=== 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: d8a4189bb71bd2b243de4999ff8a786f3563fdee
################# 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/01
job id: 50575202
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/02
job id: 50575203
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/03
job id: 50575204
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/04
job id: 50575205
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/05
job id: 50575206
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/06
job id: 50575207
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/07
job id: 50575208
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/08
job id: 50575209
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/09
job id: 50575210
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/10
job id: 50575211
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/11
job id: 50575212
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/12
job id: 50575213
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/13
job id: 50575214
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/14
job id: 50575215
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/15
job id: 50575217
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/16
job id: 50575218
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/17
job id: 50575221
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/18
job id: 50575222
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/19
job id: 50575223
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/20
job id: 50575225
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/21
job id: 50575226
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/22
job id: 50575227
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/23
job id: 50575229
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/24
job id: 50575230
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/25
job id: 50575232
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/26
job id: 50575233
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/27
job id: 50575234
--- 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 0.021 0.035 167.741 167.741
farming_run 1 2.0 166.954 166.956 167.706 167.711
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.493369E+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 0.007 0.027 146.803 146.803
qs_energies 1 2.0 0.000 0.000 146.546 146.548
mp2_main 1 3.0 0.000 0.000 143.950 143.952
mp2_gpw_main 1 4.0 0.027 0.034 142.615 142.617
mp2_ri_gpw_compute_in 1 5.0 0.171 0.173 93.081 93.443
mp2_ri_gpw_compute_in_loop 1 6.0 0.003 0.004 54.639 55.001
rpa_ri_compute_en 1 5.0 0.019 0.021 49.428 49.550
mp2_eri_3c_integrate_gpw 272 7.0 0.152 0.167 41.046 46.087
get_2c_integrals 1 6.0 0.008 0.009 37.246 38.270
rpa_num_int 1 6.0 0.890 14.238 32.913 32.922
integrate_v_rspace 273 8.0 0.436 0.453 24.628 29.490
pw_transfer 6555 10.6 0.374 0.390 26.807 27.226
fft_wrap_pw1pw2 5465 11.4 0.045 0.047 25.437 25.862
grid_integrate_task_list 273 9.0 20.533 25.850 20.533 25.850
fft_wrap_pw1pw2_100 2178 12.4 0.029 0.031 22.970 23.404
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 12.106 19.534
compute_2c_integrals 1 7.0 0.002 0.002 19.464 19.465
compute_2c_integrals_loop_lm 1 8.0 0.001 0.003 18.579 19.084
mp2_eri_2c_integrate_gpw 1 9.0 2.367 2.414 18.578 19.083
cp_fm_cholesky_decompose 12 8.2 17.855 18.819 17.855 18.819
calc_mat_Q 8 8.0 0.000 0.000 10.861 18.288
contract_S_to_Q 8 9.0 0.000 0.000 10.285 17.709
cholesky_decomp 1 7.0 0.000 0.000 16.644 17.681
parallel_gemm_fm 14 9.1 0.000 0.000 9.877 17.237
parallel_gemm_fm_cosma 14 10.1 9.877 17.237 9.877 17.237
fft3d_s 5443 13.4 16.101 16.417 16.122 16.439
mp_sync 25 8.8 14.054 16.253 14.054 16.253
create_integ_mat 1 6.0 7.109 8.135 16.060 16.070
ao_to_mo_and_store_B_mult_1 272 7.0 10.770 15.317 10.770 15.317
calculate_wavefunction 272 8.0 5.426 5.579 12.334 13.029
calc_potential_gpw 544 9.5 0.004 0.005 8.220 8.741
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.001 8.080 8.413
mp_sum_iv 50 7.4 1.101 8.274 1.101 8.274
potential_pw2rs 545 10.0 0.106 0.109 7.500 8.133
get_minimax_grid 1 7.0 2.816 3.227 6.889 7.896
collocate_single_gaussian 272 10.0 0.039 0.042 7.329 7.608
array2fm 1 7.0 0.000 0.000 6.884 7.256
get_rpa_minimax_coeff 1 8.0 4.073 4.670 4.073 4.670
pw_scatter_s 2720 13.7 4.332 4.567 4.332 4.567
pw_gather_s 2722 13.2 3.443 3.737 3.443 3.737
array2fm_buffer_send 1 8.0 2.909 3.189 2.909 3.189
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=142.616588, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2809.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 0.024 0.036 435.723 435.725
farming_run 1 2.0 434.587 434.598 435.679 435.683
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.239458E+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 13798312.
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 0.012 0.032 217.361 217.362
qs_energies 1 2.0 0.002 0.012 217.033 217.039
scf_env_do_scf 1 3.0 0.000 0.000 114.873 114.874
qs_ks_update_qs_env 5 5.0 0.000 0.000 113.727 113.736
rebuild_ks_matrix 4 6.0 0.000 0.000 113.726 113.734
qs_ks_build_kohn_sham_matrix 4 7.0 0.056 0.064 113.726 113.734
hfx_ks_matrix 4 8.0 0.001 0.001 113.306 113.311
integrate_four_center 4 9.0 0.152 0.463 113.305 113.310
integrate_four_center_main 4 10.0 0.084 0.479 102.008 104.977
integrate_four_center_bin 269 11.0 101.924 104.933 101.924 104.933
mp2_main 1 3.0 0.019 0.094 101.817 101.824
mp2_gpw_main 1 4.0 0.054 0.095 100.734 100.747
init_scf_loop 1 4.0 0.000 0.000 96.903 96.904
mp2_ri_gpw_compute_in 1 5.0 0.068 0.099 73.862 74.947
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.021 53.776 54.858
mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.163 41.552 46.344
integrate_v_rspace 95 8.0 0.397 0.564 28.078 32.734
pw_transfer 2240 10.6 0.146 0.166 29.499 29.856
fft_wrap_pw1pw2 1868 11.4 0.018 0.021 28.528 28.926
mp2_ri_gpw_compute_en 1 5.0 0.056 0.062 26.621 28.708
ao_to_mo_and_store_B_mult_1 91 7.0 10.556 28.370 10.556 28.370
grid_integrate_task_list 95 9.0 23.434 28.259 23.434 28.259
fft_wrap_pw1pw2_100 730 12.4 0.013 0.014 26.264 26.656
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.837 1.879 24.878 24.889
get_2c_integrals 1 6.0 0.006 0.058 19.973 20.024
compute_2c_integrals 1 7.0 0.002 0.004 18.931 18.955
fft3d_s 1823 13.4 18.461 18.815 18.474 18.828
compute_2c_integrals_loop_lm 1 8.0 0.001 0.001 18.565 18.761
mp2_eri_2c_integrate_gpw 1 9.0 1.738 1.899 18.564 18.761
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 17.967 17.967
calculate_wavefunction 91 8.0 2.035 2.068 9.637 9.863
mp2_ri_gpw_compute_en_expansio 172 7.0 0.555 0.581 8.732 9.026
potential_pw2rs 186 10.0 0.033 0.034 8.446 8.992
local_gemm 172 8.0 8.177 8.481 8.177 8.481
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.124 8.453
mp2_ri_gpw_compute_en_comm 22 7.0 0.503 0.522 7.923 8.293
calc_potential_gpw 182 9.5 0.002 0.002 7.908 8.096
collocate_single_gaussian 91 10.0 0.018 0.033 7.723 7.927
mp_sync 37 10.5 3.276 6.745 3.276 6.745
integrate_four_center_load 4 10.0 0.000 0.000 6.734 6.739
hfx_load_balance 1 11.0 0.000 0.000 6.734 6.739
mp2_ri_gpw_compute_en_ener 172 7.0 6.342 6.400 6.342 6.400
mp_sendrecv_dm3 2068 8.0 5.943 6.340 5.943 6.340
pw_gather_s 912 13.2 4.481 4.882 4.481 4.882
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=100.731517, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1499.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 453.455872E+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 329962.
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 0.033 123.034 123.035
qs_mol_dyn_low 1 2.0 0.004 0.014 122.752 122.760
velocity_verlet 10 3.0 0.002 0.002 92.756 93.422
mp_sum_dm 438 4.9 62.967 63.661 62.967 63.661
md_write_output 11 3.9 0.622 59.327 0.667 63.578
update_particle_set 20 4.0 0.000 0.000 62.908 63.574
qs_forces 11 3.9 0.002 0.002 59.142 59.143
md_output 10 3.0 0.000 0.000 0.606 57.979
qs_energies 11 4.9 0.002 0.002 57.565 57.576
scf_env_do_scf 11 5.9 0.001 0.001 51.249 51.249
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 48.866 48.866
qs_scf_new_mos 108 7.5 0.000 0.001 37.526 37.824
qs_scf_loop_do_ot 108 8.5 0.000 0.001 37.525 37.824
dbcsr_multiply_generic 2286 12.5 0.095 0.100 36.608 37.064
ot_scf_mini 108 9.5 0.002 0.002 35.686 35.891
multiply_cannon 2286 13.5 0.187 0.193 27.594 29.560
multiply_cannon_loop 2286 14.5 1.848 1.956 26.715 28.710
ot_mini 108 10.5 0.001 0.001 21.955 22.251
qs_ot_get_derivative 108 11.5 0.001 0.001 18.800 19.011
mp_waitall_1 245248 16.5 10.232 16.312 10.232 16.312
multiply_cannon_metrocomm3 54864 15.5 0.073 0.078 6.887 13.813
multiply_cannon_multrec 54864 15.5 3.569 5.515 7.554 10.847
qs_ot_get_p 119 10.4 0.001 0.002 8.789 9.064
rebuild_ks_matrix 119 8.3 0.000 0.000 8.579 8.746
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.579 8.746
mp_sum_l 7287 12.8 6.484 8.385 6.484 8.385
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.545 7.700
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 6.706 7.190
multiply_cannon_sync_h2d 54864 15.5 5.037 6.703 5.037 6.703
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 6.141 6.269
dbcsr_mm_accdrv_process 76910 16.1 1.859 2.908 3.900 5.569
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.470 5.509
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.945 5.115
calculate_rho_elec 119 8.7 0.011 0.016 4.944 5.114
sum_up_and_integrate 119 10.3 0.001 0.002 5.012 5.019
integrate_v_rspace 119 11.3 0.002 0.003 5.001 5.008
init_scf_run 11 5.9 0.000 0.001 4.941 4.942
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.941 4.941
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.635 4.635
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.344 4.345
cp_fm_redistribute_end 50 14.0 2.218 4.313 2.225 4.317
write_trajectory 44 4.9 0.035 3.321 0.045 4.250
cp_fm_diag_elpa_base 50 14.0 2.085 4.191 2.089 4.198
multiply_cannon_metrocomm1 54864 15.5 0.056 0.062 2.240 3.626
calculate_dm_sparse 119 9.5 0.000 0.000 3.071 3.262
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.018 3.254
apply_single 119 13.6 0.000 0.000 3.018 3.254
density_rs2pw 119 9.7 0.004 0.004 2.814 2.923
ot_diis_step 108 11.5 0.006 0.006 2.825 2.826
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.652 2.748
acc_transpose_blocks 54864 15.5 0.226 0.248 2.225 2.733
jit_kernel_multiply 13 15.8 1.976 2.579 1.976 2.579
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=123.035000, 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/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 488.902656E+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 1031941.
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 0.046 0.141 106.223 106.296
qs_mol_dyn_low 1 2.0 0.003 0.004 93.711 105.914
velocity_verlet 10 3.0 0.001 0.002 66.475 67.392
qs_forces 11 3.9 0.012 0.081 45.572 45.574
mp_sum_dm 438 4.9 43.179 44.134 43.179 44.134
md_write_output 11 3.9 0.362 17.368 0.918 43.991
update_particle_set 20 4.0 0.000 0.000 43.071 43.988
qs_energies 11 4.9 0.012 0.046 43.489 43.505
md_output 10 3.0 0.000 0.000 4.904 39.209
scf_env_do_scf 11 5.9 0.001 0.002 37.945 37.946
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 35.058 35.059
write_trajectory 44 4.9 0.096 4.607 0.555 26.623
dbcsr_multiply_generic 2286 12.5 0.101 0.104 23.178 23.766
qs_scf_new_mos 108 7.5 0.001 0.001 22.239 22.473
qs_scf_loop_do_ot 108 8.5 0.001 0.001 22.239 22.472
write_particle_coordinates 11 5.9 0.459 22.016 0.459 22.016
ot_scf_mini 108 9.5 0.002 0.003 21.248 21.418
multiply_cannon 2286 13.5 0.209 0.217 17.238 19.196
multiply_cannon_loop 2286 14.5 1.198 1.260 15.856 18.109
mp_max_l 33 2.8 12.205 16.339 12.205 16.339
write_restart 10 4.0 0.050 2.406 4.136 16.339
update_input 1 5.0 1.479 5.916 4.085 16.339
ot_mini 108 10.5 0.001 0.001 13.111 13.356
mp_waitall_1 200699 16.5 6.262 11.336 6.262 11.336
qs_ot_get_derivative 108 11.5 0.001 0.001 10.594 10.765
update_subsys 1 6.0 2.604 10.414 2.607 10.424
multiply_cannon_metrocomm3 27432 15.5 0.072 0.075 4.178 9.527
multiply_cannon_multrec 27432 15.5 1.808 4.209 6.158 9.017
rebuild_ks_matrix 119 8.3 0.000 0.000 7.630 7.848
qs_ks_build_kohn_sham_matrix 119 9.3 0.025 0.051 7.630 7.848
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.518 6.717
dbcsr_mm_accdrv_process 47894 16.0 3.575 5.709 4.269 6.356
qs_ot_get_p 119 10.4 0.001 0.002 5.079 5.305
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.923 4.815
mp_sum_l 7287 12.8 2.509 4.577 2.509 4.577
sum_up_and_integrate 119 10.3 0.001 0.003 4.410 4.433
integrate_v_rspace 119 11.3 0.006 0.037 4.395 4.420
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.344 4.397
calculate_rho_elec 119 8.7 0.021 0.024 4.344 4.397
init_scf_run 11 5.9 0.000 0.001 4.148 4.149
scf_env_initial_rho_setup 11 6.9 0.001 0.002 4.148 4.149
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.137 4.121
apply_single 119 13.6 0.000 0.000 3.137 4.121
make_m2s 4572 13.5 0.052 0.053 3.009 3.359
make_images 4572 14.5 0.206 0.246 2.919 3.267
qs_ot_p2m_diag 50 11.0 0.009 0.013 3.233 3.258
external_control 118 7.2 0.064 3.051 3.079 3.138
mp_bcast_b_src 168 9.3 3.019 3.138 3.019 3.138
init_scf_loop 11 6.9 0.001 0.004 2.858 2.860
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.799 2.800
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.631 2.720
density_rs2pw 119 9.7 0.004 0.004 2.574 2.646
ot_diis_step 108 11.5 0.011 0.011 2.442 2.445
calculate_first_density_matrix 1 7.0 0.006 0.050 2.378 2.381
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.365 2.367
cp_fm_redistribute_end 50 14.0 1.199 2.331 1.204 2.335
pw_transfer 1439 11.6 0.065 0.068 2.246 2.312
cp_fm_diag_elpa_base 50 14.0 1.097 2.230 1.126 2.266
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.154 2.223
calculate_dm_sparse 119 9.5 0.000 0.001 2.138 2.213
multiply_cannon_sync_h2d 27432 15.5 1.649 2.195 1.649 2.195
potential_pw2rs 119 12.3 0.006 0.006 2.096 2.130
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=106.296000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.818182, yerr=1.748671
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/05/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 59.051995E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3143552 0.0% 0.0% 100.0%
average stack size 0.0 0.0 46.8
marketing flops 2.107587E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 520.216576E+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 3.250 6.420 83.373 84.748
qs_mol_dyn_low 1 2.0 0.003 0.003 73.865 78.156
velocity_verlet 10 3.0 0.001 0.002 54.264 55.306
qs_forces 11 3.9 0.002 0.003 40.385 40.385
qs_energies 11 4.9 0.001 0.002 38.777 38.781
scf_env_do_scf 11 5.9 0.000 0.001 33.805 33.805
mp_sum_dm 438 4.9 32.318 33.366 32.318 33.366
md_write_output 11 3.9 0.862 27.559 1.043 33.330
update_particle_set 20 4.0 0.000 0.000 32.285 33.327
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 30.949 30.949
md_output 10 3.0 0.000 0.000 0.910 29.056
dbcsr_multiply_generic 2286 12.5 0.096 0.098 18.195 18.497
qs_scf_new_mos 108 7.5 0.001 0.001 17.333 17.355
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.333 17.354
ot_scf_mini 108 9.5 0.002 0.002 16.497 16.518
multiply_cannon 2286 13.5 0.195 0.202 14.061 14.650
multiply_cannon_loop 2286 14.5 0.867 0.930 13.118 13.880
ot_mini 108 10.5 0.001 0.001 10.123 10.156
qs_ot_get_derivative 108 11.5 0.001 0.001 8.486 8.508
multiply_cannon_multrec 18288 15.5 1.870 2.907 6.953 7.273
rebuild_ks_matrix 119 8.3 0.000 0.000 6.582 6.631
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.582 6.630
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.818 5.859
dbcsr_mm_accdrv_process 38222 16.0 4.917 5.709 4.990 5.775
write_trajectory 44 4.9 0.154 4.913 0.181 5.770
mp_bcast_b_src 168 10.2 4.735 4.926 4.735 4.926
external_control 118 7.2 0.152 4.855 4.884 4.916
mp_waitall_1 158411 16.6 3.437 4.773 3.437 4.773
mp_max_l 33 2.8 4.292 4.431 4.292 4.431
write_restart 10 4.0 0.138 4.427 0.139 4.427
sum_up_and_integrate 119 10.3 0.001 0.001 4.079 4.095
integrate_v_rspace 119 11.3 0.003 0.003 4.066 4.082
qs_ot_get_p 119 10.4 0.001 0.002 3.972 4.010
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.859 3.880
mp_sum_l 7287 12.8 3.317 3.880 3.317 3.880
calculate_rho_elec 119 8.7 0.031 0.031 3.859 3.880
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.174 3.781
init_scf_run 11 5.9 0.000 0.001 3.695 3.696
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.695 3.695
init_scf_loop 11 6.9 0.000 0.000 2.836 2.836
multiply_cannon_metrocomm3 18288 15.5 0.048 0.050 1.640 2.834
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.617 2.630
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.081 2.528
apply_single 119 13.6 0.000 0.000 2.081 2.528
density_rs2pw 119 9.7 0.004 0.004 2.394 2.483
make_m2s 4572 13.5 0.044 0.045 2.200 2.423
make_images 4572 14.5 0.193 0.207 2.114 2.337
calculate_first_density_matrix 1 7.0 0.000 0.000 2.296 2.296
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.284 2.285
pw_transfer 1439 11.6 0.065 0.069 2.113 2.140
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.020 2.048
prepare_preconditioner 11 7.9 0.000 0.000 2.026 2.030
make_preconditioner 11 8.9 0.000 0.000 2.026 2.030
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.995 2.006
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.970 1.979
cp_fm_diag_elpa_base 50 14.0 1.943 1.960 1.968 1.977
calculate_dm_sparse 119 9.5 0.000 0.001 1.954 1.966
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.853 1.954
grid_integrate_task_list 119 12.3 1.804 1.898 1.804 1.898
potential_pw2rs 119 12.3 0.007 0.008 1.838 1.855
acc_transpose_blocks 18288 15.5 0.079 0.081 1.646 1.752
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.712 1.714
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=84.748000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.272727, yerr=0.962091
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 560.373760E+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 0.007 0.025 92.871 92.872
qs_mol_dyn_low 1 2.0 0.003 0.003 92.571 92.713
velocity_verlet 10 3.0 0.001 0.002 71.617 73.657
md_write_output 11 3.9 1.586 38.039 2.047 49.085
mp_sum_dm 438 4.9 46.937 48.984 46.937 48.984
update_particle_set 20 4.0 0.000 0.000 46.908 48.948
qs_forces 11 3.9 0.002 0.002 43.602 43.602
qs_energies 11 4.9 0.001 0.001 41.811 41.817
md_output 10 3.0 0.000 0.000 1.685 40.374
scf_env_do_scf 11 5.9 0.000 0.001 36.484 36.485
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 32.745 32.745
dbcsr_multiply_generic 2286 12.5 0.101 0.103 20.598 20.712
qs_scf_new_mos 108 7.5 0.001 0.001 19.029 19.085
qs_scf_loop_do_ot 108 8.5 0.001 0.001 19.029 19.085
ot_scf_mini 108 9.5 0.002 0.003 17.949 18.003
multiply_cannon 2286 13.5 0.219 0.229 16.246 16.666
multiply_cannon_loop 2286 14.5 1.498 1.612 15.117 15.547
write_trajectory 44 4.9 0.213 5.093 0.461 11.046
ot_mini 108 10.5 0.001 0.001 10.871 10.942
multiply_cannon_multrec 27432 15.5 2.458 3.133 9.024 9.325
qs_ot_get_derivative 108 11.5 0.001 0.001 8.944 9.002
dbcsr_mm_accdrv_process 47916 15.9 6.142 7.709 6.463 7.772
rebuild_ks_matrix 119 8.3 0.000 0.000 6.888 6.934
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.888 6.933
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.089 6.131
write_particle_coordinates 11 5.9 0.248 5.953 0.248 5.953
mp_bcast_b_src 168 10.2 4.578 4.834 4.578 4.834
external_control 118 7.2 0.198 4.746 4.772 4.818
qs_ot_get_p 119 10.4 0.001 0.002 4.130 4.210
sum_up_and_integrate 119 10.3 0.001 0.001 3.984 3.992
integrate_v_rspace 119 11.3 0.003 0.003 3.973 3.981
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.890 3.928
calculate_rho_elec 119 8.7 0.040 0.046 3.890 3.928
init_scf_run 11 5.9 0.000 0.001 3.924 3.925
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.924 3.924
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.241 3.727
init_scf_loop 11 6.9 0.000 0.000 3.713 3.714
mp_waitall_1 137007 16.6 2.347 3.106 2.347 3.106
make_m2s 4572 13.5 0.054 0.057 2.722 2.939
make_images 4572 14.5 0.274 0.332 2.613 2.829
prepare_preconditioner 11 7.9 0.000 0.000 2.820 2.829
make_preconditioner 11 8.9 0.000 0.000 2.820 2.829
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.414 2.752
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.236 2.713
apply_single 119 13.6 0.000 0.000 2.236 2.713
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.544 2.556
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.507 2.541
acc_transpose_blocks 27432 15.5 0.120 0.122 2.374 2.520
density_rs2pw 119 9.7 0.003 0.004 2.311 2.416
calculate_first_density_matrix 1 7.0 0.000 0.000 2.331 2.332
pw_transfer 1439 11.6 0.065 0.069 2.225 2.265
calculate_dm_sparse 119 9.5 0.000 0.001 2.221 2.264
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.133 2.175
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.115 2.115
grid_integrate_task_list 119 12.3 1.836 1.913 1.836 1.913
ot_diis_step 108 11.5 0.012 0.012 1.878 1.879
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=92.872000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=532.181818, yerr=4.488737
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 612.028416E+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 1.645 3.952 87.855 88.896
qs_mol_dyn_low 1 2.0 0.003 0.003 75.321 84.730
velocity_verlet 10 3.0 0.019 0.020 56.444 58.706
md_output 10 3.0 0.000 0.000 2.891 46.222
md_write_output 11 3.9 2.625 41.994 2.718 43.460
qs_forces 11 3.9 0.002 0.002 38.472 38.473
qs_energies 11 4.9 0.002 0.002 36.697 36.700
mp_sum_dm 438 4.9 33.956 36.222 33.956 36.222
update_particle_set 20 4.0 0.000 0.000 33.926 36.188
scf_env_do_scf 11 5.9 0.000 0.001 31.578 31.578
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 28.715 28.716
dbcsr_multiply_generic 2286 12.5 0.094 0.096 14.916 15.020
qs_scf_new_mos 108 7.5 0.001 0.001 13.863 13.898
qs_scf_loop_do_ot 108 8.5 0.001 0.001 13.863 13.898
ot_scf_mini 108 9.5 0.002 0.002 13.074 13.104
multiply_cannon 2286 13.5 0.224 0.229 11.653 12.228
multiply_cannon_loop 2286 14.5 0.649 0.674 10.523 10.824
mp_max_l 33 2.8 9.409 10.036 9.409 10.036
ot_mini 108 10.5 0.001 0.001 7.606 7.644
multiply_cannon_multrec 9144 15.5 2.010 2.215 6.496 6.730
mp_bcast_b_src 168 10.2 6.013 6.499 6.013 6.499
external_control 118 7.2 0.399 6.383 6.408 6.485
rebuild_ks_matrix 119 8.3 0.000 0.000 6.318 6.357
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 6.318 6.357
qs_ot_get_derivative 108 11.5 0.001 0.001 6.234 6.263
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.626 5.662
dbcsr_mm_accdrv_process 12550 15.8 3.693 4.410 4.373 4.454
sum_up_and_integrate 119 10.3 0.001 0.001 3.850 3.855
integrate_v_rspace 119 11.3 0.003 0.003 3.840 3.845
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.764 3.778
calculate_rho_elec 119 8.7 0.060 0.061 3.763 3.778
init_scf_run 11 5.9 0.000 0.001 3.635 3.635
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.635 3.635
qs_ot_get_p 119 10.4 0.002 0.002 3.440 3.481
mp_waitall_1 115863 16.7 2.157 2.987 2.157 2.987
init_scf_loop 11 6.9 0.000 0.000 2.835 2.837
write_restart 10 4.0 0.173 2.769 0.174 2.770
make_m2s 4572 13.5 0.034 0.035 2.266 2.551
make_images 4572 14.5 0.269 0.302 2.176 2.459
calculate_first_density_matrix 1 7.0 0.000 0.000 2.397 2.397
pw_transfer 1439 11.6 0.066 0.068 2.301 2.317
qs_ot_p2m_diag 50 11.0 0.022 0.023 2.229 2.232
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.208 2.225
density_rs2pw 119 9.7 0.003 0.003 2.113 2.162
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.103 2.121
mp_sum_l 7287 12.8 1.801 2.075 1.801 2.075
prepare_preconditioner 11 7.9 0.000 0.000 2.058 2.064
make_preconditioner 11 8.9 0.000 0.000 2.058 2.064
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.940 1.977
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.954 1.955
grid_integrate_task_list 119 12.3 1.881 1.936 1.881 1.936
calculate_dm_sparse 119 9.5 0.000 0.000 1.888 1.911
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=88.896000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=581.000000, yerr=4.369314
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 772.169728E+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 3.126 5.675 86.071 87.281
qs_mol_dyn_low 1 2.0 0.003 0.003 78.544 81.366
velocity_verlet 10 3.0 0.002 0.002 54.433 57.750
qs_forces 11 3.9 0.002 0.002 51.592 51.593
qs_energies 11 4.9 0.001 0.002 49.522 49.527
scf_env_do_scf 11 5.9 0.001 0.001 43.419 43.419
scf_env_do_scf_inner_loop 108 6.5 0.004 0.009 34.663 34.664
mp_sum_dm 438 4.9 23.254 26.585 23.254 26.585
md_write_output 11 3.9 2.276 18.206 3.319 26.539
update_particle_set 20 4.0 0.000 0.000 23.218 26.534
md_output 10 3.0 0.000 0.000 2.632 21.038
dbcsr_multiply_generic 2286 12.5 0.113 0.114 20.571 20.845
qs_scf_new_mos 108 7.5 0.001 0.001 19.354 19.446
qs_scf_loop_do_ot 108 8.5 0.001 0.001 19.353 19.445
ot_scf_mini 108 9.5 0.002 0.002 18.109 18.208
multiply_cannon 2286 13.5 0.298 0.302 15.700 16.902
multiply_cannon_loop 2286 14.5 0.865 0.877 14.293 15.315
ot_mini 108 10.5 0.001 0.001 10.738 10.868
multiply_cannon_multrec 9144 15.5 3.398 4.804 8.894 8.980
init_scf_loop 11 6.9 0.000 0.000 8.723 8.727
qs_ot_get_derivative 108 11.5 0.001 0.001 8.545 8.644
write_trajectory 44 4.9 0.614 4.907 1.042 8.332
rebuild_ks_matrix 119 8.3 0.000 0.000 7.899 8.062
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.899 8.062
prepare_preconditioner 11 7.9 0.000 0.000 7.673 7.688
make_preconditioner 11 8.9 0.000 0.000 7.672 7.688
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.173 7.562
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.131 7.279
dbcsr_mm_accdrv_process 12550 15.8 4.542 6.577 5.354 6.612
mp_bcast_b_src 168 10.2 4.136 4.833 4.136 4.833
external_control 118 7.2 0.580 4.640 4.693 4.745
cp_fm_upper_to_full 72 14.2 3.243 4.683 3.243 4.683
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.670 4.677
calculate_rho_elec 119 8.7 0.117 0.120 4.670 4.677
mp_waitall_1 94719 16.7 3.259 4.371 3.259 4.371
qs_ot_get_p 119 10.4 0.004 0.005 4.176 4.324
sum_up_and_integrate 119 10.3 0.001 0.001 4.298 4.305
integrate_v_rspace 119 11.3 0.003 0.003 4.288 4.294
init_scf_run 11 5.9 0.000 0.001 3.975 3.975
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.975 3.975
mp_sum_l 7287 12.8 2.637 3.682 2.637 3.682
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.021 3.456
write_particle_coordinates 11 5.9 0.428 3.425 0.428 3.425
pw_transfer 1439 11.6 0.068 0.069 3.345 3.353
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 3.247 3.254
make_m2s 4572 13.5 0.037 0.038 2.995 3.239
mp_max_l 33 2.8 2.823 3.226 2.823 3.226
write_restart 10 4.0 0.403 3.221 0.404 3.222
make_images 4572 14.5 0.354 0.386 2.874 3.118
dbcsr_complete_redistribute 329 12.2 0.278 0.283 2.151 3.002
fft_wrap_pw1pw2_140 487 13.2 0.649 0.652 2.704 2.712
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.699 2.703
density_rs2pw 119 9.7 0.003 0.003 2.602 2.615
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.775 2.613
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.329 2.572
apply_single 119 13.6 0.000 0.000 2.329 2.571
qs_ot_p2m_diag 50 11.0 0.044 0.044 2.481 2.483
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.378 2.429
calculate_dm_sparse 119 9.5 0.000 0.000 2.361 2.380
mp_alltoall_i22 627 13.8 1.527 2.380 1.527 2.380
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.494 2.326
calculate_first_density_matrix 1 7.0 0.000 0.000 2.309 2.312
grid_integrate_task_list 119 12.3 2.149 2.177 2.149 2.177
fft3d_ps 1201 14.6 0.874 0.894 2.168 2.172
ot_diis_step 108 11.5 0.014 0.015 2.147 2.147
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.082 2.083
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.060 2.060
cp_fm_cholesky_invert 11 10.9 1.987 1.992 1.987 1.992
acc_transpose_blocks 9144 15.5 0.046 0.046 1.849 1.873
make_images_data 4572 15.5 0.046 0.049 1.430 1.809
hybrid_alltoall_any 4725 16.4 0.091 0.150 1.431 1.801
qs_energies_init_hamiltonians 11 5.9 0.004 0.004 1.800 1.800
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.736 1.756
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=87.281000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=724.000000, yerr=12.482715
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/09/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 198.287135E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 8410880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 117.0
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 502.484992E+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 2113911.
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 0.008 0.025 88.716 88.717
qs_mol_dyn_low 1 2.0 0.003 0.003 88.281 88.290
qs_forces 11 3.9 0.003 0.003 88.201 88.202
qs_energies 11 4.9 0.009 0.018 85.187 85.205
scf_env_do_scf 11 5.9 0.000 0.001 75.017 75.020
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 68.752 68.753
dbcsr_multiply_generic 2055 12.4 0.108 0.132 51.963 52.351
qs_scf_new_mos 99 7.5 0.000 0.001 50.081 50.255
qs_scf_loop_do_ot 99 8.5 0.000 0.001 50.081 50.254
ot_scf_mini 99 9.5 0.002 0.002 47.638 47.759
velocity_verlet 10 3.0 0.001 0.002 45.498 45.499
multiply_cannon 2055 13.4 0.183 0.197 42.691 43.703
multiply_cannon_loop 2055 14.4 1.811 1.845 41.511 42.609
ot_mini 99 10.5 0.001 0.001 27.298 27.410
qs_ot_get_derivative 99 11.5 0.001 0.001 20.401 20.514
multiply_cannon_multrec 49320 15.4 11.343 12.103 17.383 18.315
rebuild_ks_matrix 110 8.3 0.000 0.001 15.460 15.646
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 15.460 15.645
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.563 13.710
mp_waitall_1 220248 16.4 11.444 12.335 11.444 12.335
qs_ot_get_p 110 10.4 0.001 0.001 11.719 11.859
multiply_cannon_sync_h2d 49320 15.4 9.508 10.263 9.508 10.263
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.740 8.278
qs_ot_p2m_diag 48 11.0 0.013 0.019 8.114 8.131
init_scf_run 11 5.9 0.000 0.001 7.906 7.907
scf_env_initial_rho_setup 11 6.9 0.009 0.021 7.906 7.907
multiply_cannon_metrocomm3 49320 15.4 0.083 0.088 6.357 7.779
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.290 7.762
apply_single 110 13.6 0.000 0.001 7.290 7.762
sum_up_and_integrate 110 10.3 0.002 0.003 7.740 7.752
integrate_v_rspace 110 11.3 0.003 0.003 7.714 7.733
qs_rho_update_rho_low 110 7.6 0.000 0.001 7.285 7.428
calculate_rho_elec 110 8.6 0.020 0.024 7.285 7.427
cp_dbcsr_syevd 48 12.0 0.003 0.003 7.235 7.236
ot_diis_step 99 11.5 0.005 0.006 6.675 6.675
cp_fm_diag_elpa 48 13.0 0.000 0.000 6.464 6.489
cp_fm_diag_elpa_base 48 14.0 6.444 6.471 6.461 6.486
init_scf_loop 11 6.9 0.000 0.000 6.223 6.224
dbcsr_mm_accdrv_process 87628 16.1 3.077 3.159 5.909 6.189
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.691 5.754
make_m2s 4110 13.4 0.061 0.065 4.730 4.887
mp_sum_l 6594 12.7 4.080 4.828 4.080 4.828
make_images 4110 14.4 0.179 0.192 4.633 4.795
density_rs2pw 110 9.6 0.004 0.004 4.265 4.483
wfi_extrapolate 11 7.9 0.001 0.001 4.373 4.373
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 4.262 4.270
prepare_preconditioner 11 7.9 0.000 0.000 3.945 3.967
make_preconditioner 11 8.9 0.000 0.000 3.945 3.967
pw_transfer 1331 11.6 0.055 0.065 3.847 3.965
calculate_dm_sparse 110 9.5 0.001 0.001 3.822 3.921
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.759 3.878
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.698 3.760
multiply_cannon_metrocomm1 49320 15.4 0.065 0.068 2.564 3.568
grid_integrate_task_list 110 12.3 3.244 3.458 3.244 3.458
calculate_first_density_matrix 1 7.0 0.000 0.000 3.365 3.378
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.301 3.348
potential_pw2rs 110 12.3 0.005 0.006 3.279 3.323
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.234 3.316
fft_wrap_pw1pw2_140 451 13.1 0.310 0.332 3.053 3.169
fft3d_ps 1111 14.6 0.793 0.888 3.053 3.153
mp_alltoall_d11v 2046 13.8 2.383 2.828 2.383 2.828
transfer_rs2pw 451 10.6 0.005 0.006 2.461 2.631
acc_transpose_blocks 49320 15.4 0.213 0.220 2.577 2.620
jit_kernel_multiply 13 15.9 2.550 2.583 2.550 2.583
grid_collocate_task_list 110 9.6 2.154 2.263 2.154 2.263
mp_waitany 14300 13.8 1.895 2.251 1.895 2.251
transfer_pw2rs 451 13.1 0.006 0.006 2.203 2.242
make_images_data 4110 15.4 0.045 0.048 2.025 2.238
cp_fm_cholesky_invert 11 10.9 2.182 2.187 2.182 2.187
mp_sum_d 3891 11.9 1.452 2.072 1.452 2.072
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 2.031 2.067
hybrid_alltoall_any 4261 16.3 0.085 0.487 1.744 2.054
mp_alltoall_z22v 1111 16.6 1.869 2.052 1.869 2.052
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=88.717000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=477.181818, yerr=2.587318
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 593.272832E+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 2942216.
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 0.103 0.152 75.963 75.969
qs_mol_dyn_low 1 2.0 0.004 0.011 74.577 74.589
qs_forces 11 3.9 0.014 0.057 74.482 74.490
qs_energies 11 4.9 0.003 0.012 71.074 71.088
scf_env_do_scf 11 5.9 0.001 0.001 61.157 61.167
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 52.759 52.764
dbcsr_multiply_generic 2055 12.4 0.117 0.122 39.554 39.951
velocity_verlet 10 3.0 0.001 0.002 38.276 38.281
qs_scf_new_mos 99 7.5 0.001 0.001 35.942 36.076
qs_scf_loop_do_ot 99 8.5 0.001 0.001 35.942 36.075
ot_scf_mini 99 9.5 0.003 0.004 34.165 34.271
multiply_cannon 2055 13.4 0.224 0.246 31.842 32.973
multiply_cannon_loop 2055 14.4 1.170 1.203 30.325 31.601
ot_mini 99 10.5 0.001 0.001 19.605 19.744
multiply_cannon_multrec 24660 15.4 6.871 8.963 13.959 15.821
rebuild_ks_matrix 110 8.3 0.000 0.000 14.039 14.137
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.017 14.038 14.136
qs_ot_get_derivative 99 11.5 0.001 0.001 13.717 13.830
mp_waitall_1 176588 16.5 8.460 12.468 8.460 12.468
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.377 12.466
multiply_cannon_metrocomm3 24660 15.4 0.074 0.076 5.494 9.177
init_scf_loop 11 6.9 0.001 0.003 8.357 8.358
qs_ot_get_p 110 10.4 0.001 0.004 7.636 7.771
init_scf_run 11 5.9 0.000 0.001 7.372 7.373
scf_env_initial_rho_setup 11 6.9 0.001 0.002 7.372 7.373
multiply_cannon_sync_h2d 24660 15.4 6.327 7.330 6.327 7.330
dbcsr_mm_accdrv_process 52282 16.1 5.504 6.339 6.920 7.229
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.491 7.100
apply_single 110 13.6 0.000 0.001 6.490 7.100
sum_up_and_integrate 110 10.3 0.002 0.004 6.783 6.794
integrate_v_rspace 110 11.3 0.004 0.015 6.756 6.767
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.448 6.476
calculate_rho_elec 110 8.6 0.039 0.047 6.448 6.476
prepare_preconditioner 11 7.9 0.000 0.000 6.281 6.301
make_preconditioner 11 8.9 0.000 0.001 6.281 6.301
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.840 6.011
ot_diis_step 99 11.5 0.010 0.010 5.798 5.798
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 5.092 5.781
qs_ot_p2m_diag 48 11.0 0.029 0.045 5.441 5.465
make_m2s 4110 13.4 0.057 0.059 4.823 5.384
make_images 4110 14.4 0.408 0.465 4.710 5.267
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.916 4.916
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.208 4.221
cp_fm_diag_elpa_base 48 14.0 4.155 4.176 4.205 4.218
pw_transfer 1331 11.6 0.066 0.073 3.885 4.053
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.778 3.949
density_rs2pw 110 9.6 0.004 0.004 3.671 3.836
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.718 3.781
wfi_extrapolate 11 7.9 0.001 0.001 3.682 3.683
calculate_first_density_matrix 1 7.0 0.001 0.003 3.586 3.588
grid_integrate_task_list 110 12.3 3.159 3.345 3.159 3.345
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.307 3.309
make_images_data 4110 15.4 0.049 0.053 2.598 3.207
fft_wrap_pw1pw2_140 451 13.1 0.358 0.379 3.026 3.193
calculate_dm_sparse 110 9.5 0.001 0.001 3.121 3.155
hybrid_alltoall_any 4261 16.3 0.104 0.449 2.268 3.118
mp_sum_l 6594 12.7 2.269 3.069 2.269 3.069
fft3d_ps 1111 14.6 1.117 1.352 2.873 3.048
cp_fm_cholesky_invert 11 10.9 3.026 3.036 3.026 3.036
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.909 2.942
potential_pw2rs 110 12.3 0.008 0.009 2.643 2.663
grid_collocate_task_list 110 9.6 2.170 2.316 2.170 2.316
jit_kernel_multiply 12 16.4 1.046 2.304 1.046 2.304
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.155 2.200
mp_alltoall_d11v 2046 13.8 1.945 2.195 1.945 2.195
make_basis_sm 11 9.8 0.000 0.000 2.034 2.038
transfer_rs2pw 451 10.6 0.006 0.008 1.851 2.022
parallel_gemm_fm 81 9.0 0.000 0.000 2.005 2.011
parallel_gemm_fm_cosma 81 10.0 2.005 2.011 2.005 2.011
acc_transpose_blocks 24660 15.4 0.117 0.120 1.974 2.003
mp_allgather_i34 2055 14.4 0.824 1.957 0.824 1.957
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.939 1.948
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.821 1.833
multiply_cannon_metrocomm4 22605 15.4 0.074 0.079 0.784 1.680
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.597 1.607
dbcsr_complete_redistribute 325 12.2 0.233 0.287 1.310 1.580
cp_fm_cholesky_decompose 22 10.9 1.569 1.575 1.569 1.575
mp_alltoall_z22v 1111 16.6 1.469 1.573 1.469 1.573
transfer_pw2rs 451 13.1 0.005 0.007 1.510 1.532
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=75.969000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=559.909091, yerr=6.473085
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 666.370048E+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 0.112 0.335 69.311 69.320
qs_mol_dyn_low 1 2.0 0.003 0.005 67.758 67.805
qs_forces 11 3.9 0.003 0.003 67.685 67.686
qs_energies 11 4.9 0.005 0.033 64.272 64.325
scf_env_do_scf 11 5.9 0.001 0.003 54.926 54.927
scf_env_do_scf_inner_loop 99 6.5 0.003 0.006 44.740 44.741
velocity_verlet 10 3.0 0.001 0.002 36.035 36.037
dbcsr_multiply_generic 2055 12.4 0.111 0.113 30.946 31.433
qs_scf_new_mos 99 7.5 0.001 0.001 28.969 29.107
qs_scf_loop_do_ot 99 8.5 0.001 0.001 28.968 29.106
ot_scf_mini 99 9.5 0.002 0.003 27.634 27.780
multiply_cannon 2055 13.4 0.211 0.221 23.496 25.569
multiply_cannon_loop 2055 14.4 0.822 0.849 22.007 23.820
ot_mini 99 10.5 0.001 0.001 15.383 15.549
rebuild_ks_matrix 110 8.3 0.000 0.000 12.846 13.074
qs_ks_build_kohn_sham_matrix 110 9.3 0.015 0.041 12.846 13.074
mp_waitall_1 139946 16.5 8.224 11.754 8.224 11.754
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.264 11.467
qs_ot_get_derivative 99 11.5 0.001 0.001 10.775 10.924
multiply_cannon_multrec 16440 15.4 3.567 4.390 9.789 10.595
init_scf_loop 11 6.9 0.001 0.005 10.146 10.147
multiply_cannon_metrocomm3 16440 15.4 0.046 0.048 5.027 8.504
prepare_preconditioner 11 7.9 0.000 0.000 8.261 8.282
make_preconditioner 11 8.9 0.000 0.000 8.261 8.281
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.564 7.942
qs_ot_get_p 110 10.4 0.001 0.002 7.072 7.261
init_scf_run 11 5.9 0.000 0.001 6.751 6.751
scf_env_initial_rho_setup 11 6.9 0.001 0.002 6.751 6.751
sum_up_and_integrate 110 10.3 0.002 0.003 6.733 6.746
integrate_v_rspace 110 11.3 0.003 0.003 6.707 6.721
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.414 6.434
calculate_rho_elec 110 8.6 0.059 0.059 6.414 6.434
dbcsr_mm_accdrv_process 34862 16.1 5.474 5.779 6.072 6.197
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.953 5.540
apply_single 110 13.6 0.000 0.000 4.953 5.540
make_m2s 4110 13.4 0.049 0.050 4.731 5.107
qs_ot_p2m_diag 48 11.0 0.042 0.044 5.034 5.051
make_images 4110 14.4 0.400 0.522 4.614 4.994
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.571 4.572
ot_diis_step 99 11.5 0.011 0.011 4.537 4.537
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.644 4.220
pw_transfer 1331 11.6 0.065 0.071 3.985 4.010
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.878 3.906
multiply_cannon_sync_h2d 16440 15.4 3.175 3.877 3.175 3.877
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.859 3.868
cp_fm_diag_elpa_base 48 14.0 3.776 3.815 3.857 3.866
density_rs2pw 110 9.6 0.004 0.004 3.586 3.781
cp_fm_cholesky_invert 11 10.9 3.629 3.639 3.629 3.639
calculate_first_density_matrix 1 7.0 0.000 0.002 3.406 3.408
grid_integrate_task_list 110 12.3 3.180 3.359 3.180 3.359
wfi_extrapolate 11 7.9 0.001 0.001 3.234 3.235
fft_wrap_pw1pw2_140 451 13.1 0.461 0.470 3.198 3.226
mp_sum_l 6594 12.7 2.213 3.195 2.213 3.195
make_images_data 4110 15.4 0.047 0.051 2.619 3.139
hybrid_alltoall_any 4261 16.3 0.108 0.379 2.299 3.128
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.019 3.095
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.884 2.886
fft3d_ps 1111 14.6 1.126 1.154 2.798 2.822
calculate_dm_sparse 110 9.5 0.001 0.001 2.651 2.680
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.498 2.568
multiply_cannon_metrocomm4 14385 15.4 0.047 0.053 0.901 2.551
potential_pw2rs 110 12.3 0.010 0.011 2.532 2.545
mp_irecv_dv 48980 15.7 0.827 2.404 0.827 2.404
mp_alltoall_d11v 2046 13.8 2.074 2.403 2.074 2.403
grid_collocate_task_list 110 9.6 2.221 2.400 2.221 2.400
dbcsr_complete_redistribute 325 12.2 0.308 0.349 1.648 2.139
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 2.023 2.024
cp_fm_cholesky_decompose 22 10.9 1.981 2.003 1.981 2.003
mp_allgather_i34 2055 14.4 0.795 1.934 0.795 1.934
transfer_rs2pw 451 10.6 0.005 0.005 1.711 1.885
cp_fm_upper_to_full 70 14.2 1.469 1.878 1.469 1.878
parallel_gemm_fm 81 9.0 0.000 0.000 1.818 1.820
parallel_gemm_fm_cosma 81 10.0 1.818 1.820 1.818 1.820
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.782 1.807
make_basis_sm 11 9.8 0.000 0.000 1.799 1.802
acc_transpose_blocks 16440 15.4 0.078 0.081 1.580 1.624
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.078 1.558
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.519 1.535
mp_waitany 17072 13.8 1.316 1.524 1.316 1.524
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.494 1.515
transfer_pw2rs 451 13.1 0.005 0.005 1.476 1.494
mp_alltoall_z22v 1111 16.6 1.458 1.486 1.458 1.486
build_core_hamiltonian_matrix_ 11 4.9 0.003 0.024 1.369 1.469
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=69.320000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=631.363636, yerr=9.138060
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 746.409984E+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 0.010 0.031 71.693 71.693
qs_mol_dyn_low 1 2.0 0.003 0.003 71.359 71.369
qs_forces 11 3.9 0.003 0.003 71.274 71.275
qs_energies 11 4.9 0.010 0.014 67.808 67.812
scf_env_do_scf 11 5.9 0.000 0.001 58.499 58.501
scf_env_do_scf_inner_loop 99 6.5 0.002 0.006 45.463 45.463
velocity_verlet 10 3.0 0.001 0.002 39.732 39.735
dbcsr_multiply_generic 2055 12.4 0.117 0.125 31.771 31.975
qs_scf_new_mos 99 7.5 0.001 0.001 29.970 30.085
qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.970 30.084
ot_scf_mini 99 9.5 0.003 0.003 28.239 28.333
multiply_cannon 2055 13.4 0.238 0.254 23.749 24.878
multiply_cannon_loop 2055 14.4 1.414 1.498 22.280 22.919
ot_mini 99 10.5 0.001 0.001 15.627 15.751
multiply_cannon_multrec 24660 15.4 4.077 6.783 13.062 14.240
init_scf_loop 11 6.9 0.000 0.000 12.984 12.984
rebuild_ks_matrix 110 8.3 0.000 0.000 12.503 12.615
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.502 12.615
qs_ot_get_derivative 99 11.5 0.001 0.001 11.404 11.500
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.056 11.155
prepare_preconditioner 11 7.9 0.000 0.000 11.102 11.116
make_preconditioner 11 8.9 0.000 0.000 11.102 11.116
make_full_inverse_cholesky 11 9.9 0.000 0.000 9.277 10.807
dbcsr_mm_accdrv_process 52304 16.0 7.812 8.975 8.826 9.784
qs_ot_get_p 110 10.4 0.001 0.001 7.380 7.529
mp_waitall_1 121746 16.5 4.986 6.844 4.986 6.844
sum_up_and_integrate 110 10.3 0.001 0.002 6.633 6.646
integrate_v_rspace 110 11.3 0.003 0.003 6.607 6.619
make_m2s 4110 13.4 0.060 0.062 6.237 6.535
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.392 6.399
calculate_rho_elec 110 8.6 0.078 0.081 6.392 6.399
make_images 4110 14.4 0.575 0.694 6.095 6.388
init_scf_run 11 5.9 0.000 0.001 6.334 6.334
scf_env_initial_rho_setup 11 6.9 0.001 0.002 6.333 6.334
qs_ot_p2m_diag 48 11.0 0.055 0.064 5.176 5.193
cp_fm_upper_to_full 70 14.2 3.478 5.060 3.478 5.060
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.573 4.574
dbcsr_complete_redistribute 325 12.2 0.471 0.932 3.316 4.515
ot_diis_step 99 11.5 0.011 0.011 4.183 4.184
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.083 4.161
apply_single 110 13.6 0.000 0.000 4.083 4.161
pw_transfer 1331 11.6 0.065 0.073 4.021 4.063
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.914 3.959
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.892 3.902
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.720 3.901
cp_fm_diag_elpa_base 48 14.0 3.709 3.782 3.889 3.899
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.753 3.806
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.698 3.751
density_rs2pw 110 9.6 0.004 0.004 3.460 3.607
make_images_data 4110 15.4 0.049 0.053 3.130 3.522
grid_integrate_task_list 110 12.3 3.270 3.422 3.270 3.422
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.395 3.396
cp_fm_cholesky_invert 11 10.9 3.318 3.327 3.318 3.327
hybrid_alltoall_any 4261 16.3 0.123 0.462 2.557 3.305
fft_wrap_pw1pw2_140 451 13.1 0.532 0.547 3.224 3.271
wfi_extrapolate 11 7.9 0.001 0.001 3.178 3.178
calculate_dm_sparse 110 9.5 0.001 0.001 3.108 3.135
multiply_cannon_metrocomm3 24660 15.4 0.037 0.039 1.428 3.100
calculate_first_density_matrix 1 7.0 0.000 0.000 3.049 3.052
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.816 2.986
mp_alltoall_i22 605 13.7 1.740 2.958 1.740 2.958
fft3d_ps 1111 14.6 1.155 1.191 2.740 2.763
acc_transpose_blocks 24660 15.4 0.113 0.117 2.420 2.571
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.513 2.556
multiply_cannon_sync_h2d 24660 15.4 2.355 2.510 2.355 2.510
potential_pw2rs 110 12.3 0.012 0.013 2.415 2.423
grid_collocate_task_list 110 9.6 2.265 2.399 2.265 2.399
qs_energies_init_hamiltonians 11 5.9 0.018 0.026 2.304 2.305
mp_alltoall_d11v 2046 13.8 1.953 2.244 1.953 2.244
cp_fm_cholesky_decompose 22 10.9 2.110 2.171 2.110 2.171
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.869 1.908
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.615 1.713
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.689 1.700
transfer_rs2pw 451 10.6 0.005 0.006 1.514 1.678
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.638 1.651
mp_sum_l 6594 12.7 1.097 1.604 1.097 1.604
multiply_cannon_metrocomm4 20550 15.4 0.060 0.062 0.866 1.583
acc_transpose_blocks_sync 73980 16.4 1.399 1.520 1.399 1.520
mp_irecv_dv 62702 16.1 0.764 1.505 0.764 1.505
mp_allgather_i34 2055 14.4 0.556 1.464 0.556 1.464
mp_alltoall_z22v 1111 16.6 1.371 1.451 1.371 1.451
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=71.693000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=703.272727, yerr=13.369510
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 847.667200E+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 0.032 0.105 62.714 62.730
qs_mol_dyn_low 1 2.0 0.003 0.003 62.314 62.339
qs_forces 11 3.9 0.003 0.004 62.240 62.241
qs_energies 11 4.9 0.012 0.014 58.515 58.520
scf_env_do_scf 11 5.9 0.000 0.001 49.122 49.122
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 39.785 39.785
velocity_verlet 10 3.0 0.010 0.012 34.139 34.142
dbcsr_multiply_generic 2055 12.4 0.110 0.113 24.886 25.083
qs_scf_new_mos 99 7.5 0.001 0.001 24.196 24.250
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.195 24.250
ot_scf_mini 99 9.5 0.002 0.002 22.850 22.877
multiply_cannon 2055 13.4 0.239 0.250 18.453 20.038
multiply_cannon_loop 2055 14.4 0.609 0.625 17.057 17.559
rebuild_ks_matrix 110 8.3 0.000 0.000 12.200 12.254
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.199 12.254
ot_mini 99 10.5 0.001 0.001 11.896 11.915
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.876 10.924
init_scf_loop 11 6.9 0.000 0.000 9.281 9.282
multiply_cannon_multrec 8220 15.4 3.203 4.363 7.792 8.733
mp_waitall_1 103326 16.6 6.750 8.255 6.750 8.255
qs_ot_get_derivative 99 11.5 0.001 0.001 8.112 8.139
prepare_preconditioner 11 7.9 0.000 0.000 7.506 7.509
make_preconditioner 11 8.9 0.000 0.000 7.506 7.509
make_full_inverse_cholesky 11 9.9 0.000 0.000 7.080 7.164
qs_ot_get_p 110 10.4 0.001 0.002 6.833 6.863
sum_up_and_integrate 110 10.3 0.001 0.002 6.659 6.673
integrate_v_rspace 110 11.3 0.003 0.003 6.632 6.646
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.621 6.637
calculate_rho_elec 110 8.6 0.113 0.113 6.620 6.636
init_scf_run 11 5.9 0.000 0.001 6.082 6.082
scf_env_initial_rho_setup 11 6.9 0.001 0.002 6.082 6.082
dbcsr_mm_accdrv_process 17442 15.9 3.318 4.365 4.448 5.457
qs_ot_p2m_diag 48 11.0 0.081 0.085 5.086 5.092
make_m2s 4110 13.4 0.038 0.039 4.782 5.057
make_images 4110 14.4 0.646 0.708 4.652 4.928
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.663 4.664
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 2.958 4.562
pw_transfer 1331 11.6 0.066 0.071 4.457 4.468
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 4.349 4.364
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.909 3.917
cp_fm_diag_elpa_base 48 14.0 3.838 3.868 3.907 3.914
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.776 3.865
apply_single 110 13.6 0.000 0.000 3.776 3.864
cp_fm_cholesky_invert 11 10.9 3.803 3.807 3.803 3.807
ot_diis_step 99 11.5 0.012 0.012 3.743 3.743
fft_wrap_pw1pw2_140 451 13.1 0.719 0.734 3.650 3.664
density_rs2pw 110 9.6 0.004 0.004 3.508 3.613
grid_integrate_task_list 110 12.3 3.381 3.536 3.381 3.536
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.069 3.070
calculate_first_density_matrix 1 7.0 0.000 0.000 3.027 3.028
hybrid_alltoall_any 4261 16.3 0.202 0.866 2.482 2.979
wfi_extrapolate 11 7.9 0.001 0.001 2.949 2.949
make_images_data 4110 15.4 0.041 0.046 2.505 2.945
fft3d_ps 1111 14.6 1.294 1.299 2.894 2.911
calculate_dm_sparse 110 9.5 0.001 0.001 2.792 2.839
qs_energies_init_hamiltonians 11 5.9 0.002 0.004 2.685 2.687
grid_collocate_task_list 110 9.6 2.368 2.467 2.368 2.467
multiply_cannon_sync_h2d 8220 15.4 2.369 2.450 2.369 2.450
potential_pw2rs 110 12.3 0.015 0.015 2.415 2.420
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.370 2.387
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.278 2.311
mp_alltoall_d11v 2046 13.8 2.046 2.252 2.046 2.252
cp_fm_cholesky_decompose 22 10.9 2.221 2.242 2.221 2.242
dbcsr_complete_redistribute 325 12.2 0.599 1.024 2.088 2.197
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.148 2.174
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.789 2.002
multiply_cannon_metrocomm1 8220 15.4 0.022 0.023 1.256 1.680
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.663 1.673
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.655 1.661
mp_allgather_i34 2055 14.4 0.564 1.656 0.564 1.656
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.513 1.623
transfer_rs2pw 451 10.6 0.005 0.005 1.423 1.557
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.336 1.434
acc_transpose_blocks 8220 15.4 0.041 0.043 1.393 1.407
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.379 1.401
mp_alltoall_z22v 1111 16.6 1.367 1.390 1.367 1.390
qs_create_task_list 11 7.9 0.003 0.008 1.222 1.318
generate_qs_task_list 11 8.9 0.373 0.442 1.219 1.316
mp_waitany 9240 13.8 1.164 1.295 1.164 1.295
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=62.730000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=804.363636, yerr=9.138060
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 1.372479E+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 0.019 0.042 96.028 96.030
qs_mol_dyn_low 1 2.0 0.003 0.003 95.627 95.638
qs_forces 11 3.9 0.003 0.004 95.560 95.561
qs_energies 11 4.9 0.011 0.012 91.297 91.303
scf_env_do_scf 11 5.9 0.000 0.001 80.184 80.184
velocity_verlet 10 3.0 0.002 0.002 60.154 60.160
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 49.383 49.384
dbcsr_multiply_generic 2055 12.4 0.162 0.173 31.758 31.948
init_scf_loop 11 6.9 0.000 0.000 30.722 30.724
qs_scf_new_mos 99 7.5 0.001 0.001 30.177 30.225
qs_scf_loop_do_ot 99 8.5 0.001 0.001 30.176 30.224
prepare_preconditioner 11 7.9 0.000 0.000 28.532 28.540
make_preconditioner 11 8.9 0.000 0.000 28.532 28.540
ot_scf_mini 99 9.5 0.002 0.002 28.269 28.330
make_full_inverse_cholesky 11 9.9 0.000 0.000 22.737 27.959
multiply_cannon 2055 13.4 0.334 0.361 23.545 24.468
multiply_cannon_loop 2055 14.4 0.832 0.839 21.560 22.082
cp_fm_upper_to_full 70 14.2 12.738 18.216 12.738 18.216
ot_mini 99 10.5 0.001 0.001 15.439 15.501
rebuild_ks_matrix 110 8.3 0.000 0.001 14.828 14.904
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 14.828 14.903
qs_ks_update_qs_env 110 7.6 0.001 0.001 13.460 13.528
dbcsr_complete_redistribute 325 12.2 1.008 1.049 7.575 10.713
qs_ot_get_derivative 99 11.5 0.001 0.001 10.645 10.699
multiply_cannon_multrec 8220 15.4 4.046 4.210 10.136 10.231
mp_waitall_1 84994 16.7 8.683 9.725 8.683 9.725
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.430 9.565
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.779 8.887
mp_alltoall_i22 605 13.7 5.399 8.531 5.399 8.531
qs_rho_update_rho_low 110 7.6 0.001 0.001 8.134 8.164
calculate_rho_elec 110 8.6 0.222 0.223 8.133 8.163
sum_up_and_integrate 110 10.3 0.002 0.002 7.522 7.535
integrate_v_rspace 110 11.3 0.003 0.003 7.494 7.508
qs_ot_get_p 110 10.4 0.002 0.002 7.238 7.309
cp_fm_cholesky_invert 11 10.9 6.857 6.862 6.857 6.862
make_m2s 4110 13.4 0.043 0.043 6.097 6.726
init_scf_run 11 5.9 0.000 0.001 6.646 6.646
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.646 6.646
make_images 4110 14.4 0.882 0.924 5.907 6.535
pw_transfer 1331 11.6 0.075 0.075 6.400 6.407
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 6.283 6.290
dbcsr_mm_accdrv_process 11614 15.7 3.983 4.650 5.941 6.156
multiply_cannon_metrocomm3 8220 15.4 0.019 0.020 5.324 5.763
fft_wrap_pw1pw2_140 451 13.1 1.367 1.370 5.400 5.409
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.844 5.309
apply_single 110 13.6 0.000 0.000 4.844 5.309
qs_ot_p2m_diag 48 11.0 0.151 0.156 5.301 5.308
ot_diis_step 99 11.5 0.015 0.016 4.751 4.752
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.727 4.727
density_rs2pw 110 9.6 0.004 0.004 4.482 4.500
make_images_data 4110 15.4 0.046 0.049 3.270 4.150
hybrid_alltoall_any 4261 16.3 0.265 0.566 3.218 4.067
fft3d_ps 1111 14.6 1.886 1.893 3.986 3.994
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.979 3.980
cp_fm_diag_elpa_base 48 14.0 3.418 3.624 3.977 3.977
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.339 3.803
grid_integrate_task_list 110 12.3 3.693 3.755 3.693 3.755
qs_energies_init_hamiltonians 11 5.9 0.007 0.008 3.727 3.728
wfi_extrapolate 11 7.9 0.001 0.001 3.631 3.631
calculate_dm_sparse 110 9.5 0.001 0.001 3.485 3.529
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.339 3.341
multiply_cannon_sync_h2d 8220 15.4 3.115 3.135 3.115 3.135
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.069 3.101
potential_pw2rs 110 12.3 0.021 0.021 2.924 2.930
calculate_first_density_matrix 1 7.0 0.000 0.000 2.888 2.889
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.749 2.778
grid_collocate_task_list 110 9.6 2.682 2.713 2.682 2.713
cp_fm_cholesky_decompose 22 10.9 2.598 2.615 2.598 2.615
mp_alltoall_d11v 2046 13.8 2.523 2.565 2.523 2.565
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.242 2.305
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.117 2.190
jit_kernel_multiply 10 15.2 1.757 2.112 1.757 2.112
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.969 2.016
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.972 1.979
qs_create_task_list 11 7.9 0.012 0.018 1.906 1.960
generate_qs_task_list 11 8.9 0.732 0.784 1.893 1.949
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.919 1.939
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=96.030000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1236.545455, yerr=60.147340
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430460020736 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1958505086976 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986244964352 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992000282624 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753956716544 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613089636352 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239146475520 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239146475520 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911124992000 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.228663E+12 0.0% 0.0% 100.0%
flops max/rank 1.103326E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806316384 0.0% 0.0% 100.0%
number of processed stacks 12044928 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.1
marketing flops 145.647559E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 630.358016E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10431456
MPI messages size (bytes):
total size 4.526012E+12
min size 0.000000E+00
max size 4.537280E+06
average size 433.881156E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3605352 96403587072
32768 < size <= 131072 1305088 74666999808
131072 < size <= 4194304 5190114 3200148350408
4194304 < size <= 16777216 263934 1154745839544
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4024 57450.
MP_Allreduce 11139 794.
MP_Sync 87
MP_Alltoall 2242 1878090.
MP_SendRecv 24510 18752.
MP_ISendRecv 24510 18752.
MP_Wait 42802
MP_ISend 16140 108019.
MP_IRecv 16140 108019.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.031 223.503 223.516
qs_mol_dyn_low 1 2.0 0.008 0.039 222.729 222.773
qs_forces 11 3.9 0.023 0.028 222.612 222.618
qs_energies 11 4.9 0.002 0.002 216.842 216.880
scf_env_do_scf 11 5.9 0.001 0.001 198.485 198.489
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 175.989 175.991
qs_scf_new_mos 118 7.6 0.001 0.001 134.260 134.542
qs_scf_loop_do_ot 118 8.6 0.001 0.001 134.259 134.541
velocity_verlet 10 3.0 0.010 0.028 132.473 132.474
dbcsr_multiply_generic 2527 12.6 0.177 0.183 128.355 129.092
ot_scf_mini 118 9.6 0.003 0.003 127.490 127.776
multiply_cannon 2527 13.6 0.241 0.257 103.246 105.013
multiply_cannon_loop 2527 14.6 2.454 2.527 100.781 102.506
ot_mini 118 10.6 0.001 0.001 68.513 68.783
multiply_cannon_multrec 60648 15.6 32.058 34.476 41.992 43.710
qs_ot_get_derivative 118 11.6 0.001 0.001 43.296 43.593
qs_ot_get_p 129 10.4 0.001 0.001 36.420 36.783
rebuild_ks_matrix 129 8.3 0.001 0.001 34.790 35.158
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.020 34.789 35.158
mp_waitall_1 269254 16.5 30.351 33.346 30.351 33.346
qs_ks_update_qs_env 129 7.6 0.001 0.001 31.308 31.636
qs_ot_p2m_diag 83 11.4 0.079 0.092 29.264 29.324
multiply_cannon_sync_h2d 60648 15.6 26.427 28.445 26.427 28.445
cp_dbcsr_syevd 83 12.4 0.005 0.005 26.261 26.263
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 24.642 25.590
apply_single 129 13.6 0.001 0.001 24.642 25.590
ot_diis_step 118 11.6 0.007 0.008 24.949 24.950
cp_fm_diag_elpa 83 13.4 0.000 0.000 22.573 22.612
cp_fm_diag_elpa_base 83 14.4 22.465 22.500 22.568 22.607
init_scf_loop 11 6.9 0.000 0.001 22.413 22.415
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 20.744 20.973
multiply_cannon_metrocomm3 60648 15.6 0.123 0.128 16.608 19.190
prepare_preconditioner 11 7.9 0.000 0.000 17.697 17.730
make_preconditioner 11 8.9 0.000 0.000 17.697 17.729
make_full_inverse_cholesky 11 9.9 0.000 0.000 16.948 17.108
sum_up_and_integrate 129 10.3 0.002 0.005 15.072 15.088
integrate_v_rspace 129 11.3 0.003 0.004 15.013 15.032
make_m2s 5054 13.6 0.105 0.114 14.453 14.818
qs_rho_update_rho_low 129 7.7 0.001 0.001 14.516 14.665
calculate_rho_elec 129 8.7 0.045 0.064 14.516 14.665
make_images 5054 14.6 0.409 0.427 14.272 14.642
init_scf_run 11 5.9 0.000 0.001 13.796 13.797
scf_env_initial_rho_setup 11 6.9 0.001 0.001 13.796 13.797
mp_sum_l 8010 12.9 9.484 10.965 9.484 10.965
cp_fm_cholesky_invert 11 10.9 10.272 10.279 10.272 10.279
dbcsr_mm_accdrv_process 125468 16.2 4.825 4.976 9.491 10.049
wfi_extrapolate 11 7.9 0.001 0.001 9.404 9.405
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 8.662 8.820
density_rs2pw 129 9.7 0.006 0.007 8.114 8.732
calculate_dm_sparse 129 9.5 0.001 0.001 8.622 8.714
pw_transfer 1559 11.6 0.075 0.092 8.346 8.559
multiply_cannon_metrocomm1 60648 15.6 0.095 0.099 6.521 8.497
qs_ot_get_orbitals 118 10.6 0.001 0.001 8.330 8.417
fft_wrap_pw1pw2 1301 12.7 0.010 0.012 8.141 8.354
make_images_data 5054 15.6 0.069 0.073 6.801 7.675
grid_integrate_task_list 129 12.3 7.076 7.631 7.076 7.631
hybrid_alltoall_any 5240 16.5 0.300 2.303 5.904 6.972
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.774 6.786
fft_wrap_pw1pw2_140 527 13.2 0.865 0.908 6.509 6.727
fft3d_ps 1301 14.7 2.209 2.851 6.135 6.476
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.753 5.861
mp_alltoall_d11v 2423 14.1 4.413 5.606 4.413 5.606
potential_pw2rs 129 12.3 0.009 0.010 5.512 5.550
cp_fm_cholesky_decompose 22 10.9 5.283 5.296 5.283 5.296
grid_collocate_task_list 129 9.7 4.877 5.268 4.877 5.268
transfer_rs2pw 527 10.6 0.008 0.009 4.128 4.758
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=223.516000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=598.181818, yerr=5.572957
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 2.183246E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5975232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.7
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 847.581184E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100942E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703955E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70860 2317615104
32768 < size <= 131072 722992 55511613440
131072 < size <= 4194304 1375664 1398181724160
4194304 < size <= 16777216 154704 1463834332048
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58357.
MP_Allreduce 11058 960.
MP_Sync 87
MP_Alltoall 1969 5967625.
MP_SendRecv 12032 47072.
MP_ISendRecv 12032 47072.
MP_Wait 25916
MP_ISend 11748 212467.
MP_IRecv 11748 212467.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.037 0.100 199.759 199.778
qs_mol_dyn_low 1 2.0 0.003 0.006 198.584 198.614
qs_forces 11 3.9 0.004 0.005 198.491 198.492
qs_energies 11 4.9 0.006 0.036 191.612 191.631
scf_env_do_scf 11 5.9 0.001 0.002 173.784 173.794
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 138.509 138.511
velocity_verlet 10 3.0 0.001 0.002 124.294 124.295
dbcsr_multiply_generic 2507 12.6 0.188 0.194 98.735 100.324
qs_scf_new_mos 117 7.6 0.001 0.001 99.316 99.826
qs_scf_loop_do_ot 117 8.6 0.001 0.001 99.315 99.825
ot_scf_mini 117 9.6 0.004 0.004 94.467 95.072
multiply_cannon 2507 13.6 0.503 0.562 77.924 82.346
multiply_cannon_loop 2507 14.6 1.575 1.652 74.317 78.040
ot_mini 117 10.6 0.001 0.001 51.455 51.992
mp_waitall_1 214728 16.6 24.510 39.284 24.510 39.284
multiply_cannon_multrec 30084 15.6 20.958 25.261 31.684 36.504
init_scf_loop 11 6.9 0.001 0.006 35.181 35.183
rebuild_ks_matrix 128 8.3 0.001 0.001 32.678 33.297
qs_ks_build_kohn_sham_matrix 128 9.3 0.021 0.054 32.677 33.297
prepare_preconditioner 11 7.9 0.000 0.000 30.808 30.873
make_preconditioner 11 8.9 0.000 0.001 30.808 30.873
multiply_cannon_metrocomm3 30084 15.6 0.101 0.108 15.969 30.794
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.495 30.072
qs_ot_get_derivative 117 11.6 0.001 0.002 29.364 29.956
qs_ks_update_qs_env 128 7.6 0.001 0.001 29.353 29.917
qs_ot_get_p 128 10.4 0.001 0.002 24.895 25.499
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.158 23.299
apply_single 128 13.6 0.001 0.001 22.158 23.299
ot_diis_step 117 11.6 0.014 0.015 21.870 21.872
multiply_cannon_sync_h2d 30084 15.6 17.975 19.869 17.975 19.869
qs_ot_p2m_diag 83 11.4 0.189 0.217 19.738 19.802
cp_dbcsr_syevd 83 12.4 0.005 0.006 18.414 18.416
cp_fm_cholesky_invert 11 10.9 18.243 18.258 18.243 18.258
make_m2s 5014 13.6 0.087 0.093 14.269 15.871
make_images 5014 14.6 1.176 1.374 14.055 15.656
cp_fm_diag_elpa 83 13.4 0.000 0.000 14.861 14.893
cp_fm_diag_elpa_base 83 14.4 14.583 14.711 14.854 14.882
sum_up_and_integrate 128 10.3 0.002 0.005 14.401 14.430
integrate_v_rspace 128 11.3 0.003 0.004 14.341 14.371
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.767 13.819
calculate_rho_elec 128 8.7 0.088 0.105 13.766 13.818
init_scf_run 11 5.9 0.000 0.001 12.633 12.635
scf_env_initial_rho_setup 11 6.9 0.001 0.002 12.633 12.634
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 11.779 12.217
dbcsr_mm_accdrv_process 62242 16.2 5.518 6.354 10.189 10.830
multiply_cannon_metrocomm4 27577 15.6 0.100 0.117 3.850 10.821
mp_irecv_dv 69486 16.3 3.650 10.424 3.650 10.424
make_images_data 5014 15.6 0.068 0.077 8.069 9.904
hybrid_alltoall_any 5200 16.5 0.352 1.509 6.789 9.348
pw_transfer 1547 11.6 0.084 0.093 8.560 8.619
wfi_extrapolate 11 7.9 0.001 0.001 8.513 8.513
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.336 8.401
density_rs2pw 128 9.7 0.006 0.006 7.516 7.886
cp_fm_cholesky_decompose 22 10.9 7.668 7.751 7.668 7.751
grid_integrate_task_list 128 12.3 7.171 7.720 7.171 7.720
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.441 7.295
fft_wrap_pw1pw2_140 523 13.2 0.935 0.957 7.120 7.181
mp_sum_l 7950 12.9 4.903 7.015 4.903 7.015
calculate_dm_sparse 128 9.5 0.001 0.001 6.682 6.861
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.092 6.100
fft3d_ps 1291 14.7 2.831 3.009 5.892 5.967
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.532 5.612
grid_collocate_task_list 128 9.7 4.997 5.481 4.997 5.481
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.188 5.379
potential_pw2rs 128 12.3 0.015 0.017 4.970 4.994
mp_allgather_i34 2507 14.6 2.127 4.869 2.127 4.869
mp_alltoall_d11v 2415 14.1 4.071 4.529 4.071 4.529
mp_sum_d 4467 12.1 2.808 4.172 2.808 4.172
dbcsr_complete_redistribute 395 12.7 0.763 0.845 3.262 4.150
transfer_rs2pw 523 10.6 0.007 0.007 3.652 4.081
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=199.778000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=802.363636, yerr=3.844068
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528903135232 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 2.928533E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 3984192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1695.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 993.284096E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1042912
MPI messages size (bytes):
total size 2.716210E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.604448E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 281856 36943429632
131072 < size <= 4194304 660064 996105256960
4194304 < size <= 16777216 65632 931531265168
16777216 < size 28672 751619276800
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 57952.
MP_Allreduce 11127 998.
MP_Sync 87
MP_Alltoall 1712 9388896.
MP_SendRecv 7936 75008.
MP_ISendRecv 7936 75008.
MP_Wait 21820
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.013 0.035 187.232 187.234
qs_mol_dyn_low 1 2.0 0.003 0.003 186.736 186.751
qs_forces 11 3.9 0.006 0.010 186.642 186.645
qs_energies 11 4.9 0.002 0.003 180.042 180.052
scf_env_do_scf 11 5.9 0.001 0.001 163.434 163.436
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 124.965 124.966
velocity_verlet 10 3.0 0.005 0.012 119.219 119.221
qs_scf_new_mos 117 7.6 0.001 0.001 87.128 87.564
qs_scf_loop_do_ot 117 8.6 0.001 0.001 87.128 87.563
dbcsr_multiply_generic 2507 12.6 0.185 0.189 83.165 84.556
ot_scf_mini 117 9.6 0.003 0.004 82.853 83.321
multiply_cannon 2507 13.6 0.505 0.525 62.547 66.035
multiply_cannon_loop 2507 14.6 1.142 1.206 59.202 62.818
ot_mini 117 10.6 0.001 0.001 43.854 44.369
init_scf_loop 11 6.9 0.000 0.000 38.355 38.356
mp_waitall_1 170520 16.6 24.979 35.638 24.979 35.638
prepare_preconditioner 11 7.9 0.000 0.000 34.279 34.335
make_preconditioner 11 8.9 0.000 0.000 34.279 34.335
make_full_inverse_cholesky 11 9.9 0.000 0.000 31.686 33.295
rebuild_ks_matrix 128 8.3 0.001 0.001 30.581 31.157
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.017 30.581 31.156
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.601 28.128
multiply_cannon_metrocomm3 20056 15.6 0.063 0.067 15.334 25.817
multiply_cannon_multrec 20056 15.6 13.023 16.103 22.372 25.498
qs_ot_get_p 128 10.4 0.001 0.002 24.265 24.803
qs_ot_get_derivative 117 11.6 0.001 0.002 24.158 24.631
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.735 20.883
apply_single 128 13.6 0.001 0.001 19.734 20.883
qs_ot_p2m_diag 83 11.4 0.266 0.273 19.662 19.706
ot_diis_step 117 11.6 0.018 0.018 19.558 19.559
cp_dbcsr_syevd 83 12.4 0.005 0.006 18.456 18.459
cp_fm_cholesky_invert 11 10.9 16.321 16.334 16.321 16.334
make_m2s 5014 13.6 0.078 0.082 14.940 16.079
multiply_cannon_sync_h2d 20056 15.6 13.619 15.945 13.619 15.945
make_images 5014 14.6 1.161 1.249 14.707 15.845
cp_fm_diag_elpa 83 13.4 0.000 0.000 14.809 14.827
cp_fm_diag_elpa_base 83 14.4 14.350 14.529 14.803 14.822
sum_up_and_integrate 128 10.3 0.002 0.003 14.439 14.463
integrate_v_rspace 128 11.3 0.003 0.004 14.379 14.400
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.051 14.110
calculate_rho_elec 128 8.7 0.130 0.145 14.051 14.109
init_scf_run 11 5.9 0.000 0.001 11.567 11.568
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.567 11.567
make_images_data 5014 15.6 0.065 0.074 8.659 10.276
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.767 10.137
hybrid_alltoall_any 5200 16.5 0.451 2.060 7.426 9.437
multiply_cannon_metrocomm4 17549 15.6 0.065 0.074 3.463 9.346
mp_irecv_dv 50230 16.2 3.335 9.079 3.335 9.079
pw_transfer 1547 11.6 0.085 0.104 8.858 8.974
dbcsr_mm_accdrv_process 41502 16.2 5.667 5.951 8.815 8.926
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.635 8.758
cp_fm_cholesky_decompose 22 10.9 8.331 8.391 8.331 8.391
density_rs2pw 128 9.7 0.006 0.006 7.443 7.834
grid_integrate_task_list 128 12.3 7.353 7.828 7.353 7.828
cp_fm_upper_to_full 105 14.8 6.032 7.813 6.032 7.813
wfi_extrapolate 11 7.9 0.001 0.001 7.748 7.748
fft_wrap_pw1pw2_140 523 13.2 1.064 1.102 7.388 7.514
dbcsr_complete_redistribute 395 12.7 1.274 1.586 5.433 7.352
copy_fm_to_dbcsr 209 11.7 0.002 0.002 4.123 6.044
fft3d_ps 1291 14.7 2.825 3.062 5.948 6.042
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.998 6.003
calculate_dm_sparse 128 9.5 0.001 0.001 5.885 5.959
grid_collocate_task_list 128 9.7 5.152 5.715 5.152 5.715
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.799 5.494
mp_alltoall_d11v 2415 14.1 4.393 5.256 4.393 5.256
mp_sum_l 7950 12.9 3.845 5.211 3.845 5.211
potential_pw2rs 128 12.3 0.020 0.021 4.884 4.910
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.629 4.777
mp_allgather_i34 2507 14.6 1.840 4.678 1.840 4.678
transfer_fm_to_dbcsr 11 9.9 0.018 0.023 2.574 4.490
mp_alltoall_i22 716 14.1 2.232 4.223 2.232 4.223
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.159 4.205
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=187.234000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=932.909091, yerr=27.150940
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420242647040 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528903135232 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514766E+12 0.0% 0.0% 100.0%
flops max/rank 4.353791E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755941440 0.0% 0.0% 100.0%
number of processed stacks 5977344 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.167835E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1143192
MPI messages size (bytes):
total size 2.023815E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.770320E+06
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 0 0
8192 < size <= 32768 396 8650752
32768 < size <= 131072 319024 36042702848
131072 < size <= 4194304 715736 785529176064
4194304 < size <= 16777216 70320 665379475120
16777216 < size 30720 536870912000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4002 58205.
MP_Allreduce 11082 1082.
MP_Sync 87
MP_Alltoall 1712 12503107.
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 0.022 0.113 196.884 196.921
qs_mol_dyn_low 1 2.0 0.004 0.008 196.398 196.411
qs_forces 11 3.9 0.005 0.012 196.256 196.266
qs_energies 11 4.9 0.007 0.047 189.071 189.096
scf_env_do_scf 11 5.9 0.004 0.054 171.900 171.911
velocity_verlet 10 3.0 0.002 0.003 129.570 129.575
scf_env_do_scf_inner_loop 117 6.6 0.004 0.014 122.867 122.870
qs_scf_new_mos 117 7.6 0.001 0.001 85.367 85.683
qs_scf_loop_do_ot 117 8.6 0.001 0.001 85.366 85.682
dbcsr_multiply_generic 2507 12.6 0.192 0.197 81.223 81.967
ot_scf_mini 117 9.6 0.004 0.005 80.773 81.082
multiply_cannon 2507 13.6 0.549 0.583 55.790 58.659
multiply_cannon_loop 2507 14.6 1.816 1.907 52.076 53.725
init_scf_loop 11 6.9 0.001 0.005 48.856 48.864
prepare_preconditioner 11 7.9 0.000 0.000 44.624 44.651
make_preconditioner 11 8.9 0.000 0.001 44.624 44.651
ot_mini 117 10.6 0.001 0.003 43.638 43.930
make_full_inverse_cholesky 11 9.9 0.009 0.022 38.167 43.269
multiply_cannon_multrec 30084 15.6 13.715 18.681 26.644 31.270
rebuild_ks_matrix 128 8.3 0.001 0.001 30.092 30.399
qs_ks_build_kohn_sham_matrix 128 9.3 0.021 0.058 30.091 30.399
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.185 27.463
mp_waitall_1 147882 16.7 17.772 26.813 17.772 26.813
qs_ot_get_derivative 117 11.6 0.001 0.002 24.285 24.600
qs_ot_get_p 128 10.4 0.002 0.003 23.170 23.483
make_m2s 5014 13.6 0.093 0.097 20.790 21.549
make_images 5014 14.6 1.930 2.107 20.478 21.243
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 18.655 19.269
apply_single 128 13.6 0.001 0.001 18.655 19.269
ot_diis_step 117 11.6 0.018 0.021 19.212 19.213
qs_ot_p2m_diag 83 11.4 0.343 0.390 18.687 18.736
cp_fm_cholesky_invert 11 10.9 17.653 17.663 17.653 17.663
cp_fm_upper_to_full 105 14.8 11.705 17.238 11.705 17.238
cp_dbcsr_syevd 83 12.4 0.005 0.007 17.230 17.232
multiply_cannon_metrocomm3 30084 15.6 0.049 0.052 6.219 14.875
sum_up_and_integrate 128 10.3 0.002 0.004 14.541 14.570
integrate_v_rspace 128 11.3 0.003 0.004 14.481 14.512
qs_rho_update_rho_low 128 7.7 0.001 0.002 14.405 14.452
calculate_rho_elec 128 8.7 0.171 0.187 14.404 14.450
cp_fm_diag_elpa 83 13.4 0.000 0.000 13.692 13.704
cp_fm_diag_elpa_base 83 14.4 12.672 13.012 13.685 13.695
dbcsr_complete_redistribute 395 12.7 1.473 1.601 9.367 13.307
dbcsr_mm_accdrv_process 62264 16.2 8.510 9.288 12.498 12.985
make_images_data 5014 15.6 0.066 0.072 11.201 12.560
copy_fm_to_dbcsr 209 11.7 0.001 0.002 7.876 11.764
hybrid_alltoall_any 5200 16.5 0.522 2.193 9.729 11.690
multiply_cannon_sync_h2d 30084 15.6 10.479 11.485 10.479 11.485
init_scf_run 11 5.9 0.000 0.001 11.038 11.039
scf_env_initial_rho_setup 11 6.9 0.001 0.004 11.038 11.039
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 10.271 10.517
transfer_fm_to_dbcsr 11 9.9 0.002 0.009 6.434 10.231
mp_alltoall_i22 716 14.1 5.775 9.701 5.775 9.701
pw_transfer 1547 11.6 0.086 0.102 9.528 9.636
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 9.302 9.415
fft_wrap_pw1pw2_140 523 13.2 1.247 1.296 8.159 8.283
cp_fm_cholesky_decompose 22 10.9 8.174 8.243 8.174 8.243
density_rs2pw 128 9.7 0.006 0.006 7.444 7.986
grid_integrate_task_list 128 12.3 7.497 7.901 7.497 7.901
wfi_extrapolate 11 7.9 0.001 0.004 7.897 7.897
multiply_cannon_metrocomm4 25070 15.6 0.079 0.089 2.792 6.696
fft3d_ps 1291 14.7 2.997 3.065 6.353 6.434
mp_irecv_dv 76098 16.2 2.643 6.431 2.643 6.431
calculate_dm_sparse 128 9.5 0.001 0.001 6.297 6.370
mp_alltoall_d11v 2415 14.1 5.157 6.125 5.157 6.125
grid_collocate_task_list 128 9.7 5.288 5.771 5.288 5.771
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.455 5.501
potential_pw2rs 128 12.3 0.023 0.024 4.789 4.814
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.639 4.731
qs_energies_init_hamiltonians 11 5.9 0.003 0.017 4.476 4.478
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.351 4.425
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.341 4.424
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=196.921000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1103.090909, yerr=19.284730
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022950912 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444706349056 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019187724288 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019187724288 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796577E+12 0.0% 0.0% 100.0%
flops max/rank 5.820059E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705500544 0.0% 0.0% 100.0%
number of processed stacks 1944496 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3448.5
marketing flops 143.507742E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 1.539453E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 238560
MPI messages size (bytes):
total size 1.321104E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.537828E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 112800 59139686400
4194304 < size <= 16777216 104112 545846722560
16777216 < size 20064 716108613552
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8852 52.
MP_Alltoall 9584 804353.
MP_ISend 39716 2104723.
MP_IRecv 39716 2103824.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3949 58969.
MP_Allreduce 10950 1171.
MP_Sync 86
MP_Alltoall 1700 18828146.
MP_SendRecv 3810 122880.
MP_ISendRecv 3810 122880.
MP_Wait 16000
MP_ISend 10600 423612.
MP_IRecv 10600 423612.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.037 178.749 178.751
qs_mol_dyn_low 1 2.0 0.003 0.003 178.316 178.367
qs_forces 11 3.9 0.004 0.004 178.219 178.221
qs_energies 11 4.9 0.001 0.002 170.732 170.743
scf_env_do_scf 11 5.9 0.001 0.001 153.183 153.194
velocity_verlet 10 3.0 0.002 0.003 116.331 116.335
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 115.282 115.283
qs_scf_new_mos 116 7.6 0.001 0.001 77.229 77.346
qs_scf_loop_do_ot 116 8.6 0.001 0.001 77.228 77.346
dbcsr_multiply_generic 2485 12.5 0.183 0.189 73.618 74.017
ot_scf_mini 116 9.6 0.003 0.004 72.731 72.806
multiply_cannon 2485 13.5 0.557 0.577 54.381 57.452
multiply_cannon_loop 2485 14.5 0.809 0.846 51.003 51.741
ot_mini 116 10.6 0.001 0.001 38.948 39.031
init_scf_loop 11 6.9 0.000 0.000 37.742 37.743
prepare_preconditioner 11 7.9 0.000 0.000 33.786 33.813
make_preconditioner 11 8.9 0.000 0.000 33.786 33.813
mp_waitall_1 124680 16.7 26.175 32.716 26.175 32.716
make_full_inverse_cholesky 11 9.9 0.015 0.026 31.586 31.883
rebuild_ks_matrix 127 8.3 0.001 0.001 29.742 29.845
qs_ks_build_kohn_sham_matrix 127 9.3 0.017 0.017 29.741 29.844
qs_ks_update_qs_env 127 7.6 0.001 0.001 27.116 27.210
multiply_cannon_multrec 9940 15.5 10.226 16.166 17.816 22.261
qs_ot_get_derivative 116 11.6 0.002 0.002 21.714 21.782
qs_ot_get_p 127 10.4 0.002 0.002 20.940 21.018
multiply_cannon_metrocomm3 9940 15.5 0.024 0.025 13.026 20.200
cp_fm_cholesky_invert 11 10.9 19.590 19.599 19.590 19.599
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 17.265 17.477
apply_single 127 13.6 0.001 0.001 17.265 17.477
ot_diis_step 116 11.6 0.019 0.020 17.137 17.137
qs_ot_p2m_diag 82 11.4 0.490 0.495 16.947 16.966
make_m2s 4970 13.5 0.065 0.069 15.194 16.213
make_images 4970 14.5 2.156 2.581 14.891 15.912
cp_dbcsr_syevd 82 12.4 0.005 0.006 15.655 15.659
qs_rho_update_rho_low 127 7.7 0.001 0.001 14.830 14.865
calculate_rho_elec 127 8.7 0.252 0.263 14.829 14.864
sum_up_and_integrate 127 10.3 0.002 0.002 14.699 14.749
integrate_v_rspace 127 11.3 0.004 0.004 14.640 14.691
cp_fm_diag_elpa 82 13.4 0.000 0.000 12.471 12.480
cp_fm_diag_elpa_base 82 14.4 12.223 12.312 12.465 12.474
multiply_cannon_sync_h2d 9940 15.5 10.583 11.268 10.583 11.268
init_scf_run 11 5.9 0.000 0.001 10.662 10.662
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.662 10.662
make_images_data 4970 15.5 0.055 0.065 8.708 10.329
pw_transfer 1535 11.6 0.084 0.093 10.232 10.267
fft_wrap_pw1pw2 1281 12.7 0.010 0.010 10.011 10.051
hybrid_alltoall_any 5155 16.4 0.837 3.802 8.451 9.744
fft_wrap_pw1pw2_140 519 13.2 1.655 1.691 8.711 8.751
qs_ot_get_derivative_diag 76 12.4 0.002 0.003 8.627 8.683
cp_fm_cholesky_decompose 22 10.9 8.446 8.551 8.446 8.551
multiply_cannon_metrocomm1 9940 15.5 0.030 0.030 4.946 8.117
grid_integrate_task_list 127 12.3 7.631 8.081 7.631 8.081
dbcsr_mm_accdrv_process 20590 16.1 3.202 4.480 7.243 7.968
density_rs2pw 127 9.7 0.005 0.005 7.436 7.700
wfi_extrapolate 11 7.9 0.001 0.001 7.600 7.600
fft3d_ps 1281 14.7 3.117 3.235 6.501 6.540
calculate_dm_sparse 127 9.5 0.001 0.001 6.213 6.287
dbcsr_complete_redistribute 393 12.7 2.084 2.120 5.435 5.817
grid_collocate_task_list 127 9.7 5.522 5.771 5.522 5.771
mp_alltoall_d11v 2401 14.1 5.120 5.766 5.120 5.766
multiply_cannon_metrocomm4 7455 15.5 0.026 0.029 1.861 5.680
mp_irecv_dv 28618 15.9 1.821 5.589 1.821 5.589
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 5.232 5.233
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.131 5.143
mp_allgather_i34 2485 14.5 1.570 4.916 1.570 4.916
potential_pw2rs 127 12.3 0.026 0.026 4.802 4.822
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.030 4.070
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.834 3.900
copy_fm_to_dbcsr 208 11.6 0.001 0.002 3.518 3.863
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.579 3.847
qs_ot_get_orbitals 116 10.6 0.001 0.001 3.751 3.793
copy_dbcsr_to_fm 185 11.7 0.004 0.004 3.527 3.680
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=178.751000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1457.090909, yerr=17.875603
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430451062784 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986258751488 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1990717341696 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1991997444096 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 11613050155008 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239170400256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239170400256 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 94.260896E+12 0.0% 0.0% 100.0%
flops max/rank 11.789544E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806803520 0.0% 0.0% 100.0%
number of processed stacks 1985072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3429.0
marketing flops 145.678848E+12
-------------------------------------------------------------------------------
# multiplications 2542
max memory usage/rank 3.162436E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 101680
MPI messages size (bytes):
total size 1.145405E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.264802E+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 46168 35869687808
4194304 < size <= 16777216 44720 382939955200
16777216 < size 10176 726592076688
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4020 58882.
MP_Allreduce 11127 1501.
MP_Sync 87
MP_Alltoall 1724 36993678.
MP_SendRecv 1806 218624.
MP_ISendRecv 1806 218624.
MP_Wait 9876
MP_ISend 6456 1080169.
MP_IRecv 6456 1080169.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.036 0.061 306.473 306.476
qs_mol_dyn_low 1 2.0 0.003 0.003 305.907 305.921
qs_forces 11 3.9 0.004 0.005 305.675 305.680
qs_energies 11 4.9 0.002 0.002 296.725 296.733
scf_env_do_scf 11 5.9 0.001 0.001 273.657 273.664
velocity_verlet 10 3.0 0.021 0.026 221.862 221.885
scf_env_do_scf_inner_loop 118 6.6 0.044 0.048 143.144 143.147
init_scf_loop 11 6.9 0.000 0.000 130.224 130.228
prepare_preconditioner 11 7.9 0.000 0.000 125.223 125.245
make_preconditioner 11 8.9 0.000 0.000 125.223 125.245
make_full_inverse_cholesky 11 9.9 0.038 0.039 99.936 122.265
qs_scf_new_mos 118 7.6 0.001 0.001 96.402 96.510
qs_scf_loop_do_ot 118 8.6 0.001 0.001 96.401 96.509
ot_scf_mini 118 9.6 0.004 0.004 91.382 91.415
dbcsr_multiply_generic 2542 12.6 0.218 0.228 84.922 85.471
cp_fm_upper_to_full 105 14.8 54.234 78.537 54.234 78.537
multiply_cannon 2542 13.6 0.693 0.750 60.452 61.588
multiply_cannon_loop 2542 14.6 1.077 1.108 56.115 57.743
ot_mini 118 10.6 0.001 0.001 46.479 46.525
dbcsr_complete_redistribute 395 12.7 3.982 4.037 31.514 45.333
copy_fm_to_dbcsr 209 11.7 0.001 0.002 27.864 41.624
transfer_fm_to_dbcsr 11 9.9 0.030 0.030 25.243 38.778
mp_alltoall_i22 718 14.1 22.971 36.833 22.971 36.833
rebuild_ks_matrix 129 8.3 0.001 0.001 35.876 35.950
qs_ks_build_kohn_sham_matrix 129 9.3 0.017 0.017 35.875 35.949
cp_fm_cholesky_invert 11 10.9 35.806 35.815 35.806 35.815
mp_waitall_1 105100 16.8 29.552 34.592 29.552 34.592
qs_ks_update_qs_env 129 7.6 0.001 0.001 33.309 33.385
qs_ot_get_p 129 10.4 0.002 0.002 29.364 29.385
qs_ot_get_derivative 118 11.6 0.002 0.002 26.542 26.576
qs_ot_p2m_diag 83 11.4 0.880 0.886 24.843 24.885
cp_dbcsr_syevd 83 12.4 0.006 0.006 22.908 22.911
multiply_cannon_metrocomm3 10168 15.6 0.025 0.026 19.645 21.341
make_m2s 5084 13.6 0.075 0.077 19.459 21.182
make_images 5084 14.6 3.096 3.293 18.974 20.698
ot_diis_step 118 11.6 0.022 0.022 19.889 19.889
cp_fm_diag_elpa 83 13.4 0.000 0.000 19.467 19.468
cp_fm_diag_elpa_base 83 14.4 14.895 16.641 19.462 19.463
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.112 19.304
apply_single 129 13.6 0.001 0.001 19.112 19.304
multiply_cannon_multrec 10168 15.6 10.423 12.265 18.275 18.352
qs_rho_update_rho_low 129 7.7 0.001 0.001 18.291 18.308
calculate_rho_elec 129 8.7 0.482 0.482 18.290 18.307
sum_up_and_integrate 129 10.3 0.002 0.002 17.266 17.360
integrate_v_rspace 129 11.3 0.004 0.004 17.203 17.298
multiply_cannon_sync_h2d 10168 15.6 14.396 14.424 14.396 14.424
pw_transfer 1559 11.6 0.096 0.097 14.160 14.171
fft_wrap_pw1pw2 1301 12.7 0.012 0.012 13.919 13.931
hybrid_alltoall_any 5270 16.5 1.328 3.136 11.074 13.655
make_images_data 5084 15.6 0.065 0.071 10.986 13.650
init_scf_run 11 5.9 0.000 0.001 12.610 12.612
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.609 12.612
fft_wrap_pw1pw2_140 527 13.2 3.265 3.359 12.273 12.292
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 10.486 10.509
cp_fm_cholesky_decompose 22 10.9 9.848 9.869 9.848 9.869
dbcsr_mm_accdrv_process 20986 16.1 4.198 6.068 7.598 9.508
wfi_extrapolate 11 7.9 0.001 0.001 9.422 9.422
density_rs2pw 129 9.7 0.005 0.005 9.039 9.090
grid_integrate_task_list 129 12.3 8.626 8.852 8.626 8.852
fft3d_ps 1301 14.7 4.012 4.039 8.315 8.383
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 8.227 8.228
mp_alltoall_d11v 2423 14.1 7.649 7.791 7.649 7.791
calculate_dm_sparse 129 9.5 0.001 0.001 6.903 6.993
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.534 6.614
grid_collocate_task_list 129 9.7 6.472 6.570 6.472 6.570
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.368 6.489
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=306.476000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2863.363636, yerr=147.498088
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 1.260745E+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 39548608.
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 0.013 0.029 85.612 85.613
qs_energies 1 2.0 0.000 0.000 85.132 85.139
ls_scf 1 3.0 0.000 0.000 84.230 84.237
dbcsr_multiply_generic 111 6.7 0.015 0.016 73.075 73.209
multiply_cannon 111 7.7 0.018 0.020 56.289 57.561
multiply_cannon_loop 111 8.7 0.229 0.246 52.817 54.157
ls_scf_main 1 4.0 0.000 0.000 52.627 52.631
density_matrix_trs4 2 5.0 0.002 0.003 47.000 47.061
ls_scf_init_scf 1 4.0 0.000 0.000 28.538 28.539
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.391 27.454
mp_waitall_1 11031 10.9 22.942 26.316 22.942 26.316
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.242 25.256
multiply_cannon_multrec 2664 9.7 8.137 8.889 15.629 17.294
multiply_cannon_sync_h2d 2664 9.7 13.354 15.048 13.354 15.048
make_m2s 222 7.7 0.009 0.012 13.082 13.625
make_images 222 8.7 0.100 0.111 13.059 13.606
multiply_cannon_metrocomm1 2664 9.7 0.011 0.012 10.191 13.196
multiply_cannon_metrocomm3 2664 9.7 0.010 0.012 5.436 8.343
make_images_data 222 9.7 0.005 0.006 7.653 8.231
dbcsr_mm_accdrv_process 4760 10.4 0.593 0.703 7.108 8.079
hybrid_alltoall_any 227 10.6 0.218 1.854 6.523 7.798
dbcsr_mm_accdrv_process_sort 4760 11.4 6.316 7.206 6.316 7.206
calculate_norms 4752 9.8 5.471 6.094 5.471 6.094
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.094 5.233
mp_sum_l 887 5.1 3.160 4.566 3.160 4.566
make_images_sizes 222 9.7 0.000 0.000 0.761 3.423
mp_alltoall_i44 222 10.7 0.761 3.422 0.761 3.422
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.393 3.409
arnoldi_extremal 4 6.8 0.000 0.000 3.294 3.320
arnoldi_normal_ev 4 7.8 0.001 0.003 3.293 3.320
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.079 3.274
mp_irecv_dv 6231 10.9 2.062 3.248 2.062 3.248
build_subspace 16 8.4 0.009 0.012 3.198 3.200
ls_scf_post 1 4.0 0.000 0.000 3.065 3.073
ls_scf_store_result 1 5.0 0.000 0.000 2.890 2.927
dbcsr_special_finalize 555 9.7 0.005 0.006 2.307 2.831
dbcsr_merge_single_wm 555 10.7 0.471 0.587 2.299 2.823
make_images_pack 222 9.7 2.211 2.625 2.213 2.627
dbcsr_matrix_vector_mult 304 9.0 0.006 0.014 2.344 2.579
dbcsr_sort_data 658 11.4 2.086 2.513 2.086 2.513
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.342 2.461
dbcsr_matrix_vector_mult_local 304 10.0 2.062 2.453 2.065 2.455
buffer_matrices_ensure_size 222 8.7 1.768 2.058 1.768 2.058
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.813 1.819
rebuild_ks_matrix 3 7.3 0.000 0.000 1.803 1.809
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.803 1.809
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=85.613000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1141.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 2.158801E+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 0.060 0.170 89.960 90.013
qs_energies 1 2.0 0.004 0.028 89.055 89.101
ls_scf 1 3.0 0.004 0.017 87.717 87.767
dbcsr_multiply_generic 111 6.7 0.015 0.016 73.409 73.662
multiply_cannon 111 7.7 0.027 0.037 52.509 56.645
ls_scf_main 1 4.0 0.002 0.030 53.898 53.909
multiply_cannon_loop 111 8.7 0.136 0.146 49.738 52.851
density_matrix_trs4 2 5.0 0.002 0.004 48.248 48.454
ls_scf_init_scf 1 4.0 0.001 0.003 30.074 30.093
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.697 28.798
mp_waitall_1 9105 10.9 19.827 28.529 19.827 28.529
multiply_cannon_multrec 1332 9.7 13.303 16.651 22.684 26.721
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 26.288 26.304
multiply_cannon_metrocomm3 1332 9.7 0.007 0.009 11.147 20.390
make_m2s 222 7.7 0.007 0.008 14.513 15.027
make_images 222 8.7 1.381 1.704 14.483 14.998
dbcsr_mm_accdrv_process 4041 10.4 0.347 0.530 8.981 10.534
dbcsr_mm_accdrv_process_sort 4041 11.4 8.494 10.012 8.494 10.012
make_images_data 222 9.7 0.004 0.005 8.160 8.978
hybrid_alltoall_any 227 10.6 0.545 2.566 7.499 8.732
mp_sum_l 887 5.1 5.015 7.845 5.015 7.845
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.199 7.714
mp_irecv_dv 3311 11.0 3.179 7.661 3.179 7.661
calculate_norms 2376 9.8 6.089 6.710 6.089 6.710
multiply_cannon_sync_h2d 1332 9.7 4.938 6.252 4.938 6.252
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.681 6.102
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.929 5.128
arnoldi_extremal 4 6.8 0.000 0.000 4.888 4.909
arnoldi_normal_ev 4 7.8 0.001 0.005 4.888 4.909
build_subspace 16 8.4 0.014 0.021 4.635 4.637
ls_scf_post 1 4.0 0.004 0.018 3.742 3.779
ls_scf_store_result 1 5.0 0.000 0.000 3.409 3.541
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.256 3.513
dbcsr_matrix_vector_mult_local 304 10.0 2.778 3.275 2.780 3.277
mp_allgather_i34 111 8.7 0.748 2.704 0.748 2.704
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.570 2.669
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.170 2.645
dbcsr_data_new 4174 10.1 2.118 2.407 2.118 2.407
make_images_pack 222 9.7 1.819 2.127 1.822 2.130
dbcsr_sort_data 436 11.2 1.832 2.076 1.832 2.076
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.966 1.986
rebuild_ks_matrix 3 7.3 0.000 0.000 1.953 1.974
qs_ks_build_kohn_sham_matrix 3 8.3 0.018 0.056 1.953 1.974
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=90.013000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1774.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 2.857902E+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 0.024 0.044 92.458 92.459
qs_energies 1 2.0 0.000 0.000 91.887 91.891
ls_scf 1 3.0 0.000 0.000 90.460 90.464
dbcsr_multiply_generic 111 6.7 0.016 0.016 74.969 75.214
ls_scf_main 1 4.0 0.000 0.000 56.298 56.302
multiply_cannon 111 7.7 0.033 0.082 51.983 56.012
multiply_cannon_loop 111 8.7 0.118 0.133 49.268 53.448
density_matrix_trs4 2 5.0 0.002 0.003 50.376 50.559
mp_waitall_1 7281 11.0 22.982 32.655 22.982 32.655
ls_scf_init_scf 1 4.0 0.000 0.000 30.499 30.502
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.291 29.384
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 26.939 26.953
multiply_cannon_multrec 888 9.7 12.712 15.505 21.396 24.803
multiply_cannon_metrocomm3 888 9.7 0.004 0.005 10.881 22.391
make_m2s 222 7.7 0.006 0.008 16.014 16.709
make_images 222 8.7 1.589 1.867 15.976 16.668
make_images_data 222 9.7 0.004 0.005 9.097 10.172
hybrid_alltoall_any 227 10.6 0.644 2.964 8.707 9.752
dbcsr_mm_accdrv_process 3754 10.4 0.293 0.510 8.216 9.510
dbcsr_mm_accdrv_process_sort 3754 11.4 7.775 9.000 7.775 9.000
mp_sum_l 887 5.1 5.280 8.709 5.280 8.709
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.649 7.761
multiply_cannon_sync_h2d 888 9.7 6.063 7.201 6.063 7.201
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.145 7.012
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.449 6.685
mp_irecv_dv 2335 11.1 2.434 6.626 2.434 6.626
arnoldi_extremal 4 6.8 0.000 0.000 5.279 5.297
arnoldi_normal_ev 4 7.8 0.001 0.005 5.279 5.297
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.755 5.010
build_subspace 16 8.4 0.014 0.020 4.969 4.976
calculate_norms 1584 9.8 4.363 4.768 4.363 4.768
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.565 3.863
ls_scf_post 1 4.0 0.000 0.000 3.662 3.668
dbcsr_matrix_vector_mult_local 304 10.0 3.053 3.635 3.055 3.637
mp_allgather_i34 111 8.7 0.822 3.508 0.822 3.508
ls_scf_store_result 1 5.0 0.000 0.000 3.420 3.484
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.846 2.956
dbcsr_data_new 4116 9.9 2.111 2.448 2.111 2.448
make_images_sizes 222 9.7 0.000 0.000 1.106 2.327
mp_alltoall_i44 222 10.7 1.105 2.327 1.105 2.327
dbcsr_sort_data 325 11.1 1.842 2.167 1.842 2.167
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.886 1.888
make_images_pack 222 9.7 1.624 1.885 1.628 1.888
rebuild_ks_matrix 3 7.3 0.000 0.000 1.868 1.870
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.868 1.870
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=92.459000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2273.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 3.395174E+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 0.045 97.812 97.813
qs_energies 1 2.0 0.000 0.000 97.205 97.213
ls_scf 1 3.0 0.000 0.000 95.539 95.545
dbcsr_multiply_generic 111 6.7 0.017 0.017 78.798 79.047
ls_scf_main 1 4.0 0.000 0.000 59.314 59.315
multiply_cannon 111 7.7 0.048 0.115 51.671 55.754
density_matrix_trs4 2 5.0 0.002 0.003 53.079 53.211
multiply_cannon_loop 111 8.7 0.153 0.168 46.667 49.310
ls_scf_init_scf 1 4.0 0.000 0.000 32.996 33.000
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.740 31.832
mp_waitall_1 6369 11.0 22.791 29.279 22.791 29.279
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.234 29.246
multiply_cannon_multrec 1332 9.7 14.195 17.265 22.190 24.949
make_m2s 222 7.7 0.008 0.009 21.342 22.745
make_images 222 8.7 3.147 3.599 21.291 22.696
multiply_cannon_metrocomm3 1332 9.7 0.004 0.004 9.020 16.919
make_images_data 222 9.7 0.004 0.005 11.988 13.667
hybrid_alltoall_any 227 10.6 0.796 3.758 11.272 12.970
dbcsr_mm_accdrv_process 3641 10.4 0.286 0.476 7.644 9.190
dbcsr_mm_accdrv_process_sort 3641 11.4 7.204 8.697 7.204 8.697
mp_sum_l 887 5.1 4.188 8.185 4.188 8.185
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.202 6.466
multiply_cannon_sync_h2d 1332 9.7 5.475 6.191 5.475 6.191
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.071 5.945
mp_irecv_dv 3229 10.9 2.047 5.864 2.047 5.864
arnoldi_extremal 4 6.8 0.000 0.000 5.580 5.593
arnoldi_normal_ev 4 7.8 0.001 0.005 5.580 5.593
build_subspace 16 8.4 0.014 0.021 5.239 5.247
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.612 5.204
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.562 4.744
calculate_norms 2376 9.8 4.189 4.523 4.189 4.523
mp_allgather_i34 111 8.7 2.112 4.417 2.112 4.417
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.746 4.089
dbcsr_matrix_vector_mult_local 304 10.0 3.236 3.752 3.238 3.754
dbcsr_sort_data 658 11.4 3.053 3.483 3.053 3.483
ls_scf_post 1 4.0 0.000 0.000 3.228 3.236
dbcsr_special_finalize 555 9.7 0.006 0.007 2.798 3.230
dbcsr_merge_single_wm 555 10.7 0.539 0.669 2.789 3.222
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.089 3.159
ls_scf_store_result 1 5.0 0.000 0.000 2.986 3.031
dbcsr_data_release 10477 10.7 1.597 2.409 1.597 2.409
dbcsr_finalize 304 7.8 0.049 0.061 1.816 2.023
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=97.813000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2751.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 4.738826E+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 0.036 0.054 92.286 92.287
qs_energies 1 2.0 0.000 0.000 91.503 91.556
ls_scf 1 3.0 0.000 0.000 89.550 89.603
dbcsr_multiply_generic 111 6.7 0.017 0.018 70.791 71.009
ls_scf_main 1 4.0 0.000 0.000 56.694 56.695
multiply_cannon 111 7.7 0.097 0.176 52.803 55.991
multiply_cannon_loop 111 8.7 0.088 0.094 50.173 51.771
density_matrix_trs4 2 5.0 0.002 0.003 49.711 49.777
ls_scf_init_scf 1 4.0 0.000 0.000 29.447 29.448
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.189 28.220
mp_waitall_1 5436 11.0 24.257 27.850 24.257 27.850
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 26.091 26.102
multiply_cannon_multrec 444 9.7 13.754 16.114 20.901 22.043
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.402 16.172
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 5.674 14.485
make_m2s 222 7.7 0.005 0.005 13.228 14.205
make_images 222 8.7 2.045 2.485 13.160 14.135
make_images_data 222 9.7 0.003 0.004 7.999 9.505
hybrid_alltoall_any 227 10.6 0.805 3.829 7.855 9.451
multiply_cannon_sync_h2d 444 9.7 6.678 8.367 6.678 8.367
dbcsr_mm_accdrv_process 3003 10.4 0.353 0.406 6.855 7.976
dbcsr_mm_accdrv_process_sort 3003 11.4 6.491 7.581 6.491 7.581
arnoldi_extremal 4 6.8 0.000 0.000 6.102 6.113
arnoldi_normal_ev 4 7.8 0.002 0.005 6.102 6.113
build_subspace 16 8.4 0.015 0.019 5.706 5.719
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.433 4.590
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.289 4.473
mp_sum_l 887 5.1 2.756 4.428 2.756 4.428
dbcsr_matrix_vector_mult_local 304 10.0 3.726 4.180 3.728 4.182
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.573 3.930
mp_irecv_dv 1241 11.2 1.560 3.885 1.560 3.885
calculate_norms 792 9.8 3.596 3.719 3.596 3.719
mp_allgather_i34 111 8.7 1.197 3.709 1.197 3.709
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.532 3.576
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.960 3.546
ls_scf_post 1 4.0 0.000 0.000 3.409 3.462
ls_scf_store_result 1 5.0 0.000 0.000 3.193 3.228
make_images_sizes 222 9.7 0.000 0.000 0.828 2.941
mp_alltoall_i44 222 10.7 0.827 2.941 0.827 2.941
dbcsr_finalize 304 7.8 0.062 0.077 2.205 2.276
dbcsr_data_new 4608 9.7 1.764 2.181 1.764 2.181
dbcsr_merge_all 275 8.9 0.482 0.524 2.061 2.118
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.044 2.045
rebuild_ks_matrix 3 7.3 0.000 0.000 2.011 2.013
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.011 2.013
qs_energies_init_hamiltonians 1 3.0 0.000 0.001 1.936 1.936
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=92.287000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3737.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 8.938930E+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 0.071 0.089 105.797 105.799
qs_energies 1 2.0 0.000 0.000 104.405 104.414
ls_scf 1 3.0 0.000 0.000 101.426 101.434
dbcsr_multiply_generic 111 6.7 0.023 0.027 74.545 74.731
ls_scf_main 1 4.0 0.000 0.000 63.874 63.875
density_matrix_trs4 2 5.0 0.002 0.003 54.704 54.766
multiply_cannon 111 7.7 0.114 0.224 48.193 49.852
multiply_cannon_loop 111 8.7 0.099 0.101 45.271 46.157
ls_scf_init_scf 1 4.0 0.000 0.000 33.748 33.748
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.156 32.182
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 29.463 29.471
mp_waitall_1 4527 11.1 21.408 25.528 21.408 25.528
make_m2s 222 7.7 0.005 0.005 22.481 23.475
make_images 222 8.7 3.577 3.866 22.371 23.362
multiply_cannon_multrec 444 9.7 17.858 18.398 22.558 23.058
hybrid_alltoall_any 227 10.6 1.654 3.628 12.692 15.444
make_images_data 222 9.7 0.004 0.004 12.836 14.928
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 9.983 10.454
multiply_cannon_sync_h2d 444 9.7 8.788 8.843 8.788 8.843
arnoldi_extremal 4 6.8 0.000 0.000 7.645 7.657
arnoldi_normal_ev 4 7.8 0.003 0.009 7.645 7.657
build_subspace 16 8.4 0.025 0.035 7.084 7.098
dbcsr_matrix_vector_mult 304 9.0 0.017 0.035 5.618 5.793
dbcsr_matrix_vector_mult_local 304 10.0 5.115 5.403 5.118 5.406
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.249 5.345
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.909 5.165
dbcsr_mm_accdrv_process 1814 10.4 0.247 0.370 4.520 4.654
dbcsr_mm_accdrv_process_sort 1814 11.4 4.177 4.306 4.177 4.306
ls_scf_post 1 4.0 0.000 0.000 3.804 3.813
make_images_sizes 222 9.7 0.000 0.000 1.496 3.612
mp_alltoall_i44 222 10.7 1.495 3.612 1.495 3.612
ls_scf_store_result 1 5.0 0.000 0.000 3.519 3.557
mp_allgather_i34 111 8.7 1.029 3.381 1.029 3.381
calculate_norms 792 9.8 3.236 3.284 3.236 3.284
dbcsr_finalize 304 7.8 0.082 0.090 3.070 3.115
qs_energies_init_hamiltonians 1 3.0 0.001 0.001 2.949 2.949
dbcsr_complete_redistribute 5 7.6 1.426 1.491 2.761 2.904
dbcsr_merge_all 275 8.9 0.886 0.912 2.854 2.897
matrix_ls_to_qs 2 6.0 0.000 0.000 2.426 2.566
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.502 2.505
dbcsr_sort_data 325 11.1 2.437 2.491 2.437 2.491
rebuild_ks_matrix 3 7.3 0.000 0.000 2.436 2.439
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.436 2.439
mp_sum_l 887 5.1 1.475 2.411 1.475 2.411
dbcsr_new_transposed 4 7.5 0.241 0.251 2.340 2.361
dbcsr_data_new 6591 9.6 1.889 2.342 1.889 2.342
dbcsr_frobenius_norm 74 6.6 2.058 2.138 2.199 2.243
dbcsr_add_d 103 6.2 0.000 0.000 2.127 2.202
dbcsr_add_anytype 103 7.2 0.858 0.890 2.126 2.202
dbcsr_data_release 12724 10.6 1.975 2.197 1.975 2.197
mp_alltoall_d11v 48 9.2 2.078 2.158 2.078 2.158
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=105.799000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=7088.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/d8a4189bb71bd2b243de4999ff8a786f3563fdee_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 594.231296E+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
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 0.043 0.138 266.909 266.921
qs_mol_dyn_low 1 2.0 0.003 0.006 265.609 265.659
qs_forces 5 3.8 0.005 0.058 265.494 265.495
qs_energies 5 4.8 0.006 0.070 261.985 262.066
scf_env_do_scf 5 5.8 0.000 0.003 246.870 246.873
scf_env_do_scf_inner_loop 105 6.6 0.003 0.012 214.667 214.676
qs_scf_new_mos 105 7.6 0.000 0.001 168.563 168.734
qs_scf_loop_do_ot 105 8.6 0.001 0.001 168.563 168.733
ot_scf_mini 105 9.6 0.003 0.005 158.488 158.632
dbcsr_multiply_generic 1445 12.2 0.126 0.133 136.413 137.054
multiply_cannon 1445 13.2 0.275 0.286 115.557 118.382
multiply_cannon_loop 1445 14.2 2.850 3.016 113.232 114.970
velocity_verlet 4 3.0 0.007 0.057 113.894 113.895
qs_ot_get_p 112 10.4 0.001 0.002 65.611 65.881
ot_mini 105 10.6 0.001 0.001 62.178 62.293
qs_ot_p2m_diag 40 11.0 0.020 0.030 53.711 53.819
cp_dbcsr_syevd 40 12.0 0.002 0.004 50.271 50.273
mp_waitall_1 488190 16.1 37.635 44.890 37.635 44.890
multiply_cannon_multrec 69360 15.2 29.528 34.663 39.622 44.862
cp_fm_syevd 40 13.0 0.000 0.002 44.492 44.631
qs_ot_get_derivative 55 11.6 0.001 0.001 40.291 40.416
cp_fm_redistribute_end 40 14.0 19.618 39.154 19.626 39.157
cp_fm_syevd_base 40 14.0 19.520 39.060 19.520 39.060
multiply_cannon_metrocomm3 69360 15.2 0.205 0.215 26.649 35.600
multiply_cannon_sync_h2d 69360 15.2 28.772 32.550 28.772 32.550
init_scf_loop 7 6.6 0.001 0.018 32.160 32.163
rebuild_ks_matrix 110 8.4 0.000 0.000 31.522 31.695
qs_ks_build_kohn_sham_matrix 110 9.4 0.020 0.099 31.521 31.694
qs_ks_update_qs_env 112 7.6 0.001 0.001 28.812 28.968
prepare_preconditioner 7 7.6 0.000 0.000 26.769 26.800
make_preconditioner 7 8.6 0.000 0.011 26.769 26.800
apply_preconditioner_dbcsr 62 12.6 0.000 0.000 23.180 23.463
apply_single 62 13.6 0.000 0.000 23.180 23.463
qs_rho_update_rho_low 110 7.6 0.001 0.002 21.228 21.547
calculate_rho_elec 110 8.6 0.030 0.032 21.228 21.546
ot_new_cg_direction 55 11.6 0.001 0.001 21.214 21.215
make_full_inverse_cholesky 7 9.6 0.000 0.002 18.650 18.719
density_rs2pw 110 9.6 0.005 0.006 15.093 15.594
qs_ot_get_orbitals 105 10.6 0.001 0.001 15.353 15.541
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 14.279 14.397
mp_sum_l 4764 12.2 12.649 13.351 12.649 13.351
pw_transfer 1645 12.4 0.081 0.105 12.774 13.038
fft_wrap_pw1pw2 1425 13.5 0.012 0.016 12.634 12.898
cp_fm_cholesky_invert 7 10.6 12.826 12.840 12.826 12.840
init_scf_run 5 5.8 0.000 0.001 12.268 12.269
scf_env_initial_rho_setup 5 6.8 0.001 0.003 12.268 12.269
calculate_dm_sparse 110 9.5 0.000 0.001 11.775 11.966
transfer_rs2pw 445 10.6 0.007 0.008 11.346 11.848
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 11.135 11.204
dbcsr_mm_accdrv_process 154766 15.8 6.214 6.392 9.963 11.201
fft_wrap_pw1pw2_240 915 15.0 0.850 0.940 10.898 11.146
qs_vxc_create 110 10.4 0.004 0.037 10.885 10.943
check_diag 80 13.5 8.684 8.935 10.459 10.597
sum_up_and_integrate 60 10.3 0.001 0.003 10.467 10.532
integrate_v_rspace 60 11.3 0.002 0.051 10.450 10.516
fft3d_pb 915 16.0 2.406 2.704 9.040 9.361
make_m2s 2890 13.2 0.078 0.085 7.748 8.356
multiply_cannon_metrocomm1 69360 15.2 0.095 0.102 4.927 8.283
make_images 2890 14.2 0.242 0.262 7.643 8.244
acc_transpose_blocks 69360 15.2 0.361 0.382 7.287 7.874
make_full_single_inverse 7 9.6 0.001 0.021 7.801 7.833
xc_rho_set_and_dset_create 110 12.4 0.076 0.098 7.554 7.812
xc_vxc_pw_create 60 11.3 0.039 0.052 7.322 7.382
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.001 7.325 7.340
calculate_first_density_matrix 1 7.0 0.000 0.005 6.961 6.972
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 6.816 6.872
xc_pw_derive 510 13.4 0.006 0.007 6.555 6.624
mp_alltoall_z22v 2340 17.7 6.092 6.555 6.092 6.555
potential_pw2rs 60 12.3 0.002 0.003 6.426 6.498
transfer_rs2pw_30 110 11.6 1.264 1.324 5.519 5.759
mp_sendrecv_dv 168740 12.6 5.418 5.613 5.418 5.613
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=266.921000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=563.600000, yerr=3.322650
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: d8a4189bb71bd2b243de4999ff8a786f3563fdee
Summary: empty
Status: OK