_ISendRecv 92040 11136.
MP_Wait 102830
MP_comm_split 40
MP_ISend 26090 85106.
MP_IRecv 37890 59644.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.035 222.749 222.750
qs_mol_dyn_low 1 2.0 0.003 0.004 221.785 221.894
qs_forces 5 3.8 0.004 0.006 221.719 221.721
qs_energies 5 4.8 0.003 0.005 218.820 218.841
scf_env_do_scf 5 5.8 0.000 0.001 204.315 204.318
scf_env_do_scf_inner_loop 105 6.6 0.002 0.006 178.161 178.163
qs_scf_new_mos 105 7.6 0.000 0.001 139.380 139.527
qs_scf_loop_do_ot 105 8.6 0.001 0.001 139.379 139.526
dbcsr_multiply_generic 1445 12.2 0.127 0.136 132.959 133.399
ot_scf_mini 105 9.6 0.003 0.003 129.559 129.676
multiply_cannon 1445 13.2 0.273 0.289 114.416 116.520
multiply_cannon_loop 1445 14.2 2.840 2.995 112.821 114.308
velocity_verlet 4 3.0 0.001 0.001 105.147 105.148
ot_mini 105 10.6 0.001 0.001 59.614 59.725
multiply_cannon_multrec 69360 15.2 29.643 34.914 39.548 44.515
mp_waitall_1 488190 16.1 34.538 42.073 34.538 42.073
qs_ot_get_p 112 10.4 0.001 0.001 39.988 40.226
qs_ot_get_derivative 55 11.6 0.001 0.001 37.834 37.997
multiply_cannon_metrocomm3 69360 15.2 0.197 0.208 25.683 33.523
multiply_cannon_sync_h2d 69360 15.2 29.126 33.185 29.126 33.185
qs_ot_p2m_diag 40 11.0 0.020 0.030 28.972 29.044
rebuild_ks_matrix 110 8.4 0.000 0.000 28.294 28.455
qs_ks_build_kohn_sham_matrix 110 9.4 0.011 0.014 28.294 28.454
qs_ks_update_qs_env 112 7.6 0.001 0.001 25.981 26.132
init_scf_loop 7 6.6 0.000 0.000 26.124 26.126
cp_dbcsr_syevd 40 12.0 0.002 0.002 25.723 25.724
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 22.952 23.240
apply_single 62 13.6 0.000 0.000 22.952 23.240
prepare_preconditioner 7 7.6 0.000 0.000 21.389 21.420
make_preconditioner 7 8.6 0.000 0.000 21.389 21.420
ot_new_cg_direction 55 11.6 0.001 0.001 21.041 21.041
cp_fm_syevd 40 13.0 0.000 0.000 20.654 20.783
qs_rho_update_rho_low 110 7.6 0.001 0.001 16.036 16.398
calculate_rho_elec 110 8.6 0.030 0.032 16.035 16.398
cp_fm_redistribute_end 40 14.0 8.010 15.983 8.014 15.984
cp_fm_syevd_base 40 14.0 7.963 15.938 7.963 15.938
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.806 15.020
make_full_inverse_cholesky 7 9.6 0.000 0.000 14.388 14.456
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 13.603 13.713
mp_sum_l 4764 12.2 12.023 12.876 12.023 12.876
init_scf_run 5 5.8 0.000 0.000 11.973 11.974
scf_env_initial_rho_setup 5 6.8 0.002 0.003 11.973 11.974
calculate_dm_sparse 110 9.5 0.000 0.001 11.379 11.581
pw_transfer 1645 12.4 0.079 0.100 11.201 11.373
fft_wrap_pw1pw2 1425 13.5 0.012 0.014 11.062 11.240
density_rs2pw 110 9.6 0.005 0.006 10.346 10.775
dbcsr_mm_accdrv_process 154766 15.8 6.226 6.390 9.773 10.566
qs_vxc_create 110 10.4 0.002 0.004 10.203 10.239
cp_fm_cholesky_invert 7 10.6 10.188 10.196 10.188 10.196
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 10.066 10.136
fft_wrap_pw1pw2_240 915 15.0 0.839 0.920 9.738 9.928
check_diag 80 13.5 8.687 8.967 9.390 9.534
multiply_cannon_metrocomm1 69360 15.2 0.095 0.099 4.714 8.903
acc_transpose_blocks 69360 15.2 0.360 0.379 7.792 8.388
sum_up_and_integrate 60 10.3 0.001 0.003 8.146 8.155
integrate_v_rspace 60 11.3 0.001 0.002 8.128 8.138
fft3d_pb 915 16.0 2.411 2.608 7.902 8.061
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.001 7.682 7.699
transfer_rs2pw 445 10.6 0.007 0.008 7.193 7.664
xc_rho_set_and_dset_create 110 12.4 0.076 0.096 7.165 7.409
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 7.328 7.379
calculate_first_density_matrix 1 7.0 0.000 0.000 7.170 7.183
xc_vxc_pw_create 60 11.3 0.039 0.049 6.797 6.833
make_full_single_inverse 7 9.6 0.001 0.001 6.736 6.767
make_m2s 2890 13.2 0.078 0.087 6.013 6.584
make_images 2890 14.2 0.236 0.258 5.905 6.478
xc_pw_derive 510 13.4 0.005 0.007 5.984 6.044
acc_transpose_blocks_kernels 69360 16.2 0.852 0.891 5.033 5.545
mp_alltoall_z22v 2340 17.7 4.878 5.150 4.878 5.150
mp_waitany 7680 13.5 4.368 4.995 4.368 4.995
jit_kernel_transpose 5 15.0 4.181 4.671 4.181 4.671
multiply_cannon_metrocomm4 67915 15.2 0.187 0.203 2.020 4.651
wfi_extrapolate 5 7.8 0.000 0.001 4.517 4.517
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=222.750000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=562.000000, yerr=2.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 644f5a76d68b492ea6ab77ab84438411b373c60e
Summary: empty
Status: OK