=== 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: c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1, Cray-FFTW 3.3.8.10,
# COSMA 2.6.2, ELPA 2021.11.002, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 5.2.3, LIBVORI 220621, LIBXSMM 1.17, PLUMED 2.8.0,
# SIRIUS 7.3.2, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Author: Matthias Krack (15.09.2022)
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.2
USE_ELPA := 2021.11.002
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 5.2.3
USE_LIBXSMM := 1.17
USE_PLUMED := 2.8.0
#USE_QUIP := b4336484fb65b0e73211a8f920ae4361c7c353fd
USE_SIRIUS := 7.3.2
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.4
HDF5_VER := 1.12.0
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MPI_VERSION=3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
HDF5_VER := $(strip $(HDF5_VER))
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(HDF5_VER)/lib
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__HDF5
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta_prefixed_scalapack.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/01
job id: 41962516
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/02
job id: 41962517
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/03
job id: 41962518
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/04
job id: 41962520
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/05
job id: 41962521
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/06
job id: 41962523
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/07
job id: 41962526
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/08
job id: 41962529
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/09
job id: 41962532
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/10
job id: 41962534
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/11
job id: 41962536
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/12
job id: 41962538
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/13
job id: 41962540
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/14
job id: 41962541
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/15
job id: 41962542
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/16
job id: 41962543
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/17
job id: 41962544
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/18
job id: 41962545
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/19
job id: 41962546
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/20
job id: 41962547
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/21
job id: 41962549
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/22
job id: 41962550
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/23
job id: 41962551
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/24
job id: 41962552
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/25
job id: 41962553
--- 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/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/26
job id: 41962554
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/01/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 15 177869.
MP_Allreduce 344 9.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.036 135.938 135.939
farming_run 1 2.0 135.455 135.456 135.905 135.910
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.450156E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 228 1113141.
MP_Allreduce 489 2263609.
MP_Sync 27
MP_Alltoall 38 9316958.
MP_SendRecv 30 829726.
MP_ISendRecv 135 235435.
MP_Wait 281
MP_comm_split 8
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.032 116.769 116.770
qs_energies 1 2.0 0.000 0.000 116.489 116.492
mp2_main 1 3.0 0.000 0.000 114.470 114.474
mp2_gpw_main 1 4.0 0.020 0.027 113.440 113.443
mp2_ri_gpw_compute_in 1 5.0 0.172 0.173 93.766 94.328
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.004 55.474 56.037
mp2_eri_3c_integrate_gpw 272 7.0 0.154 0.172 41.781 47.320
get_2c_integrals 1 6.0 0.000 0.001 37.257 38.119
integrate_v_rspace 273 8.0 0.437 0.449 25.136 30.278
pw_transfer 6555 10.6 0.372 0.381 27.356 27.973
fft_wrap_pw1pw2 5465 11.4 0.044 0.046 26.031 26.645
grid_integrate_task_list 273 9.0 20.947 26.565 20.947 26.565
fft_wrap_pw1pw2_100 2178 12.4 1.191 1.470 23.594 24.203
rpa_ri_compute_en 1 5.0 0.010 0.012 19.562 19.752
compute_2c_integrals 1 7.0 0.002 0.003 19.726 19.726
compute_2c_integrals_loop_lm 1 8.0 0.004 0.005 18.883 19.384
mp2_eri_2c_integrate_gpw 1 9.0 2.385 2.438 18.879 19.381
cp_fm_cholesky_decompose 12 8.2 17.713 18.586 17.713 18.586
cholesky_decomp 1 7.0 0.000 0.000 16.372 17.221
fft3d_s 5443 13.4 16.123 16.705 16.145 16.726
ao_to_mo_and_store_B_mult_1 272 7.0 10.860 15.598 10.860 15.598
calculate_wavefunction 272 8.0 5.428 5.585 12.558 13.175
rpa_num_int 1 6.0 0.000 0.002 10.816 10.825
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.766 10.781
calc_mat_Q 8 8.0 0.000 0.000 9.374 9.468
contract_S_to_Q 8 9.0 0.000 0.000 8.794 8.886
calc_potential_gpw 544 9.5 0.005 0.006 8.297 8.662
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.002 8.229 8.567
parallel_gemm_fm 14 9.1 0.000 0.000 8.382 8.450
parallel_gemm_fm_cosma 14 10.1 8.382 8.450 8.382 8.450
potential_pw2rs 545 10.0 0.108 0.110 7.666 8.327
create_integ_mat 1 6.0 0.014 0.028 8.078 8.087
collocate_single_gaussian 272 10.0 0.040 0.043 7.442 7.716
array2fm 1 7.0 0.000 0.000 6.640 7.168
pw_scatter_s 2720 13.7 4.453 4.636 4.453 4.636
pw_gather_s 2722 13.2 3.894 4.319 3.894 4.319
array2fm_buffer_send 1 8.0 3.013 3.134 3.013 3.134
pw_poisson_solve 545 10.5 1.135 1.192 2.167 2.353
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=113.442291, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2721.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/02/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 0.000000E+00 0.0% 0.0% 0.0%
flops max/rank 0.000000E+00 0.0% 0.0% 0.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 0 0.0% 0.0% 0.0%
number of processed stacks 0 0.0% 0.0% 0.0%
average stack size 0.0 0.0 0.0
marketing flops 0.000000E+00
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 1 12.
MP_Allreduce 19 21.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 22 205321.
MP_Allreduce 344 10.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.028 0.035 397.662 397.664
farming_run 1 2.0 396.998 397.003 397.623 397.625
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.224897E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 703 408373.
MP_Allreduce 1825 23678.
MP_Sync 38
MP_Alltoall 77
MP_SendRecv 2171 2843495.
MP_ISendRecv 1739 144022.
MP_Wait 2051
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.029 210.541 210.541
qs_energies 1 2.0 0.000 0.000 210.285 210.291
scf_env_do_scf 1 3.0 0.000 0.000 106.419 106.419
qs_ks_update_qs_env 5 5.0 0.000 0.000 105.477 105.485
rebuild_ks_matrix 4 6.0 0.000 0.000 105.476 105.484
qs_ks_build_kohn_sham_matrix 4 7.0 0.057 0.065 105.476 105.484
hfx_ks_matrix 4 8.0 0.001 0.001 105.095 105.099
integrate_four_center 4 9.0 0.144 0.457 105.095 105.098
mp2_main 1 3.0 0.000 0.000 103.584 103.590
mp2_gpw_main 1 4.0 0.045 0.085 102.649 102.659
integrate_four_center_main 4 10.0 0.113 0.503 96.403 99.574
integrate_four_center_bin 264 11.0 96.290 99.225 96.290 99.225
init_scf_loop 1 4.0 0.000 0.000 92.067 92.067
mp2_ri_gpw_compute_in 1 5.0 0.072 0.109 75.356 76.521
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.011 54.619 55.774
mp2_eri_3c_integrate_gpw 91 7.0 0.145 0.163 42.229 47.464
integrate_v_rspace 95 8.0 0.398 0.571 28.582 33.616
pw_transfer 2240 10.6 0.143 0.160 29.882 30.287
fft_wrap_pw1pw2 1868 11.4 0.018 0.019 28.894 29.294
grid_integrate_task_list 95 9.0 23.865 29.099 23.865 29.099
mp2_ri_gpw_compute_en 1 5.0 0.055 0.062 27.114 29.023
ao_to_mo_and_store_B_mult_1 91 7.0 10.705 27.079 10.705 27.079
fft_wrap_pw1pw2_100 730 12.4 1.273 1.459 26.646 27.071
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.836 1.889 25.304 25.315
get_2c_integrals 1 6.0 0.004 0.033 20.354 20.684
compute_2c_integrals 1 7.0 0.004 0.008 19.278 19.301
compute_2c_integrals_loop_lm 1 8.0 0.009 0.039 18.850 19.162
mp2_eri_2c_integrate_gpw 1 9.0 1.755 1.879 18.841 19.161
fft3d_s 1823 13.4 18.434 18.940 18.448 18.952
scf_env_do_scf_inner_loop 4 4.0 0.000 0.000 14.350 14.350
calculate_wavefunction 91 8.0 2.032 2.069 9.756 9.984
mp2_ri_gpw_compute_en_expansio 172 7.0 0.557 0.583 8.748 9.377
potential_pw2rs 186 10.0 0.034 0.035 8.654 9.230
local_gemm 172 8.0 8.191 8.819 8.191 8.819
mp2_ri_gpw_compute_en_comm 22 7.0 0.512 0.540 8.329 8.693
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.227 8.575
collocate_single_gaussian 91 10.0 0.017 0.034 7.869 8.146
calc_potential_gpw 182 9.5 0.002 0.002 7.892 8.077
mp_sendrecv_dm3 2068 8.0 6.333 6.722 6.333 6.722
mp2_ri_gpw_compute_en_ener 172 7.0 6.347 6.432 6.347 6.432
mp_sync 38 10.4 3.560 5.889 3.560 5.889
pw_gather_s 912 13.2 4.922 5.400 4.922 5.400
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=102.641959, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1512.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/03/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 29.277748E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 5055360 0.0% 0.0% 100.0%
average stack size 0.0 0.0 29.1
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 450.904064E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62385.
MP_Allreduce 10249 271.
MP_Sync 580
MP_Alltoall 2083
MP_ISendRecv 45220 5520.
MP_Wait 60486
MP_comm_split 50
MP_ISend 20771 42672.
MP_IRecv 20771 42672.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.026 59.523 59.524
qs_mol_dyn_low 1 2.0 0.003 0.003 59.336 59.344
qs_forces 11 3.9 0.002 0.003 59.231 59.232
qs_energies 11 4.9 0.002 0.002 57.732 57.743
scf_env_do_scf 11 5.9 0.000 0.001 51.527 51.527
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 49.396 49.396
dbcsr_multiply_generic 2286 12.5 0.093 0.096 36.655 37.089
qs_scf_new_mos 108 7.5 0.000 0.001 36.734 37.005
qs_scf_loop_do_ot 108 8.5 0.000 0.001 36.734 37.004
ot_scf_mini 108 9.5 0.002 0.002 34.988 35.161
velocity_verlet 10 3.0 0.001 0.001 29.614 29.615
multiply_cannon 2286 13.5 0.178 0.185 27.328 28.755
multiply_cannon_loop 2286 14.5 1.482 1.549 26.150 27.699
ot_mini 108 10.5 0.001 0.001 21.858 22.125
qs_ot_get_derivative 108 11.5 0.001 0.001 18.510 18.697
mp_waitall_1 267858 16.1 11.090 16.732 11.090 16.732
multiply_cannon_metrocomm3 54864 15.5 0.065 0.070 6.356 13.406
multiply_cannon_multrec 54864 15.5 4.153 6.464 7.410 10.816
rebuild_ks_matrix 119 8.3 0.000 0.000 8.947 9.075
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 8.946 9.074
qs_ot_get_p 119 10.4 0.001 0.001 8.103 8.411
qs_ks_update_qs_env 119 7.6 0.001 0.001 7.993 8.114
mp_sum_l 7207 12.9 5.730 7.470 5.730 7.470
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 6.708 7.152
multiply_cannon_sync_h2d 54864 15.5 5.796 6.991 5.796 6.991
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 6.167 6.265
qs_rho_update_rho_low 119 7.7 0.000 0.001 5.661 5.792
calculate_rho_elec 119 8.7 0.011 0.016 5.661 5.792
sum_up_and_integrate 119 10.3 0.012 0.015 5.350 5.356
integrate_v_rspace 119 11.3 0.002 0.002 5.337 5.344
init_scf_run 11 5.9 0.000 0.001 4.978 4.978
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.978 4.978
dbcsr_mm_accdrv_process 76910 16.1 1.180 1.809 3.182 4.404
multiply_cannon_metrocomm1 54864 15.5 0.051 0.057 2.566 4.015
qs_ot_p2m_diag 50 11.0 0.004 0.006 3.967 4.011
rs_pw_transfer 974 11.9 0.012 0.013 3.747 3.930
make_m2s 4572 13.5 0.053 0.056 3.284 3.409
density_rs2pw 119 9.7 0.004 0.004 3.246 3.337
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.133 3.335
apply_single 119 13.6 0.000 0.000 3.132 3.335
make_images 4572 14.5 0.133 0.138 3.202 3.327
calculate_dm_sparse 119 9.5 0.000 0.000 2.955 3.114
cp_dbcsr_syevd 50 12.0 0.002 0.003 3.085 3.086
ot_diis_step 108 11.5 0.006 0.006 3.024 3.027
calculate_first_density_matrix 1 7.0 0.000 0.001 2.825 2.829
mp_alltoall_d11v 2130 13.8 2.548 2.676 2.548 2.676
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.602 2.671
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.510 2.515
jit_kernel_multiply 13 15.8 1.939 2.513 1.939 2.513
pw_transfer 1439 11.6 0.051 0.056 2.314 2.486
cp_fm_redistribute_end 50 14.0 2.221 2.422 2.230 2.430
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 2.239 2.414
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.402 2.404
potential_pw2rs 119 12.3 0.004 0.004 2.332 2.371
cp_fm_diag_elpa_base 50 14.0 0.192 2.326 0.193 2.336
acc_transpose_blocks 54864 15.5 0.222 0.237 1.856 2.323
fft3d_ps 1201 14.6 0.359 0.459 2.019 2.183
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.123 2.175
mp_sum_d 4129 12.0 1.589 2.149 1.589 2.149
grid_integrate_task_list 119 12.3 2.019 2.120 2.019 2.120
init_scf_loop 11 6.9 0.000 0.000 2.112 2.112
make_images_sizes 4572 15.5 0.004 0.004 1.688 2.107
mp_alltoall_i44 4572 16.5 1.684 2.103 1.684 2.103
wfi_extrapolate 11 7.9 0.001 0.001 2.097 2.097
fft_wrap_pw1pw2_140 487 13.2 0.078 0.092 1.769 1.946
arnoldi_extremal 119 11.4 0.001 0.001 1.422 1.563
arnoldi_normal_ev 119 12.4 0.002 0.002 1.422 1.562
mp_waitany 12084 13.8 1.358 1.550 1.358 1.550
make_images_data 4572 15.5 0.037 0.042 0.931 1.407
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.373 1.398
mp_alltoall_z22v 1201 16.6 1.223 1.373 1.223 1.373
grid_collocate_task_list 119 9.7 1.286 1.350 1.286 1.350
dbcsr_dot_sd 1205 11.9 0.047 0.057 0.865 1.290
rs_scatter_matrices 130 9.7 0.039 0.049 1.116 1.245
acc_transpose_blocks_kernels 54864 16.5 0.249 0.378 0.876 1.204
dbcsr_new_transposed 954 13.3 0.007 0.008 1.028 1.196
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=59.524000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=429.909091, yerr=1.504813
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 486.883328E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62664.
MP_Allreduce 10226 305.
MP_Sync 104
MP_Alltoall 2060 717021.
MP_ISendRecv 33558 37093.
MP_Wait 40318
MP_comm_split 50
MP_ISend 5720 128509.
MP_IRecv 5720 128509.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.035 39.944 39.946
qs_mol_dyn_low 1 2.0 0.003 0.006 39.673 39.682
qs_forces 11 3.9 0.002 0.003 39.600 39.602
qs_energies 11 4.9 0.002 0.004 37.879 37.884
scf_env_do_scf 11 5.9 0.000 0.001 32.451 32.451
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 29.848 29.848
dbcsr_multiply_generic 2286 12.5 0.100 0.104 21.994 22.436
qs_scf_new_mos 108 7.5 0.001 0.001 20.450 20.703
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.450 20.702
ot_scf_mini 108 9.5 0.002 0.003 19.499 19.677
velocity_verlet 10 3.0 0.001 0.002 19.026 19.027
multiply_cannon 2286 13.5 0.209 0.215 16.593 18.194
multiply_cannon_loop 2286 14.5 0.899 0.984 15.389 17.024
ot_mini 108 10.5 0.001 0.002 12.227 12.465
mp_waitall_1 217478 16.2 6.561 11.670 6.561 11.670
qs_ot_get_derivative 108 11.5 0.001 0.003 9.714 9.891
multiply_cannon_metrocomm3 27432 15.5 0.067 0.070 4.256 9.841
multiply_cannon_multrec 27432 15.5 1.969 4.462 5.871 8.572
rebuild_ks_matrix 119 8.3 0.000 0.000 7.413 7.554
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.014 7.412 7.554
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.541 6.670
dbcsr_mm_accdrv_process 47894 16.0 2.937 5.109 3.832 5.667
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.724 4.571
qs_ot_get_p 119 10.4 0.001 0.002 4.305 4.531
sum_up_and_integrate 119 10.3 0.024 0.027 4.420 4.425
integrate_v_rspace 119 11.3 0.002 0.002 4.396 4.402
mp_sum_l 7207 12.9 2.219 4.216 2.219 4.216
apply_preconditioner_dbcsr 119 12.6 0.000 0.001 3.070 4.171
apply_single 119 13.6 0.000 0.000 3.070 4.171
init_scf_run 11 5.9 0.000 0.001 4.083 4.083
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.082 4.083
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.815 3.862
calculate_rho_elec 119 8.7 0.021 0.024 3.814 3.861
rs_pw_transfer 974 11.9 0.010 0.011 2.936 3.093
make_m2s 4572 13.5 0.052 0.054 2.741 3.025
make_images 4572 14.5 0.201 0.237 2.654 2.936
multiply_cannon_sync_h2d 27432 15.5 2.195 2.826 2.195 2.826
qs_ot_p2m_diag 50 11.0 0.009 0.013 2.758 2.779
init_scf_loop 11 6.9 0.000 0.000 2.580 2.581
jit_kernel_multiply 13 16.0 0.842 2.502 0.842 2.502
calculate_first_density_matrix 1 7.0 0.000 0.001 2.483 2.484
ot_diis_step 108 11.5 0.011 0.013 2.451 2.452
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.320 2.415
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.315 2.315
density_rs2pw 119 9.7 0.004 0.004 2.160 2.293
calculate_dm_sparse 119 9.5 0.000 0.000 2.110 2.188
potential_pw2rs 119 12.3 0.006 0.006 2.138 2.157
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 2.063 2.066
grid_integrate_task_list 119 12.3 1.832 1.947 1.832 1.947
pw_transfer 1439 11.6 0.063 0.067 1.905 1.934
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.929 1.930
cp_fm_redistribute_end 50 14.0 1.587 1.898 1.590 1.900
cp_fm_diag_elpa_base 50 14.0 0.300 1.834 0.307 1.863
fft_wrap_pw1pw2 1201 12.6 0.007 0.008 1.815 1.845
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.804 1.841
make_images_data 4572 15.5 0.045 0.051 1.235 1.738
prepare_preconditioner 11 7.9 0.000 0.000 1.670 1.696
make_preconditioner 11 8.9 0.000 0.000 1.670 1.696
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.570 1.626
hybrid_alltoall_any 4725 16.4 0.050 0.111 1.073 1.606
acc_transpose_blocks 27432 15.5 0.106 0.110 1.258 1.594
fft3d_ps 1201 14.6 0.497 0.549 1.521 1.549
wfi_extrapolate 11 7.9 0.001 0.002 1.547 1.547
mp_alltoall_d11v 2130 13.8 1.312 1.445 1.312 1.445
mp_allgather_i34 2286 14.5 0.646 1.435 0.646 1.435
fft_wrap_pw1pw2_140 487 13.2 0.076 0.083 1.359 1.390
grid_collocate_task_list 119 9.7 1.243 1.300 1.243 1.300
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.276 1.284
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.202 1.250
mp_sum_d 4129 12.0 0.634 1.068 0.634 1.068
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 1.049 1.051
acc_transpose_blocks_kernels 27432 16.5 0.180 0.268 0.732 0.994
make_images_sizes 4572 15.5 0.005 0.005 0.695 0.992
mp_alltoall_i44 4572 16.5 0.691 0.987 0.691 0.987
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.929 0.943
rs_pw_transfer_PW2RS_50 119 14.3 0.587 0.606 0.827 0.871
mp_alltoall_z22v 1201 16.6 0.780 0.852 0.780 0.852
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=39.946000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=462.727273, yerr=1.135454
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 520.871936E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62660.
MP_Allreduce 10225 303.
MP_Sync 104
MP_Alltoall 1821 1607811.
MP_ISendRecv 22134 57667.
MP_Wait 33054
MP_comm_split 50
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.033 33.757 33.759
qs_mol_dyn_low 1 2.0 0.003 0.005 33.499 33.507
qs_forces 11 3.9 0.002 0.003 33.432 33.433
qs_energies 11 4.9 0.002 0.004 31.779 31.783
scf_env_do_scf 11 5.9 0.001 0.001 26.751 26.751
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 23.950 23.953
dbcsr_multiply_generic 2286 12.5 0.094 0.098 17.119 17.249
velocity_verlet 10 3.0 0.001 0.003 16.105 16.107
qs_scf_new_mos 108 7.5 0.001 0.001 15.455 15.479
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.455 15.478
ot_scf_mini 108 9.5 0.002 0.003 14.701 14.715
multiply_cannon 2286 13.5 0.195 0.201 13.352 14.241
multiply_cannon_loop 2286 14.5 0.638 0.664 12.444 13.416
ot_mini 108 10.5 0.001 0.001 9.139 9.154
qs_ot_get_derivative 108 11.5 0.001 0.001 7.584 7.596
multiply_cannon_multrec 18288 15.5 1.939 2.903 6.731 7.013
rebuild_ks_matrix 119 8.3 0.000 0.000 6.704 6.737
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.017 6.703 6.736
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.914 5.944
dbcsr_mm_accdrv_process 38222 16.0 4.000 5.277 4.709 5.478
mp_waitall_1 169478 16.3 3.460 4.549 3.460 4.549
sum_up_and_integrate 119 10.3 0.030 0.032 4.250 4.256
integrate_v_rspace 119 11.3 0.002 0.003 4.220 4.230
init_scf_run 11 5.9 0.000 0.001 3.800 3.800
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.799 3.800
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.493 3.499
calculate_rho_elec 119 8.7 0.030 0.031 3.492 3.499
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.800 3.430
qs_ot_get_p 119 10.4 0.001 0.002 3.305 3.326
init_scf_loop 11 6.9 0.001 0.002 2.776 2.783
rs_pw_transfer 974 11.9 0.009 0.010 2.665 2.752
calculate_first_density_matrix 1 7.0 0.000 0.001 2.482 2.485
apply_preconditioner_dbcsr 119 12.6 0.000 0.001 2.067 2.393
apply_single 119 13.6 0.000 0.000 2.067 2.393
multiply_cannon_metrocomm3 18288 15.5 0.045 0.046 1.447 2.329
make_m2s 4572 13.5 0.044 0.046 2.069 2.226
make_images 4572 14.5 0.193 0.210 1.985 2.140
density_rs2pw 119 9.7 0.004 0.004 2.072 2.129
qs_ot_p2m_diag 50 11.0 0.012 0.015 2.099 2.107
potential_pw2rs 119 12.3 0.007 0.008 2.007 2.014
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.945 1.947
prepare_preconditioner 11 7.9 0.000 0.000 1.914 1.916
make_preconditioner 11 8.9 0.000 0.001 1.914 1.916
grid_integrate_task_list 119 12.3 1.806 1.914 1.806 1.914
pw_transfer 1439 11.6 0.063 0.065 1.885 1.897
calculate_dm_sparse 119 9.5 0.000 0.000 1.849 1.865
jit_kernel_multiply 10 16.1 0.657 1.859 0.657 1.859
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.767 1.851
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.815 1.816
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.795 1.809
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 1.763 1.774
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.714 1.728
mp_sum_l 7207 12.9 1.316 1.692 1.316 1.692
multiply_cannon_sync_h2d 18288 15.5 1.373 1.547 1.373 1.547
ot_diis_step 108 11.5 0.012 0.015 1.524 1.524
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.519 1.523
fft3d_ps 1201 14.6 0.507 0.522 1.482 1.499
cp_fm_redistribute_end 50 14.0 1.124 1.490 1.126 1.492
acc_transpose_blocks 18288 15.5 0.074 0.076 1.402 1.446
cp_fm_diag_elpa_base 50 14.0 0.346 1.402 0.360 1.444
fft_wrap_pw1pw2_140 487 13.2 0.085 0.089 1.352 1.368
grid_collocate_task_list 119 9.7 1.203 1.285 1.203 1.285
wfi_extrapolate 11 7.9 0.002 0.006 1.232 1.233
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.189 1.196
make_images_data 4572 15.5 0.045 0.049 0.904 1.109
multiply_cannon_metrocomm1 18288 15.5 0.029 0.031 0.458 1.076
acc_transpose_blocks_kernels 18288 16.5 0.209 0.217 0.963 1.000
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 0.981 0.982
hybrid_alltoall_any 4725 16.4 0.054 0.112 0.775 0.979
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.956 0.978
mp_alltoall_d11v 2130 13.8 0.784 0.884 0.784 0.884
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.863 0.866
mp_alltoall_z22v 1201 16.6 0.798 0.863 0.798 0.863
cp_fm_cholesky_invert 11 10.9 0.834 0.838 0.834 0.838
jit_kernel_transpose 5 15.6 0.754 0.790 0.754 0.790
mp_allgather_i34 2286 14.5 0.368 0.773 0.368 0.773
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.653 0.731
mp_waitany 9880 13.7 0.585 0.680 0.585 0.680
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=33.759000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=495.272727, yerr=1.656442
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 555.802624E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62659.
MP_Allreduce 10224 344.
MP_Sync 104
MP_Alltoall 1582 2412273.
MP_ISendRecv 16422 74133.
MP_Wait 24482
MP_comm_split 50
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.030 41.026 41.027
qs_mol_dyn_low 1 2.0 0.003 0.004 40.806 40.814
qs_forces 11 3.9 0.002 0.003 40.697 40.700
qs_energies 11 4.9 0.002 0.003 37.752 37.760
scf_env_do_scf 11 5.9 0.001 0.001 31.809 31.810
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 27.235 27.243
velocity_verlet 10 3.0 0.001 0.001 21.120 21.123
dbcsr_multiply_generic 2286 12.5 0.100 0.104 19.751 19.884
qs_scf_new_mos 108 7.5 0.001 0.001 18.190 18.239
qs_scf_loop_do_ot 108 8.5 0.001 0.001 18.189 18.238
ot_scf_mini 108 9.5 0.002 0.003 17.084 17.135
multiply_cannon 2286 13.5 0.225 0.249 15.128 15.514
multiply_cannon_loop 2286 14.5 0.934 0.966 13.972 14.264
ot_mini 108 10.5 0.001 0.001 9.653 9.718
multiply_cannon_multrec 27432 15.5 2.335 3.045 8.587 8.945
qs_ot_get_derivative 108 11.5 0.001 0.001 7.825 7.878
rebuild_ks_matrix 119 8.3 0.000 0.000 7.381 7.440
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.015 7.380 7.440
dbcsr_mm_accdrv_process 47916 15.9 5.285 7.074 6.156 7.370
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.922 5.967
qs_ot_get_p 119 10.4 0.001 0.001 4.718 4.802
init_scf_loop 11 6.9 0.001 0.002 4.552 4.561
init_scf_run 11 5.9 0.000 0.001 4.203 4.204
scf_env_initial_rho_setup 11 6.9 0.001 0.001 4.203 4.204
sum_up_and_integrate 119 10.3 0.034 0.038 4.197 4.203
integrate_v_rspace 119 11.3 0.002 0.005 4.162 4.169
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.068 4.102
calculate_rho_elec 119 8.7 0.040 0.046 4.068 4.101
mp_waitall_1 145218 16.4 3.458 3.930 3.458 3.930
prepare_preconditioner 11 7.9 0.000 0.000 3.258 3.267
make_preconditioner 11 8.9 0.000 0.002 3.258 3.267
qs_ot_p2m_diag 50 11.0 0.015 0.023 3.202 3.228
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.805 3.197
make_m2s 4572 13.5 0.053 0.055 2.994 3.096
make_images 4572 14.5 0.274 0.332 2.888 2.991
rs_pw_transfer 974 11.9 0.009 0.009 2.588 2.685
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.663 2.666
density_rs2pw 119 9.7 0.003 0.004 2.441 2.560
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.186 2.520
calculate_first_density_matrix 1 7.0 0.000 0.000 2.443 2.445
calculate_dm_sparse 119 9.5 0.000 0.000 2.196 2.255
pw_transfer 1439 11.6 0.062 0.066 2.202 2.233
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.102 2.229
apply_single 119 13.6 0.000 0.000 2.102 2.229
dbcsr_complete_redistribute 329 12.2 0.185 0.229 1.915 2.186
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.178 2.181
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.113 2.145
mp_alltoall_d11v 2130 13.8 1.967 2.076 1.967 2.076
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.998 2.028
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.893 1.927
cp_fm_redistribute_end 50 14.0 1.327 1.896 1.335 1.898
grid_integrate_task_list 119 12.3 1.817 1.883 1.817 1.883
copy_fm_to_dbcsr 176 11.2 0.001 0.002 1.605 1.877
fft3d_ps 1201 14.6 0.534 0.585 1.794 1.822
jit_kernel_multiply 10 16.0 0.809 1.819 0.809 1.819
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 1.790 1.805
ot_diis_step 108 11.5 0.012 0.012 1.787 1.787
fft_wrap_pw1pw2_140 487 13.2 0.084 0.091 1.731 1.765
cp_fm_diag_elpa_base 50 14.0 0.506 1.667 0.527 1.727
wfi_extrapolate 11 7.9 0.001 0.001 1.710 1.711
potential_pw2rs 119 12.3 0.008 0.010 1.639 1.646
acc_transpose_blocks 27432 15.5 0.110 0.114 1.563 1.593
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.550 1.565
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.479 1.484
mp_sum_l 7207 12.9 1.206 1.480 1.206 1.480
make_images_data 4572 15.5 0.045 0.048 1.272 1.430
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.377 1.405
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.068 1.340
grid_collocate_task_list 119 9.7 1.220 1.310 1.220 1.310
multiply_cannon_metrocomm3 27432 15.5 0.038 0.039 0.799 1.289
hybrid_alltoall_any 4725 16.4 0.061 0.150 1.051 1.275
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.210 1.219
mp_alltoall_z22v 1201 16.6 1.143 1.168 1.143 1.168
cp_fm_upper_to_full 72 13.5 0.828 1.151 0.828 1.151
multiply_cannon_sync_h2d 27432 15.5 0.990 1.064 0.990 1.064
multiply_cannon_metrocomm1 27432 15.5 0.033 0.035 0.585 1.052
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.973 1.013
acc_transpose_blocks_kernels 27432 16.5 0.266 0.274 0.941 0.960
make_images_sizes 4572 15.5 0.005 0.005 0.696 0.929
mp_alltoall_i44 4572 16.5 0.691 0.924 0.691 0.924
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.792 0.866
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=41.027000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=528.363636, yerr=3.599816
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 592.412672E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10224 344.
MP_Sync 104
MP_Alltoall 1582 3682667.
MP_ISendRecv 10710 94533.
MP_Wait 16690
MP_comm_split 50
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.021 0.038 29.502 29.502
qs_mol_dyn_low 1 2.0 0.003 0.003 29.205 29.216
qs_forces 11 3.9 0.003 0.006 29.077 29.079
qs_energies 11 4.9 0.001 0.001 27.351 27.356
scf_env_do_scf 11 5.9 0.000 0.001 22.323 22.323
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 19.656 19.657
velocity_verlet 10 3.0 0.001 0.001 14.890 14.896
dbcsr_multiply_generic 2286 12.5 0.092 0.095 13.059 13.135
qs_scf_new_mos 108 7.5 0.001 0.001 11.802 11.834
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.801 11.833
ot_scf_mini 108 9.5 0.002 0.002 11.100 11.144
multiply_cannon 2286 13.5 0.230 0.236 10.174 10.724
multiply_cannon_loop 2286 14.5 0.330 0.340 9.173 9.364
ot_mini 108 10.5 0.001 0.001 6.220 6.273
multiply_cannon_multrec 9144 15.5 1.647 1.902 5.969 6.190
rebuild_ks_matrix 119 8.3 0.000 0.000 5.993 6.036
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 5.992 6.035
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.345 5.385
qs_ot_get_derivative 108 11.5 0.001 0.001 4.866 4.909
dbcsr_mm_accdrv_process 12550 15.8 3.217 4.031 4.222 4.294
sum_up_and_integrate 119 10.3 0.037 0.041 3.686 3.693
integrate_v_rspace 119 11.3 0.003 0.003 3.649 3.656
init_scf_run 11 5.9 0.000 0.001 3.594 3.594
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.594 3.594
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.330 3.348
calculate_rho_elec 119 8.7 0.060 0.061 3.330 3.347
qs_ot_get_p 119 10.4 0.001 0.001 3.059 3.108
init_scf_loop 11 6.9 0.000 0.000 2.645 2.646
mp_waitall_1 121218 16.5 2.057 2.618 2.057 2.618
calculate_first_density_matrix 1 7.0 0.000 0.000 2.440 2.443
make_m2s 4572 13.5 0.034 0.035 1.941 2.136
make_images 4572 14.5 0.269 0.302 1.852 2.046
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.980 1.983
grid_integrate_task_list 119 12.3 1.854 1.932 1.854 1.932
jit_kernel_multiply 10 15.7 0.966 1.922 0.966 1.922
prepare_preconditioner 11 7.9 0.000 0.000 1.907 1.912
make_preconditioner 11 8.9 0.000 0.000 1.907 1.912
rs_pw_transfer 974 11.9 0.008 0.008 1.812 1.892
density_rs2pw 119 9.7 0.003 0.004 1.741 1.834
calculate_dm_sparse 119 9.5 0.000 0.000 1.803 1.825
pw_transfer 1439 11.6 0.063 0.065 1.800 1.816
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.788 1.816
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.800 1.802
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.768 1.769
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.703 1.729
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.709 1.723
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.629 1.638
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.461 1.462
potential_pw2rs 119 12.3 0.010 0.010 1.445 1.448
cp_fm_redistribute_end 50 14.0 0.727 1.434 0.728 1.435
cp_fm_diag_elpa_base 50 14.0 0.658 1.351 0.705 1.412
fft3d_ps 1201 14.6 0.539 0.549 1.376 1.390
fft_wrap_pw1pw2_140 487 13.2 0.082 0.086 1.353 1.366
grid_collocate_task_list 119 9.7 1.274 1.362 1.274 1.362
ot_diis_step 108 11.5 0.013 0.013 1.328 1.329
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.254 1.273
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.217 1.221
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.197 1.215
apply_single 119 13.6 0.000 0.000 1.196 1.215
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.199 1.207
hybrid_alltoall_any 4725 16.4 0.062 0.176 0.857 1.152
make_images_data 4572 15.5 0.038 0.042 0.914 1.141
wfi_extrapolate 11 7.9 0.001 0.001 1.106 1.107
acc_transpose_blocks 9144 15.5 0.038 0.040 1.095 1.103
cp_fm_cholesky_invert 11 10.9 0.992 0.996 0.992 0.996
mp_alltoall_d11v 2130 13.8 0.864 0.951 0.864 0.951
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.866 0.921
acc_transpose_blocks_kernels 9144 16.5 0.116 0.120 0.855 0.858
mp_allgather_i34 2286 14.5 0.333 0.803 0.333 0.803
multiply_cannon_sync_h2d 9144 15.5 0.711 0.784 0.711 0.784
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.770 0.773
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.752 0.763
mp_alltoall_z22v 1201 16.6 0.711 0.743 0.711 0.743
jit_kernel_transpose 5 15.6 0.739 0.740 0.739 0.740
multiply_cannon_metrocomm3 9144 15.5 0.018 0.019 0.405 0.727
qs_env_update_s_mstruct 11 6.9 0.000 0.000 0.672 0.724
mp_sum_l 7207 12.9 0.514 0.632 0.514 0.632
dbcsr_complete_redistribute 329 12.2 0.159 0.168 0.567 0.598
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=29.502000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=563.090909, yerr=3.117532
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 762.437632E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63729.
MP_Allreduce 10074 433.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_ISendRecv 4998 189067.
MP_Wait 8898
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.013 0.032 49.944 49.944
qs_mol_dyn_low 1 2.0 0.003 0.003 49.737 49.745
qs_forces 11 3.9 0.001 0.002 49.676 49.677
qs_energies 11 4.9 0.001 0.002 47.727 47.731
scf_env_do_scf 11 5.9 0.001 0.001 41.849 41.849
scf_env_do_scf_inner_loop 108 6.5 0.003 0.006 32.654 32.663
velocity_verlet 10 3.0 0.001 0.001 29.951 29.957
qs_scf_new_mos 108 7.5 0.001 0.001 22.369 22.461
qs_scf_loop_do_ot 108 8.5 0.001 0.001 22.368 22.461
dbcsr_multiply_generic 2286 12.5 0.109 0.111 22.020 22.145
ot_scf_mini 108 9.5 0.002 0.002 21.174 21.278
multiply_cannon 2286 13.5 0.310 0.316 15.634 16.794
multiply_cannon_loop 2286 14.5 0.344 0.352 13.883 14.628
ot_mini 108 10.5 0.001 0.001 13.478 13.595
qs_ot_get_derivative 108 11.5 0.001 0.001 10.848 10.948
init_scf_loop 11 6.9 0.000 0.000 9.161 9.170
multiply_cannon_multrec 9144 15.5 3.459 4.959 8.557 8.645
prepare_preconditioner 11 7.9 0.000 0.000 7.788 7.802
make_preconditioner 11 8.9 0.000 0.000 7.788 7.802
rebuild_ks_matrix 119 8.3 0.000 0.000 7.484 7.615
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.483 7.615
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.650 7.236
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.826 6.942
dbcsr_mm_accdrv_process 12550 15.8 4.121 5.616 4.972 6.337
mp_waitall_1 97218 16.6 5.301 6.176 5.301 6.176
cp_fm_upper_to_full 72 14.2 3.321 4.973 3.321 4.973
qs_ot_get_p 119 10.4 0.001 0.001 4.571 4.679
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.480 4.533
calculate_rho_elec 119 8.7 0.118 0.121 4.479 4.533
make_m2s 4572 13.5 0.037 0.038 3.969 4.449
make_images 4572 14.5 0.359 0.392 3.849 4.329
dbcsr_complete_redistribute 329 12.2 0.287 0.291 3.363 4.328
sum_up_and_integrate 119 10.3 0.064 0.066 4.064 4.071
integrate_v_rspace 119 11.3 0.003 0.003 4.000 4.008
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.456 3.908
init_scf_run 11 5.9 0.000 0.001 3.843 3.843
scf_env_initial_rho_setup 11 6.9 0.004 0.005 3.843 3.843
copy_fm_to_dbcsr 176 11.2 0.001 0.001 2.696 3.649
multiply_cannon_metrocomm3 9144 15.5 0.019 0.020 2.570 3.162
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.832 3.149
apply_single 119 13.6 0.000 0.000 2.832 3.148
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 2.133 3.055
make_images_data 4572 15.5 0.043 0.046 2.145 2.856
mp_alltoall_i22 627 13.8 1.764 2.764 1.764 2.764
ot_diis_step 108 11.5 0.014 0.014 2.583 2.586
hybrid_alltoall_any 4725 16.4 0.087 0.149 1.988 2.550
pw_transfer 1439 11.6 0.066 0.067 2.529 2.537
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.425 2.482
fft_wrap_pw1pw2 1201 12.6 0.008 0.009 2.433 2.441
density_rs2pw 119 9.7 0.003 0.003 2.422 2.437
mp_sum_l 7207 12.9 1.865 2.436 1.865 2.436
calculate_first_density_matrix 1 7.0 0.000 0.000 2.367 2.424
qs_ot_p2m_diag 50 11.0 0.043 0.044 2.418 2.420
calculate_dm_sparse 119 9.5 0.000 0.000 2.313 2.358
mp_alltoall_d11v 2130 13.8 2.221 2.311 2.221 2.311
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.130 2.132
fft3d_ps 1201 14.6 0.568 0.577 2.066 2.074
grid_integrate_task_list 119 12.3 2.043 2.068 2.043 2.068
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.042 2.042
fft_wrap_pw1pw2_140 487 13.2 0.088 0.090 1.919 1.931
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.776 1.824
rs_pw_transfer 974 11.9 0.009 0.009 1.793 1.820
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.773 1.775
cp_fm_diag_elpa_base 50 14.0 1.625 1.679 1.771 1.772
qs_energies_init_hamiltonians 11 5.9 0.004 0.006 1.750 1.751
mp_allgather_i34 2286 14.5 0.727 1.601 0.727 1.601
cp_fm_cholesky_invert 11 10.9 1.525 1.529 1.525 1.529
grid_collocate_task_list 119 9.7 1.478 1.497 1.478 1.497
potential_pw2rs 119 12.3 0.014 0.014 1.420 1.424
make_images_sizes 4572 15.5 0.005 0.005 0.870 1.416
mp_alltoall_i44 4572 16.5 0.866 1.412 0.866 1.412
mp_alltoall_z22v 1201 16.6 1.361 1.380 1.361 1.380
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.347 1.365
wfi_extrapolate 11 7.9 0.001 0.001 1.359 1.359
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.275 1.326
copy_dbcsr_to_fm 153 11.3 0.002 0.002 1.179 1.233
acc_transpose_blocks 9144 15.5 0.038 0.038 1.133 1.149
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.083 1.100
multiply_cannon_sync_h2d 9144 15.5 1.046 1.050 1.046 1.050
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.987 1.031
jit_kernel_multiply 6 15.6 0.824 1.018 0.824 1.018
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=49.944000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=713.545455, yerr=16.686214
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 501.784576E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65372.
MP_Allreduce 9840 486.
MP_Sync 100
MP_Alltoall 1938 306090.
MP_ISendRecv 41800 9096.
MP_Wait 58168
MP_comm_split 48
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.050 98.595 98.596
qs_mol_dyn_low 1 2.0 0.003 0.003 98.244 98.254
qs_forces 11 3.9 0.003 0.003 98.172 98.173
qs_energies 11 4.9 0.001 0.002 95.260 95.276
scf_env_do_scf 11 5.9 0.000 0.001 84.090 84.093
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 78.215 78.215
dbcsr_multiply_generic 2055 12.4 0.106 0.109 58.988 59.668
qs_scf_new_mos 99 7.5 0.000 0.001 54.926 55.054
qs_scf_loop_do_ot 99 8.5 0.000 0.001 54.926 55.053
ot_scf_mini 99 9.5 0.002 0.002 52.211 52.329
velocity_verlet 10 3.0 0.001 0.001 49.208 49.209
multiply_cannon 2055 13.4 0.174 0.181 46.123 47.655
multiply_cannon_loop 2055 14.4 1.534 1.584 44.526 45.924
ot_mini 99 10.5 0.001 0.001 32.102 32.235
qs_ot_get_derivative 99 11.5 0.001 0.001 24.798 24.952
mp_waitall_1 241148 16.1 18.221 20.422 18.221 20.422
multiply_cannon_multrec 49320 15.4 12.043 12.808 17.001 17.871
rebuild_ks_matrix 110 8.3 0.000 0.000 17.543 17.640
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 17.543 17.639
qs_ks_update_qs_env 110 7.6 0.001 0.001 15.721 15.806
qs_ot_get_p 110 10.4 0.001 0.001 10.720 10.859
multiply_cannon_sync_h2d 49320 15.4 9.889 10.640 9.889 10.640
qs_rho_update_rho_low 110 7.6 0.001 0.001 9.700 9.835
calculate_rho_elec 110 8.6 0.020 0.024 9.699 9.835
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 9.270 9.369
sum_up_and_integrate 110 10.3 0.037 0.044 9.125 9.141
integrate_v_rspace 110 11.3 0.003 0.003 9.088 9.112
init_scf_run 11 5.9 0.000 0.001 9.066 9.066
scf_env_initial_rho_setup 11 6.9 0.001 0.001 9.066 9.066
multiply_cannon_metrocomm3 49320 15.4 0.076 0.079 7.351 8.556
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.482 8.042
apply_single 110 13.6 0.000 0.000 7.482 8.042
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.325 7.879
mp_sum_l 6514 12.8 6.127 7.125 6.127 7.125
rs_pw_transfer 902 11.9 0.012 0.013 6.744 6.997
ot_diis_step 99 11.5 0.006 0.006 6.986 6.989
multiply_cannon_metrocomm1 49320 15.4 0.060 0.063 5.287 6.956
density_rs2pw 110 9.6 0.004 0.004 6.604 6.770
make_m2s 4110 13.4 0.060 0.065 6.276 6.410
qs_ot_p2m_diag 48 11.0 0.012 0.018 6.327 6.374
make_images 4110 14.4 0.178 0.191 6.180 6.317
wfi_extrapolate 11 7.9 0.001 0.001 6.058 6.058
init_scf_loop 11 6.9 0.000 0.000 5.827 5.828
pw_transfer 1331 11.6 0.053 0.062 5.649 5.762
fft_wrap_pw1pw2 1111 12.6 0.007 0.008 5.562 5.678
fft3d_ps 1111 14.6 0.752 0.842 5.102 5.207
dbcsr_mm_accdrv_process 87628 16.1 2.013 2.152 4.839 5.175
cp_dbcsr_syevd 48 12.0 0.002 0.003 5.132 5.132
fft_wrap_pw1pw2_140 451 13.1 0.166 0.186 4.519 4.636
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 4.571 4.582
cp_fm_diag_elpa 48 13.0 0.000 0.000 4.577 4.579
cp_fm_redistribute_end 48 14.0 3.980 4.543 3.985 4.545
cp_fm_diag_elpa_base 48 14.0 0.552 4.431 0.556 4.456
potential_pw2rs 110 12.3 0.006 0.007 4.362 4.446
calculate_dm_sparse 110 9.5 0.000 0.001 4.100 4.242
mp_alltoall_z22v 1111 16.6 3.809 3.988 3.809 3.988
qs_ot_get_orbitals 99 10.5 0.000 0.001 3.855 3.915
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.766 3.853
prepare_preconditioner 11 7.9 0.000 0.000 3.556 3.577
make_preconditioner 11 8.9 0.000 0.000 3.556 3.577
mp_alltoall_d11v 2046 13.8 3.064 3.483 3.064 3.483
grid_integrate_task_list 110 12.3 3.215 3.397 3.215 3.397
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.322 3.381
make_images_data 4110 15.4 0.043 0.046 2.770 3.178
calculate_first_density_matrix 1 7.0 0.000 0.000 2.912 2.920
hybrid_alltoall_any 4261 16.3 0.081 0.473 2.336 2.745
make_images_sizes 4110 15.4 0.004 0.005 1.923 2.624
mp_alltoall_i44 4110 16.4 1.919 2.620 1.919 2.620
jit_kernel_multiply 13 15.9 2.544 2.571 2.544 2.571
mp_waitany 14300 13.8 2.351 2.544 2.351 2.544
acc_transpose_blocks 49320 15.4 0.203 0.219 2.386 2.498
parallel_gemm_fm 81 9.0 0.000 0.000 2.424 2.432
parallel_gemm_fm_cosma 81 10.0 2.424 2.431 2.424 2.431
mp_sum_d 3883 11.9 1.809 2.379 1.809 2.379
grid_collocate_task_list 110 9.6 2.088 2.222 2.088 2.222
yz_to_x 341 15.4 0.061 0.071 2.070 2.222
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.967 1.985
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=98.596000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=475.818182, yerr=2.854603
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/10/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 390.715586E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 5019072 0.0% 0.0% 100.0%
average stack size 0.0 0.0 196.1
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 589.000704E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65587.
MP_Allreduce 9839 562.
MP_Sync 100
MP_Alltoall 1717 2800018.
MP_ISendRecv 20680 26400.
MP_Wait 32692
MP_comm_split 48
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.026 70.104 70.105
qs_mol_dyn_low 1 2.0 0.003 0.003 69.898 69.907
qs_forces 11 3.9 0.002 0.002 69.821 69.822
qs_energies 11 4.9 0.001 0.002 66.494 66.500
scf_env_do_scf 11 5.9 0.001 0.001 57.639 57.643
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 49.512 49.513
dbcsr_multiply_generic 2055 12.4 0.117 0.124 38.174 38.393
velocity_verlet 10 3.0 0.001 0.001 37.084 37.085
qs_scf_new_mos 99 7.5 0.001 0.001 33.410 33.550
qs_scf_loop_do_ot 99 8.5 0.001 0.001 33.409 33.549
multiply_cannon 2055 13.4 0.221 0.244 31.189 32.582
ot_scf_mini 99 9.5 0.003 0.003 31.713 31.814
multiply_cannon_loop 2055 14.4 0.927 0.950 29.791 30.670
ot_mini 99 10.5 0.001 0.001 18.501 18.604
multiply_cannon_multrec 24660 15.4 7.604 9.218 13.771 15.375
rebuild_ks_matrix 110 8.3 0.000 0.000 13.861 13.930
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.015 13.861 13.930
qs_ot_get_derivative 99 11.5 0.001 0.001 12.657 12.761
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.226 12.289
mp_waitall_1 186928 16.3 8.114 11.029 8.114 11.029
multiply_cannon_sync_h2d 24660 15.4 6.971 8.128 6.971 8.128
init_scf_loop 11 6.9 0.000 0.000 8.076 8.077
multiply_cannon_metrocomm3 24660 15.4 0.070 0.071 5.191 7.956
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.509 7.186
apply_single 110 13.6 0.000 0.001 6.509 7.186
sum_up_and_integrate 110 10.3 0.052 0.060 6.737 6.750
integrate_v_rspace 110 11.3 0.002 0.003 6.685 6.701
qs_ot_get_p 110 10.4 0.001 0.001 6.485 6.601
init_scf_run 11 5.9 0.000 0.001 6.460 6.461
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.460 6.461
dbcsr_mm_accdrv_process 52282 16.1 4.689 5.453 6.008 6.309
prepare_preconditioner 11 7.9 0.000 0.000 5.957 5.979
make_preconditioner 11 8.9 0.000 0.000 5.957 5.979
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.941 5.950
calculate_rho_elec 110 8.6 0.039 0.047 5.940 5.950
ot_diis_step 99 11.5 0.010 0.010 5.786 5.786
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.528 5.694
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.717 5.442
make_m2s 4110 13.4 0.056 0.059 4.439 4.939
make_images 4110 14.4 0.402 0.443 4.331 4.827
qs_ot_p2m_diag 48 11.0 0.028 0.044 4.485 4.508
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.048 4.048
wfi_extrapolate 11 7.9 0.001 0.001 3.712 3.712
pw_transfer 1331 11.6 0.065 0.070 3.478 3.617
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.372 3.510
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 3.482 3.484
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.469 3.471
cp_fm_redistribute_end 48 14.0 2.583 3.435 2.584 3.436
density_rs2pw 110 9.6 0.004 0.004 3.212 3.429
cp_fm_diag_elpa_base 48 14.0 0.815 3.312 0.847 3.395
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.336 3.392
grid_integrate_task_list 110 12.3 3.151 3.355 3.151 3.355
rs_pw_transfer 902 11.9 0.012 0.014 3.126 3.342
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.097 3.122
calculate_dm_sparse 110 9.5 0.001 0.001 2.994 3.055
make_images_data 4110 15.4 0.047 0.051 2.440 2.950
hybrid_alltoall_any 4261 16.3 0.102 0.442 2.161 2.922
fft_wrap_pw1pw2_140 451 13.1 0.199 0.221 2.776 2.914
cp_fm_cholesky_invert 11 10.9 2.887 2.895 2.887 2.895
fft3d_ps 1111 14.6 1.075 1.272 2.731 2.860
calculate_first_density_matrix 1 7.0 0.000 0.000 2.657 2.659
potential_pw2rs 110 12.3 0.008 0.009 2.560 2.580
mp_sum_l 6514 12.8 1.915 2.501 1.915 2.501
grid_collocate_task_list 110 9.6 2.121 2.259 2.121 2.259
mp_alltoall_d11v 2046 13.8 1.951 2.156 1.951 2.156
jit_kernel_multiply 11 16.2 0.965 2.066 0.965 2.066
qs_ot_get_orbitals 99 10.5 0.001 0.001 2.000 2.024
qs_energies_init_hamiltonians 11 5.9 0.000 0.001 1.849 1.849
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.801 1.813
mp_allgather_i34 2055 14.4 0.727 1.793 0.727 1.793
acc_transpose_blocks 24660 15.4 0.105 0.110 1.691 1.734
multiply_cannon_metrocomm4 22605 15.4 0.075 0.079 0.799 1.650
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.541 1.555
mp_irecv_dv 57340 16.2 0.675 1.533 0.675 1.533
dbcsr_complete_redistribute 325 12.2 0.238 0.298 1.225 1.489
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.368 1.473
mp_alltoall_z22v 1111 16.6 1.390 1.470 1.390 1.470
mp_waitany 10164 13.8 1.217 1.419 1.217 1.419
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=70.105000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=555.909091, yerr=7.292020
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 658.792448E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65578.
MP_Allreduce 9838 559.
MP_Sync 100
MP_Alltoall 1496 4511006.
MP_ISendRecv 13640 27424.
MP_Wait 32318
MP_comm_split 48
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.057 0.151 70.985 70.987
qs_mol_dyn_low 1 2.0 0.004 0.007 70.521 70.530
qs_forces 11 3.9 0.003 0.004 70.233 70.235
qs_energies 11 4.9 0.002 0.003 67.041 67.046
scf_env_do_scf 11 5.9 0.001 0.001 57.144 57.144
scf_env_do_scf_inner_loop 99 6.5 0.004 0.010 48.112 48.116
velocity_verlet 10 3.0 0.003 0.005 37.614 37.622
dbcsr_multiply_generic 2055 12.4 0.108 0.112 33.133 33.470
qs_scf_new_mos 99 7.5 0.001 0.001 29.810 29.904
qs_scf_loop_do_ot 99 8.5 0.001 0.001 29.809 29.903
ot_scf_mini 99 9.5 0.003 0.003 28.381 28.507
multiply_cannon 2055 13.4 0.220 0.282 24.478 25.554
multiply_cannon_loop 2055 14.4 0.619 0.631 22.937 24.274
ot_mini 99 10.5 0.012 0.028 16.498 16.638
mp_waitall_1 146766 16.3 11.286 14.723 11.286 14.723
rebuild_ks_matrix 110 8.3 0.000 0.001 14.238 14.374
qs_ks_build_kohn_sham_matrix 110 9.3 0.016 0.035 14.238 14.373
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.854 12.994
qs_ot_get_derivative 99 11.5 0.001 0.002 11.783 11.898
multiply_cannon_multrec 16440 15.4 3.953 4.875 10.004 10.962
init_scf_loop 11 6.9 0.026 0.109 8.991 8.998
multiply_cannon_metrocomm3 16440 15.4 0.042 0.043 4.845 8.660
sum_up_and_integrate 110 10.3 0.060 0.066 7.849 7.865
integrate_v_rspace 110 11.3 0.003 0.004 7.789 7.805
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.459 7.504
calculate_rho_elec 110 8.6 0.058 0.058 7.458 7.504
prepare_preconditioner 11 7.9 0.000 0.000 6.893 6.908
make_preconditioner 11 8.9 0.000 0.001 6.893 6.908
init_scf_run 11 5.9 0.000 0.001 6.820 6.821
scf_env_initial_rho_setup 11 6.9 0.026 0.056 6.820 6.821
qs_ot_get_p 110 10.4 0.002 0.004 6.649 6.779
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.241 6.599
make_m2s 4110 13.4 0.049 0.051 5.964 6.401
make_images 4110 14.4 0.396 0.515 5.849 6.286
dbcsr_mm_accdrv_process 34862 16.1 4.821 5.742 5.904 6.142
apply_preconditioner_dbcsr 110 12.6 0.000 0.001 5.111 5.575
apply_single 110 13.6 0.000 0.000 5.111 5.575
ot_diis_step 99 11.5 0.037 0.084 4.661 4.673
pw_transfer 1331 11.6 0.065 0.071 4.597 4.621
density_rs2pw 110 9.6 0.004 0.004 4.372 4.562
rs_pw_transfer 902 11.9 0.011 0.012 4.312 4.518
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 4.491 4.516
multiply_cannon_sync_h2d 16440 15.4 3.679 4.221 3.679 4.221
make_images_data 4110 15.4 0.043 0.048 3.432 4.148
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 4.018 4.101
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.427 4.067
wfi_extrapolate 11 7.9 0.003 0.016 3.872 3.873
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.833 3.837
fft3d_ps 1111 14.6 1.066 1.093 3.810 3.832
fft_wrap_pw1pw2_140 451 13.1 0.207 0.209 3.749 3.777
hybrid_alltoall_any 4261 16.3 0.104 0.370 2.884 3.515
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 3.482 3.485
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.463 3.464
grid_integrate_task_list 110 12.3 3.164 3.336 3.164 3.336
potential_pw2rs 110 12.3 0.010 0.010 3.223 3.264
mp_alltoall_d11v 2046 13.8 2.920 3.200 2.920 3.200
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 2.816 2.938
calculate_dm_sparse 110 9.5 0.001 0.001 2.825 2.866
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.819 2.823
calculate_first_density_matrix 1 7.0 0.000 0.001 2.809 2.810
mp_sum_l 6514 12.8 2.123 2.787 2.123 2.787
cp_fm_redistribute_end 48 14.0 1.739 2.767 1.742 2.770
cp_fm_diag_elpa_base 48 14.0 0.959 2.610 1.018 2.718
cp_fm_cholesky_invert 11 10.9 2.698 2.705 2.698 2.705
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 2.613 2.618
mp_alltoall_z22v 1111 16.6 2.531 2.562 2.531 2.562
grid_collocate_task_list 110 9.6 2.176 2.330 2.176 2.330
multiply_cannon_metrocomm1 16440 15.4 0.029 0.029 1.470 2.304
multiply_cannon_metrocomm4 14385 15.4 0.046 0.049 0.865 2.297
mp_allgather_i34 2055 14.4 0.854 2.262 0.854 2.262
dbcsr_complete_redistribute 325 12.2 0.318 0.358 1.772 2.243
mp_irecv_dv 48980 15.7 0.797 2.177 0.797 2.177
make_images_sizes 4110 15.4 0.005 0.005 1.181 1.934
mp_alltoall_i44 4110 16.4 1.176 1.929 1.176 1.929
cp_fm_upper_to_full 70 13.6 1.394 1.843 1.394 1.843
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.709 1.786
cp_fm_cholesky_decompose 22 10.9 1.729 1.761 1.729 1.761
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.645 1.663
rs_gather_matrices 110 12.3 0.137 0.150 1.349 1.612
yz_to_x 561 15.1 0.099 0.101 1.573 1.599
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.582 1.593
copy_fm_to_dbcsr 174 11.2 0.001 0.001 1.116 1.585
mp_waitany 17072 13.8 1.345 1.546 1.345 1.546
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.354 1.486
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=70.987000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=622.636364, yerr=9.819024
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 725.315584E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65576.
MP_Allreduce 9838 600.
MP_Sync 100
MP_Alltoall 1496 5863162.
MP_ISendRecv 10120 43184.
MP_Wait 25102
MP_comm_split 48
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.072 0.135 66.976 66.977
qs_mol_dyn_low 1 2.0 0.003 0.005 66.530 66.539
qs_forces 11 3.9 0.002 0.004 66.406 66.407
qs_energies 11 4.9 0.002 0.004 62.986 62.989
scf_env_do_scf 11 5.9 0.001 0.002 54.267 54.270
scf_env_do_scf_inner_loop 99 6.5 0.002 0.008 41.987 41.991
velocity_verlet 10 3.0 0.001 0.002 37.665 37.670
dbcsr_multiply_generic 2055 12.4 0.116 0.119 30.260 30.505
qs_scf_new_mos 99 7.5 0.001 0.001 27.306 27.409
qs_scf_loop_do_ot 99 8.5 0.001 0.006 27.305 27.408
ot_scf_mini 99 9.5 0.003 0.007 25.651 25.754
multiply_cannon 2055 13.4 0.244 0.302 22.687 23.640
multiply_cannon_loop 2055 14.4 0.881 0.905 21.256 21.943
ot_mini 99 10.5 0.001 0.001 14.544 14.661
multiply_cannon_multrec 24660 15.4 4.164 6.807 12.709 13.898
rebuild_ks_matrix 110 8.3 0.000 0.000 12.168 12.268
qs_ks_build_kohn_sham_matrix 110 9.3 0.017 0.049 12.167 12.268
init_scf_loop 11 6.9 0.003 0.016 12.232 12.243
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.738 10.829
qs_ot_get_derivative 99 11.5 0.001 0.001 10.290 10.397
prepare_preconditioner 11 7.9 0.000 0.000 10.233 10.247
make_preconditioner 11 8.9 0.001 0.004 10.233 10.247
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.466 9.925
dbcsr_mm_accdrv_process 52304 16.0 6.959 8.395 8.398 9.309
mp_waitall_1 126806 16.4 5.013 7.066 5.013 7.066
sum_up_and_integrate 110 10.3 0.070 0.088 6.471 6.489
integrate_v_rspace 110 11.3 0.003 0.003 6.401 6.420
qs_ot_get_p 110 10.4 0.001 0.002 6.187 6.321
make_m2s 4110 13.4 0.058 0.060 5.765 6.049
init_scf_run 11 5.9 0.000 0.001 5.940 5.940
scf_env_initial_rho_setup 11 6.9 0.001 0.004 5.939 5.940
make_images 4110 14.4 0.581 0.704 5.627 5.910
qs_rho_update_rho_low 110 7.6 0.001 0.006 5.887 5.896
calculate_rho_elec 110 8.6 0.077 0.081 5.886 5.895
cp_fm_upper_to_full 70 13.8 3.320 4.710 3.320 4.710
ot_diis_step 99 11.5 0.012 0.013 4.208 4.209
apply_preconditioner_dbcsr 110 12.6 0.000 0.001 4.080 4.171
apply_single 110 13.6 0.000 0.000 4.080 4.171
qs_ot_p2m_diag 48 11.0 0.055 0.064 4.048 4.061
dbcsr_complete_redistribute 325 12.2 0.410 0.447 2.768 3.899
multiply_cannon_metrocomm3 24660 15.4 0.035 0.036 1.609 3.571
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.549 3.550
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.431 3.486
pw_transfer 1331 11.6 0.064 0.072 3.434 3.469
grid_integrate_task_list 110 12.3 3.275 3.444 3.275 3.444
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.329 3.367
multiply_cannon_sync_h2d 24660 15.4 3.197 3.366 3.197 3.366
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.168 3.283
make_images_data 4110 15.4 0.046 0.050 2.913 3.260
density_rs2pw 110 9.6 0.004 0.004 3.013 3.221
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.164 3.212
hybrid_alltoall_any 4261 16.3 0.119 0.460 2.506 3.165
wfi_extrapolate 11 7.9 0.001 0.005 3.154 3.154
calculate_dm_sparse 110 9.5 0.001 0.001 3.124 3.151
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.983 2.986
cp_fm_redistribute_end 48 14.0 1.490 2.956 1.492 2.957
cp_fm_diag_elpa_base 48 14.0 1.381 2.814 1.461 2.926
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.910 2.912
cp_fm_cholesky_invert 11 10.9 2.892 2.901 2.892 2.901
rs_pw_transfer 902 11.9 0.010 0.011 2.654 2.875
mp_alltoall_i22 605 13.7 1.688 2.868 1.688 2.868
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.757 2.855
fft_wrap_pw1pw2_140 451 13.1 0.201 0.213 2.781 2.819
fft3d_ps 1111 14.6 1.062 1.096 2.634 2.656
calculate_first_density_matrix 1 7.0 0.000 0.001 2.602 2.605
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 2.466 2.507
grid_collocate_task_list 110 9.6 2.221 2.375 2.221 2.375
mp_alltoall_d11v 2046 13.8 1.947 2.354 1.947 2.354
jit_kernel_multiply 12 15.7 1.107 2.352 1.107 2.352
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 2.258 2.259
potential_pw2rs 110 12.3 0.013 0.013 2.227 2.242
cp_fm_cholesky_decompose 22 10.9 1.759 1.800 1.759 1.800
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.762 1.793
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.716 1.733
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.686 1.699
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 1.597 1.696
mp_sum_l 6514 12.8 1.126 1.658 1.126 1.658
multiply_cannon_metrocomm4 20550 15.4 0.057 0.060 0.853 1.546
acc_transpose_blocks 24660 15.4 0.102 0.105 1.497 1.516
mp_irecv_dv 62702 16.1 0.756 1.470 0.756 1.470
mp_allgather_i34 2055 14.4 0.532 1.447 0.532 1.447
mp_alltoall_z22v 1111 16.6 1.357 1.426 1.357 1.426
mp_waitany 13376 13.8 1.125 1.411 1.125 1.411
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=66.977000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=686.181818, yerr=9.053560
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 834.994176E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3521 65574.
MP_Allreduce 9838 640.
MP_Sync 100
MP_Alltoall 1496 8504061.
MP_ISendRecv 6600 54848.
MP_Wait 17226
MP_comm_split 48
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.013 0.029 59.184 59.184
qs_mol_dyn_low 1 2.0 0.003 0.003 58.936 58.945
qs_forces 11 3.9 0.002 0.002 58.858 58.860
qs_energies 11 4.9 0.001 0.001 55.095 55.099
scf_env_do_scf 11 5.9 0.000 0.001 46.320 46.320
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 37.183 37.183
velocity_verlet 10 3.0 0.001 0.001 32.126 32.130
dbcsr_multiply_generic 2055 12.4 0.105 0.107 23.735 23.881
qs_scf_new_mos 99 7.5 0.001 0.001 22.473 22.526
qs_scf_loop_do_ot 99 8.5 0.001 0.001 22.472 22.526
ot_scf_mini 99 9.5 0.002 0.002 21.189 21.219
multiply_cannon 2055 13.4 0.248 0.264 17.770 19.207
multiply_cannon_loop 2055 14.4 0.322 0.335 16.344 16.638
rebuild_ks_matrix 110 8.3 0.000 0.000 11.977 12.000
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 11.977 12.000
ot_mini 99 10.5 0.001 0.001 10.914 10.934
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.626 10.649
init_scf_loop 11 6.9 0.000 0.000 9.084 9.086
mp_waitall_1 106626 16.5 6.936 8.729 6.936 8.729
multiply_cannon_multrec 8220 15.4 3.221 4.465 7.567 8.499
prepare_preconditioner 11 7.9 0.000 0.000 7.255 7.258
make_preconditioner 11 8.9 0.000 0.000 7.255 7.258
qs_ot_get_derivative 99 11.5 0.001 0.001 7.052 7.078
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.810 6.892
sum_up_and_integrate 110 10.3 0.079 0.081 6.587 6.599
integrate_v_rspace 110 11.3 0.003 0.003 6.508 6.519
qs_ot_get_p 110 10.4 0.001 0.001 6.284 6.314
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.989 6.003
calculate_rho_elec 110 8.6 0.115 0.115 5.988 6.003
init_scf_run 11 5.9 0.000 0.001 5.615 5.615
scf_env_initial_rho_setup 11 6.9 0.001 0.002 5.614 5.615
dbcsr_mm_accdrv_process 17442 15.9 3.027 3.737 4.216 5.095
make_m2s 4110 13.4 0.038 0.040 4.504 4.783
make_images 4110 14.4 0.637 0.698 4.374 4.654
qs_ot_p2m_diag 48 11.0 0.081 0.084 4.612 4.617
multiply_cannon_metrocomm3 8220 15.4 0.017 0.017 3.064 4.464
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.268 4.269
ot_diis_step 99 11.5 0.012 0.012 3.836 3.837
cp_fm_cholesky_invert 11 10.9 3.773 3.778 3.773 3.778
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.727 3.745
apply_single 110 13.6 0.000 0.000 3.726 3.745
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.609 3.613
cp_fm_redistribute_end 48 14.0 0.916 3.545 0.923 3.549
pw_transfer 1331 11.6 0.064 0.069 3.539 3.546
cp_fm_diag_elpa_base 48 14.0 2.434 3.328 2.611 3.490
grid_integrate_task_list 110 12.3 3.369 3.463 3.369 3.463
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.432 3.440
density_rs2pw 110 9.6 0.004 0.004 3.007 3.164
wfi_extrapolate 11 7.9 0.001 0.001 3.101 3.101
multiply_cannon_sync_h2d 8220 15.4 2.901 3.027 2.901 3.027
make_images_data 4110 15.4 0.038 0.043 2.533 2.988
fft_wrap_pw1pw2_140 451 13.1 0.211 0.214 2.847 2.858
hybrid_alltoall_any 4261 16.3 0.199 0.861 2.434 2.848
rs_pw_transfer 902 11.9 0.010 0.010 2.584 2.758
calculate_dm_sparse 110 9.5 0.001 0.001 2.695 2.730
fft3d_ps 1111 14.6 1.116 1.139 2.691 2.706
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.675 2.676
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.639 2.641
grid_collocate_task_list 110 9.6 2.331 2.427 2.331 2.427
calculate_first_density_matrix 1 7.0 0.000 0.000 2.394 2.396
potential_pw2rs 110 12.3 0.015 0.016 2.224 2.228
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.164 2.182
mp_alltoall_d11v 2046 13.8 1.992 2.123 1.992 2.123
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.019 2.042
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.024 2.035
cp_fm_cholesky_decompose 22 10.9 1.980 1.988 1.980 1.988
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.770 1.977
dbcsr_complete_redistribute 325 12.2 0.553 0.574 1.730 1.850
mp_allgather_i34 2055 14.4 0.609 1.759 0.609 1.759
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.686 1.688
qs_env_update_s_mstruct 11 6.9 0.000 0.000 1.522 1.640
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.624 1.637
multiply_cannon_metrocomm1 8220 15.4 0.021 0.022 1.018 1.444
mp_alltoall_z22v 1111 16.6 1.342 1.359 1.342 1.359
qs_create_task_list 11 7.9 0.000 0.000 1.244 1.345
generate_qs_task_list 11 8.9 0.378 0.446 1.244 1.345
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.322 1.340
mp_waitany 9240 13.8 1.109 1.297 1.109 1.297
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=59.184000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=787.727273, yerr=8.057644
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 1.374196E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67104.
MP_Allreduce 9672 819.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_ISendRecv 4620 360267.
MP_Wait 7524
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.016 0.036 87.450 87.450
qs_mol_dyn_low 1 2.0 0.003 0.003 86.836 87.134
qs_forces 11 3.9 0.002 0.003 86.706 86.708
qs_energies 11 4.9 0.001 0.001 82.504 82.506
scf_env_do_scf 11 5.9 0.000 0.001 72.581 72.581
velocity_verlet 10 3.0 0.001 0.001 55.830 55.838
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 43.540 43.541
init_scf_loop 11 6.9 0.000 0.000 28.964 28.965
dbcsr_multiply_generic 2055 12.4 0.120 0.121 28.686 28.741
prepare_preconditioner 11 7.9 0.000 0.000 26.997 27.003
make_preconditioner 11 8.9 0.000 0.000 26.997 27.003
qs_scf_new_mos 99 7.5 0.001 0.001 26.854 26.896
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.853 26.895
make_full_inverse_cholesky 11 9.9 0.000 0.000 21.102 26.481
ot_scf_mini 99 9.5 0.002 0.002 25.095 25.127
multiply_cannon 2055 13.4 0.339 0.360 21.487 22.360
multiply_cannon_loop 2055 14.4 0.341 0.343 19.709 20.025
cp_fm_upper_to_full 70 14.2 12.874 18.509 12.874 18.509
ot_mini 99 10.5 0.001 0.001 13.817 13.841
rebuild_ks_matrix 110 8.3 0.001 0.001 13.419 13.430
qs_ks_build_kohn_sham_matrix 110 9.3 0.014 0.014 13.419 13.430
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.108 12.117
dbcsr_complete_redistribute 325 12.2 1.021 1.039 7.523 10.819
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.473 9.769
multiply_cannon_multrec 8220 15.4 4.456 4.637 9.654 9.731
qs_ot_get_derivative 99 11.5 0.001 0.001 9.331 9.362
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.880 9.149
mp_alltoall_i22 605 13.7 5.519 8.808 5.519 8.808
mp_waitall_1 87304 16.6 7.955 8.699 7.955 8.699
sum_up_and_integrate 110 10.3 0.150 0.152 6.861 6.872
integrate_v_rspace 110 11.3 0.004 0.004 6.710 6.722
qs_rho_update_rho_low 110 7.6 0.001 0.001 6.665 6.703
calculate_rho_elec 110 8.6 0.227 0.227 6.665 6.703
qs_ot_get_p 110 10.4 0.001 0.001 6.089 6.113
make_m2s 4110 13.4 0.042 0.043 5.415 5.879
init_scf_run 11 5.9 0.000 0.001 5.693 5.694
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.693 5.693
make_images 4110 14.4 0.880 0.922 5.228 5.693
cp_fm_cholesky_invert 11 10.9 5.611 5.617 5.611 5.617
dbcsr_mm_accdrv_process 11614 15.7 3.417 3.651 5.054 5.309
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.608 5.087
apply_single 110 13.6 0.000 0.000 4.608 5.087
multiply_cannon_metrocomm3 8220 15.4 0.018 0.018 4.698 5.020
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.461 4.469
ot_diis_step 99 11.5 0.015 0.016 4.465 4.466
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.015 4.016
pw_transfer 1331 11.6 0.074 0.074 3.987 3.995
multiply_cannon_sync_h2d 8220 15.4 3.951 3.961 3.951 3.961
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 3.871 3.878
grid_integrate_task_list 110 12.3 3.666 3.710 3.666 3.710
hybrid_alltoall_any 4261 16.3 0.257 0.554 2.951 3.683
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 3.623 3.624
make_images_data 4110 15.4 0.041 0.045 2.958 3.600
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.392 3.393
cp_fm_diag_elpa_base 48 14.0 2.841 3.045 3.390 3.391
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.922 3.363
fft_wrap_pw1pw2_140 451 13.1 0.216 0.220 3.347 3.359
wfi_extrapolate 11 7.9 0.001 0.001 3.291 3.292
calculate_dm_sparse 110 9.5 0.001 0.001 3.155 3.181
density_rs2pw 110 9.6 0.004 0.004 3.151 3.175
fft3d_ps 1111 14.6 1.268 1.276 3.079 3.089
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.874 2.878
grid_collocate_task_list 110 9.6 2.635 2.662 2.635 2.662
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.466 2.481
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.355 2.363
rs_pw_transfer 902 11.9 0.010 0.011 2.311 2.363
calculate_first_density_matrix 1 7.0 0.000 0.000 2.294 2.295
potential_pw2rs 110 12.3 0.022 0.022 2.241 2.253
qs_env_update_s_mstruct 11 6.9 0.000 0.000 2.168 2.226
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.104 2.217
mp_alltoall_d11v 2046 13.8 2.147 2.199 2.147 2.199
cp_fm_cholesky_decompose 22 10.9 2.097 2.122 2.097 2.122
qs_create_task_list 11 7.9 0.000 0.000 1.898 1.945
generate_qs_task_list 11 8.9 0.738 0.792 1.898 1.945
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.919 1.921
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.807 1.834
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=87.450000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1222.909091, yerr=70.134228
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410022121472 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1924145348608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957871443968 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1963544850432 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2714615709696 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4377645416448 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5350455508992 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5395653328896 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6594687401984 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11444702699520 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019188129792 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019188129792 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19624853225472 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 92.796573E+12 0.0% 0.0% 100.0%
flops max/rank 1.086553E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705499744 0.0% 0.0% 100.0%
number of processed stacks 11851392 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.8
marketing flops 143.508480E+12
-------------------------------------------------------------------------------
# multiplications 2485
max memory usage/rank 630.411264E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10258080
MPI messages size (bytes):
total size 4.456715E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.459031E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3542056 94711185408
32768 < size <= 131072 1282176 73356279808
131072 < size <= 4194304 5107038 3151762421624
4194304 < size <= 16777216 259842 1136842803272
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4085 56756.
MP_Allreduce 11253 786.
MP_Sync 168
MP_Alltoall 2210 2557061.
MP_ISendRecv 48260 18752.
MP_Wait 66280
MP_comm_split 82
MP_ISend 15900 108037.
MP_IRecv 15900 108037.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.029 239.188 239.190
qs_mol_dyn_low 1 2.0 0.004 0.007 238.792 238.806
qs_forces 11 3.9 0.003 0.005 238.657 238.660
qs_energies 11 4.9 0.001 0.002 232.306 232.375
scf_env_do_scf 11 5.9 0.001 0.001 214.524 214.528
scf_env_do_scf_inner_loop 116 6.6 0.002 0.008 191.667 191.671
velocity_verlet 10 3.0 0.001 0.002 143.239 143.243
dbcsr_multiply_generic 2485 12.5 0.174 0.178 140.870 141.784
qs_scf_new_mos 116 7.6 0.001 0.001 139.991 140.572
qs_scf_loop_do_ot 116 8.6 0.001 0.001 139.990 140.572
ot_scf_mini 116 9.6 0.003 0.004 130.789 131.081
multiply_cannon 2485 13.5 0.236 0.245 111.814 114.250
multiply_cannon_loop 2485 14.5 2.092 2.156 108.610 111.072
ot_mini 116 10.6 0.001 0.002 72.374 72.737
mp_waitall_1 288940 16.2 47.235 51.518 47.235 51.518
qs_ot_get_derivative 116 11.6 0.001 0.005 47.834 48.225
multiply_cannon_multrec 59640 15.5 32.546 35.172 40.926 42.717
rebuild_ks_matrix 127 8.3 0.001 0.001 37.269 37.578
qs_ks_build_kohn_sham_matrix 127 9.3 0.016 0.017 37.269 37.577
qs_ot_get_p 127 10.4 0.001 0.001 34.054 34.334
qs_ks_update_qs_env 127 7.6 0.001 0.001 33.133 33.421
multiply_cannon_sync_h2d 59640 15.5 26.867 29.675 26.867 29.675
apply_preconditioner_dbcsr 127 12.6 0.000 0.011 24.181 24.933
apply_single 127 13.6 0.001 0.001 24.181 24.933
ot_diis_step 116 11.6 0.008 0.010 24.216 24.217
qs_rho_update_rho_low 127 7.7 0.001 0.001 23.467 24.172
calculate_rho_elec 127 8.7 0.044 0.063 23.466 24.172
qs_ot_get_derivative_diag 76 12.4 0.002 0.002 23.100 23.458
qs_ot_p2m_diag 82 11.4 0.076 0.090 23.226 23.363
init_scf_loop 11 6.9 0.000 0.000 22.639 22.640
cp_dbcsr_syevd 82 12.4 0.004 0.005 20.157 20.164
multiply_cannon_metrocomm3 59640 15.5 0.109 0.113 16.865 18.956
multiply_cannon_metrocomm1 59640 15.5 0.085 0.089 15.925 18.876
make_m2s 4970 13.5 0.101 0.111 17.909 18.360
make_images 4970 14.5 0.400 0.417 17.731 18.195
prepare_preconditioner 11 7.9 0.000 0.000 16.480 16.526
make_preconditioner 11 8.9 0.000 0.000 16.480 16.526
make_full_inverse_cholesky 11 9.9 0.000 0.000 15.707 15.910
density_rs2pw 127 9.7 0.006 0.007 14.804 15.300
sum_up_and_integrate 127 10.3 0.089 0.108 15.271 15.286
cp_fm_diag_elpa 82 13.4 0.000 0.001 15.254 15.274
integrate_v_rspace 127 11.3 0.003 0.004 15.181 15.199
cp_fm_redistribute_end 82 14.4 11.953 15.011 11.984 15.025
cp_fm_diag_elpa_base 82 14.4 2.931 14.165 2.974 14.342
pw_transfer 1535 11.6 0.074 0.105 11.886 12.523
rs_pw_transfer 1038 11.9 0.017 0.019 11.443 12.328
fft_wrap_pw1pw2 1281 12.7 0.010 0.013 11.683 12.299
init_scf_run 11 5.9 0.000 0.001 12.282 12.283
scf_env_initial_rho_setup 11 6.9 0.001 0.002 12.282 12.283
mp_sum_l 7804 13.0 9.901 11.836 9.901 11.836
calculate_dm_sparse 127 9.5 0.001 0.001 10.997 11.583
fft3d_ps 1281 14.7 2.068 2.544 10.491 11.053
qs_ot_get_orbitals 116 10.6 0.001 0.001 10.411 10.538
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 10.190 10.370
make_images_data 4970 15.5 0.069 0.076 8.734 9.879
cp_fm_cholesky_invert 11 10.9 9.485 9.493 9.485 9.493
wfi_extrapolate 11 7.9 0.001 0.002 9.072 9.072
fft_wrap_pw1pw2_140 519 13.2 0.436 0.502 8.359 8.924
hybrid_alltoall_any 5155 16.4 0.286 2.224 7.464 8.794
mp_alltoall_d11v 2401 14.1 7.345 8.789 7.345 8.789
dbcsr_mm_accdrv_process 123452 16.2 3.302 3.441 7.954 8.608
mp_alltoall_z22v 1281 16.7 7.304 7.849 7.304 7.849
grid_integrate_task_list 127 12.3 7.060 7.463 7.060 7.463
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.003 6.664 6.671
make_images_sizes 4970 15.5 0.006 0.007 3.652 6.006
mp_alltoall_i44 4970 16.5 3.646 6.000 3.646 6.000
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.799 5.918
potential_pw2rs 127 12.3 0.009 0.010 5.736 5.791
grid_collocate_task_list 127 9.7 4.697 5.013 4.697 5.013
rs_pw_transfer_RS2PW_20 127 11.7 0.205 0.218 4.454 4.944
cp_fm_cholesky_decompose 22 10.9 4.796 4.809 4.796 4.809
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=239.190000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=595.636364, yerr=8.347811
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430457200640 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986252263424 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613077360640 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239167967232 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239167967232 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 94.233007E+12 0.0% 0.0% 100.0%
flops max/rank 2.200017E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383584 0.0% 0.0% 100.0%
number of processed stacks 6024768 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1129.7
marketing flops 145.651870E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 827.346944E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2427840
MPI messages size (bytes):
total size 4.132587E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.702166E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 71532 2339635200
32768 < size <= 131072 729952 56049532928
131072 < size <= 4194304 1387568 1410045313024
4194304 < size <= 16777216 155760 1473828432560
16777216 < size 68112 1190343475200
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4127 56635.
MP_Allreduce 11357 942.
MP_Sync 172
MP_Alltoall 1983 4792572.
MP_ISendRecv 24252 47072.
MP_Wait 38240
MP_comm_split 84
MP_ISend 11836 212447.
MP_IRecv 11836 212447.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.022 0.039 193.970 193.972
qs_mol_dyn_low 1 2.0 0.003 0.006 193.553 193.567
qs_forces 11 3.9 0.003 0.006 193.448 193.450
qs_energies 11 4.9 0.002 0.004 186.573 186.584
scf_env_do_scf 11 5.9 0.001 0.001 169.525 169.535
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 135.492 135.497
velocity_verlet 10 3.0 0.001 0.002 122.409 122.412
dbcsr_multiply_generic 2529 12.6 0.186 0.191 99.145 100.320
qs_scf_new_mos 118 7.6 0.001 0.001 96.253 96.570
qs_scf_loop_do_ot 118 8.6 0.001 0.003 96.252 96.569
ot_scf_mini 118 9.6 0.004 0.006 91.373 91.754
multiply_cannon 2529 13.6 0.485 0.534 78.405 82.377
multiply_cannon_loop 2529 14.6 1.266 1.304 74.807 77.535
ot_mini 118 10.6 0.001 0.001 50.920 51.300
mp_waitall_1 228724 16.4 25.689 39.199 25.689 39.199
multiply_cannon_multrec 30348 15.6 22.081 27.223 31.896 37.191
init_scf_loop 11 6.9 0.000 0.001 33.943 33.954
rebuild_ks_matrix 129 8.3 0.001 0.001 33.359 33.864
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.023 33.358 33.863
qs_ks_update_qs_env 129 7.6 0.001 0.001 29.953 30.414
prepare_preconditioner 11 7.9 0.000 0.000 29.477 29.534
make_preconditioner 11 8.9 0.000 0.001 29.477 29.534
qs_ot_get_derivative 118 11.6 0.001 0.002 28.677 29.056
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.046 28.604
multiply_cannon_metrocomm3 30348 15.6 0.096 0.102 15.985 28.350
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 22.312 23.423
apply_single 129 13.6 0.001 0.001 22.312 23.422
qs_ot_get_p 129 10.4 0.001 0.002 21.967 22.510
ot_diis_step 118 11.6 0.015 0.017 22.063 22.065
multiply_cannon_sync_h2d 30348 15.6 19.342 21.117 19.342 21.117
cp_fm_cholesky_invert 11 10.9 17.234 17.247 17.234 17.247
qs_ot_p2m_diag 84 11.4 0.190 0.218 17.005 17.046
make_m2s 5058 13.6 0.088 0.094 14.653 16.439
make_images 5058 14.6 1.179 1.374 14.447 16.232
cp_dbcsr_syevd 84 12.4 0.005 0.006 15.805 15.807
sum_up_and_integrate 129 10.3 0.117 0.131 14.733 14.795
integrate_v_rspace 129 11.3 0.003 0.004 14.615 14.680
qs_rho_update_rho_low 129 7.7 0.001 0.001 13.216 13.260
calculate_rho_elec 129 8.7 0.088 0.106 13.215 13.259
cp_fm_diag_elpa 84 13.4 0.000 0.001 12.666 12.682
cp_fm_redistribute_end 84 14.4 7.402 12.600 7.415 12.609
cp_fm_diag_elpa_base 84 14.4 4.941 12.061 5.164 12.420
init_scf_run 11 5.9 0.000 0.001 12.142 12.143
scf_env_initial_rho_setup 11 6.9 0.001 0.002 12.142 12.143
qs_ot_get_derivative_diag 78 12.4 0.002 0.002 11.273 11.569
make_images_data 5058 15.6 0.067 0.074 8.841 10.873
multiply_cannon_metrocomm4 27819 15.6 0.099 0.113 3.784 10.489
mp_irecv_dv 70084 16.3 3.587 10.103 3.587 10.103
hybrid_alltoall_any 5245 16.5 0.343 1.504 7.548 10.085
dbcsr_mm_accdrv_process 62758 16.2 4.685 5.612 9.274 9.840
wfi_extrapolate 11 7.9 0.002 0.008 8.429 8.430
pw_transfer 1559 11.6 0.085 0.102 8.014 8.127
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 7.789 7.895
grid_integrate_task_list 129 12.3 7.231 7.645 7.231 7.645
density_rs2pw 129 9.7 0.006 0.006 6.971 7.470
cp_fm_cholesky_decompose 22 10.9 7.257 7.342 7.257 7.342
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.265 7.034
fft_wrap_pw1pw2_140 527 13.2 0.472 0.514 6.771 6.863
calculate_dm_sparse 129 9.5 0.001 0.001 6.566 6.732
rs_pw_transfer 1054 12.0 0.015 0.017 5.942 6.493
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.344 6.354
fft3d_ps 1301 14.7 2.794 2.944 6.153 6.245
mp_sum_l 7936 13.1 4.298 5.858 4.298 5.858
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.001 5.474 5.609
qs_ot_get_orbitals 118 10.6 0.001 0.001 5.457 5.537
grid_collocate_task_list 129 9.7 4.954 5.353 4.954 5.353
mp_allgather_i34 2529 14.6 2.156 5.218 2.156 5.218
potential_pw2rs 129 12.3 0.015 0.016 5.100 5.159
mp_alltoall_d11v 2429 14.1 4.178 4.726 4.178 4.726
dbcsr_complete_redistribute 397 12.7 0.773 0.860 3.204 4.106
mp_sum_d 4503 12.2 2.696 3.956 2.696 3.956
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=193.972000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=788.909091, yerr=1.621141
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 931.663872E+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 4103 56951.
MP_Allreduce 11296 984.
MP_Sync 170
MP_Alltoall 1712 9388896.
MP_ISendRecv 15872 75008.
MP_Wait 29756
MP_comm_split 83
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.032 0.057 193.971 193.972
qs_mol_dyn_low 1 2.0 0.003 0.006 193.475 193.490
qs_forces 11 3.9 0.006 0.028 193.357 193.364
qs_energies 11 4.9 0.002 0.005 186.722 186.733
scf_env_do_scf 11 5.9 0.001 0.001 169.640 169.640
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 132.409 132.411
velocity_verlet 10 3.0 0.001 0.002 124.042 124.047
qs_scf_new_mos 117 7.6 0.001 0.001 93.794 94.138
qs_scf_loop_do_ot 117 8.6 0.001 0.001 93.794 94.137
ot_scf_mini 117 9.6 0.004 0.004 89.576 89.944
dbcsr_multiply_generic 2507 12.6 0.182 0.185 86.529 87.702
multiply_cannon 2507 13.6 0.497 0.522 64.011 69.360
multiply_cannon_loop 2507 14.6 0.861 0.886 60.422 63.049
ot_mini 117 10.6 0.001 0.001 46.229 46.599
mp_waitall_1 178456 16.5 28.709 37.871 28.709 37.871
init_scf_loop 11 6.9 0.000 0.001 37.100 37.105
prepare_preconditioner 11 7.9 0.000 0.000 32.926 32.977
make_preconditioner 11 8.9 0.000 0.001 32.926 32.977
rebuild_ks_matrix 128 8.3 0.001 0.001 31.616 32.125
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.022 31.616 32.124
make_full_inverse_cholesky 11 9.9 0.000 0.000 29.760 31.408
qs_ks_update_qs_env 128 7.6 0.001 0.001 28.582 29.044
qs_ot_get_p 128 10.4 0.001 0.002 27.698 28.181
qs_ot_get_derivative 117 11.6 0.003 0.018 25.368 25.744
multiply_cannon_metrocomm3 20056 15.6 0.058 0.061 16.133 25.368
multiply_cannon_multrec 20056 15.6 13.540 16.436 22.264 25.091
qs_ot_p2m_diag 83 11.4 0.266 0.272 22.513 22.534
apply_preconditioner_dbcsr 128 12.6 0.001 0.009 20.472 21.742
apply_single 128 13.6 0.001 0.001 20.471 21.741
cp_dbcsr_syevd 83 12.4 0.005 0.005 21.436 21.441
ot_diis_step 117 11.6 0.020 0.026 20.699 20.701
cp_fm_diag_elpa 83 13.4 0.000 0.001 17.881 17.884
cp_fm_redistribute_end 83 14.4 6.736 17.724 6.753 17.730
make_m2s 5014 13.6 0.079 0.085 16.456 17.584
cp_fm_diag_elpa_base 83 14.4 10.482 16.977 10.936 17.555
make_images 5014 14.6 1.191 1.274 16.226 17.353
multiply_cannon_sync_h2d 20056 15.6 14.239 16.273 14.239 16.273
sum_up_and_integrate 128 10.3 0.134 0.147 15.425 15.450
cp_fm_cholesky_invert 11 10.9 15.338 15.352 15.338 15.352
integrate_v_rspace 128 11.3 0.003 0.004 15.290 15.320
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.717 13.755
calculate_rho_elec 128 8.7 0.132 0.146 13.716 13.754
make_images_data 5014 15.6 0.060 0.068 10.332 12.034
init_scf_run 11 5.9 0.000 0.001 11.420 11.421
scf_env_initial_rho_setup 11 6.9 0.001 0.003 11.420 11.420
hybrid_alltoall_any 5200 16.5 0.430 1.951 8.920 11.334
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.431 9.705
pw_transfer 1547 11.6 0.085 0.104 8.902 9.010
multiply_cannon_metrocomm4 17549 15.6 0.064 0.073 3.414 8.965
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.679 8.792
mp_irecv_dv 50230 16.2 3.292 8.720 3.292 8.720
wfi_extrapolate 11 7.9 0.005 0.034 8.452 8.452
dbcsr_mm_accdrv_process 41502 16.2 4.432 5.236 8.181 8.386
dbcsr_complete_redistribute 395 12.7 1.168 1.198 6.001 7.883
cp_fm_upper_to_full 105 14.5 5.934 7.829 5.934 7.829
fft_wrap_pw1pw2_140 523 13.2 0.478 0.527 7.591 7.712
grid_integrate_task_list 128 12.3 7.275 7.693 7.275 7.693
cp_fm_cholesky_decompose 22 10.9 7.516 7.553 7.516 7.553
density_rs2pw 128 9.7 0.006 0.006 7.023 7.429
fft3d_ps 1291 14.7 2.696 2.903 6.947 7.016
mp_alltoall_d11v 2415 14.1 6.023 6.560 6.023 6.560
copy_fm_to_dbcsr 209 11.7 0.002 0.002 4.427 6.303
rs_pw_transfer 1046 11.9 0.014 0.015 5.763 6.242
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 5.533 6.203
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.178 6.185
calculate_dm_sparse 128 9.5 0.001 0.001 5.835 5.929
mp_sum_l 7870 13.0 4.237 5.853 4.237 5.853
mp_allgather_i34 2507 14.6 2.110 5.608 2.110 5.608
grid_collocate_task_list 128 9.7 5.062 5.446 5.062 5.446
potential_pw2rs 128 12.3 0.020 0.023 5.267 5.284
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.936 5.055
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 3.146 4.969
mp_alltoall_i22 716 14.1 2.354 4.377 2.354 4.377
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.297 4.299
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.238 4.264
mp_sum_d 4471 12.2 2.916 3.911 2.916 3.911
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=193.972000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=889.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 1.164591E+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 4093 57084.
MP_Allreduce 11271 1067.
MP_Sync 170
MP_Alltoall 1712 12503107.
MP_ISendRecv 11776 75008.
MP_Wait 28330
MP_comm_split 83
MP_ISend 14952 244818.
MP_IRecv 14952 244818.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.040 0.088 204.004 204.004
qs_mol_dyn_low 1 2.0 0.005 0.017 203.345 203.361
qs_forces 11 3.9 0.013 0.018 203.132 203.147
qs_energies 11 4.9 0.003 0.016 196.105 196.119
scf_env_do_scf 11 5.9 0.001 0.001 177.545 177.557
velocity_verlet 10 3.0 0.001 0.002 134.945 134.951
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 125.185 125.191
qs_scf_new_mos 117 7.6 0.001 0.001 87.266 87.555
qs_scf_loop_do_ot 117 8.6 0.001 0.001 87.265 87.554
ot_scf_mini 117 9.6 0.003 0.004 82.664 82.965
dbcsr_multiply_generic 2507 12.6 0.189 0.194 81.816 82.651
multiply_cannon 2507 13.6 0.558 0.603 55.425 58.361
multiply_cannon_loop 2507 14.6 1.180 1.207 51.193 53.218
init_scf_loop 11 6.9 0.001 0.003 52.235 52.242
prepare_preconditioner 11 7.9 0.000 0.000 47.277 47.309
make_preconditioner 11 8.9 0.000 0.000 47.277 47.309
make_full_inverse_cholesky 11 9.9 0.000 0.000 40.295 45.547
ot_mini 117 10.6 0.001 0.002 43.526 43.826
rebuild_ks_matrix 128 8.3 0.001 0.001 31.073 31.392
qs_ks_build_kohn_sham_matrix 128 9.3 0.018 0.025 31.073 31.391
multiply_cannon_multrec 30084 15.6 14.084 19.822 26.004 31.162
mp_waitall_1 153770 16.5 19.392 29.357 19.392 29.357
qs_ks_update_qs_env 128 7.6 0.001 0.001 28.300 28.592
qs_ot_get_p 128 10.4 0.001 0.002 24.956 25.269
qs_ot_get_derivative 117 11.6 0.001 0.002 23.716 24.037
make_m2s 5014 13.6 0.094 0.098 21.680 22.931
make_images 5014 14.6 1.964 2.263 21.379 22.632
qs_ot_p2m_diag 83 11.4 0.343 0.389 20.226 20.277
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 19.191 19.763
apply_single 128 13.6 0.001 0.001 19.191 19.763
ot_diis_step 117 11.6 0.018 0.020 19.677 19.680
cp_dbcsr_syevd 83 12.4 0.005 0.005 18.671 18.674
cp_fm_cholesky_invert 11 10.9 17.422 17.430 17.422 17.430
cp_fm_upper_to_full 105 14.7 11.409 17.008 11.409 17.008
sum_up_and_integrate 128 10.3 0.142 0.157 15.907 15.935
integrate_v_rspace 128 11.3 0.004 0.004 15.765 15.799
multiply_cannon_metrocomm3 30084 15.6 0.046 0.048 6.452 15.139
dbcsr_complete_redistribute 395 12.7 1.505 1.637 10.907 14.859
cp_fm_diag_elpa 83 13.4 0.000 0.000 14.645 14.660
cp_fm_redistribute_end 83 14.4 2.515 14.456 2.549 14.461
cp_fm_diag_elpa_base 83 14.4 11.224 13.723 11.848 14.243
qs_rho_update_rho_low 128 7.7 0.001 0.005 14.180 14.218
calculate_rho_elec 128 8.7 0.175 0.190 14.178 14.217
make_images_data 5014 15.6 0.063 0.067 11.882 14.209
hybrid_alltoall_any 5200 16.5 0.524 2.195 10.669 12.897
copy_fm_to_dbcsr 209 11.7 0.001 0.002 8.898 12.859
multiply_cannon_sync_h2d 30084 15.6 11.714 12.547 11.714 12.547
init_scf_run 11 5.9 0.000 0.001 12.169 12.171
scf_env_initial_rho_setup 11 6.9 0.001 0.004 12.169 12.171
dbcsr_mm_accdrv_process 62264 16.2 7.276 8.422 11.496 12.029
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 6.960 10.785
cp_fm_cholesky_decompose 22 10.9 10.344 10.440 10.344 10.440
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.734 9.966
pw_transfer 1547 11.6 0.084 0.099 9.867 9.921
mp_alltoall_i22 716 14.1 5.941 9.782 5.941 9.782
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 9.643 9.706
fft_wrap_pw1pw2_140 523 13.2 0.481 0.491 8.572 8.661
wfi_extrapolate 11 7.9 0.001 0.004 8.452 8.452
fft3d_ps 1291 14.7 2.798 2.861 7.871 7.938
grid_integrate_task_list 128 12.3 7.510 7.846 7.510 7.846
density_rs2pw 128 9.7 0.005 0.006 7.066 7.322
mp_alltoall_d11v 2415 14.1 6.743 7.062 6.743 7.062
multiply_cannon_metrocomm4 25070 15.6 0.076 0.085 2.800 6.958
mp_irecv_dv 76098 16.2 2.657 6.694 2.657 6.694
calculate_dm_sparse 128 9.5 0.001 0.001 6.442 6.525
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.993 6.036
potential_pw2rs 128 12.3 0.022 0.023 5.874 5.896
rs_pw_transfer 1046 11.9 0.013 0.014 5.448 5.670
grid_collocate_task_list 128 9.7 5.210 5.651 5.210 5.651
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.848 4.854
mp_alltoall_z22v 1291 16.7 4.489 4.563 4.489 4.563
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.386 4.472
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.369 4.453
qs_ot_get_orbitals 117 10.6 0.001 0.001 4.327 4.383
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=204.004000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1079.727273, yerr=25.096508
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430458527744 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 11613083000832 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 94.228655E+12 0.0% 0.0% 100.0%
flops max/rank 5.909858E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806314816 0.0% 0.0% 100.0%
number of processed stacks 1976192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3444.2
marketing flops 145.646636E+12
-------------------------------------------------------------------------------
# multiplications 2527
max memory usage/rank 1.512694E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 242592
MPI messages size (bytes):
total size 1.341705E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.530704E+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 114816 60196651008
4194304 < size <= 16777216 105840 554906419200
16777216 < size 20352 726591807936
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 9003 51.
MP_Alltoall 9708 795382.
MP_ISend 40388 2101966.
MP_IRecv 40388 2101070.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4030 57813.
MP_Allreduce 11072 1172.
MP_Sync 87
MP_Alltoall 1724 18848449.
MP_ISendRecv 7740 122880.
MP_Wait 20114
MP_ISend 10760 423501.
MP_IRecv 10760 423501.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.032 193.556 193.556
qs_mol_dyn_low 1 2.0 0.003 0.003 193.182 193.196
qs_forces 11 3.9 0.003 0.003 192.999 193.004
qs_energies 11 4.9 0.002 0.002 183.895 183.904
scf_env_do_scf 11 5.9 0.001 0.001 163.926 163.943
velocity_verlet 10 3.0 0.001 0.001 125.993 126.001
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 124.242 124.244
qs_scf_new_mos 118 7.6 0.001 0.001 85.019 85.140
qs_scf_loop_do_ot 118 8.6 0.001 0.001 85.018 85.139
dbcsr_multiply_generic 2527 12.6 0.196 0.210 82.309 82.681
ot_scf_mini 118 9.6 0.004 0.004 80.291 80.392
multiply_cannon 2527 13.6 0.592 0.639 59.112 64.859
multiply_cannon_loop 2527 14.6 0.451 0.460 53.912 54.881
ot_mini 118 10.6 0.001 0.001 43.500 43.600
init_scf_loop 11 6.9 0.000 0.000 39.530 39.533
mp_waitall_1 130650 16.6 31.748 39.494 31.748 39.494
prepare_preconditioner 11 7.9 0.000 0.000 35.365 35.573
make_preconditioner 11 8.9 0.000 0.000 35.365 35.573
make_full_inverse_cholesky 11 9.9 0.000 0.000 33.243 33.561
rebuild_ks_matrix 129 8.3 0.001 0.001 31.889 32.059
qs_ks_build_kohn_sham_matrix 129 9.3 0.018 0.018 31.888 32.058
qs_ks_update_qs_env 129 7.6 0.001 0.001 28.096 28.247
qs_ot_get_derivative 118 11.6 0.002 0.002 22.538 22.637
qs_ot_get_p 129 10.4 0.001 0.001 22.187 22.379
multiply_cannon_metrocomm3 10108 15.6 0.022 0.023 13.588 21.580
ot_diis_step 118 11.6 0.020 0.021 20.874 20.876
multiply_cannon_multrec 10108 15.6 10.457 14.215 17.841 20.736
make_m2s 5054 13.6 0.066 0.069 18.407 20.677
make_images 5054 14.6 2.348 2.881 18.102 20.375
apply_preconditioner_dbcsr 129 12.6 0.000 0.000 19.997 20.298
apply_single 129 13.6 0.001 0.001 19.997 20.298
cp_fm_cholesky_invert 11 10.9 19.348 19.355 19.348 19.355
qs_ot_p2m_diag 83 11.4 0.496 0.501 16.669 16.684
sum_up_and_integrate 129 10.3 0.181 0.192 16.109 16.161
integrate_v_rspace 129 11.3 0.004 0.004 15.928 15.989
cp_dbcsr_syevd 83 12.4 0.005 0.005 15.535 15.537
qs_rho_update_rho_low 129 7.7 0.001 0.001 15.080 15.098
calculate_rho_elec 129 8.7 0.260 0.271 15.079 15.098
hybrid_alltoall_any 5240 16.5 0.842 3.836 11.159 13.901
make_images_data 5054 15.6 0.053 0.061 11.174 13.829
init_scf_run 11 5.9 0.000 0.001 12.936 12.938
scf_env_initial_rho_setup 11 6.9 0.014 0.020 12.936 12.938
cp_fm_diag_elpa 83 13.4 0.000 0.000 12.407 12.411
cp_fm_diag_elpa_base 83 14.4 12.149 12.241 12.401 12.405
multiply_cannon_sync_h2d 10108 15.6 11.694 12.172 11.694 12.172
multiply_cannon_metrocomm1 10108 15.6 0.029 0.030 6.784 11.011
pw_transfer 1559 11.6 0.084 0.091 9.709 9.730
fft_wrap_pw1pw2 1301 12.7 0.010 0.010 9.488 9.516
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.220 9.306
cp_fm_cholesky_decompose 22 10.9 8.852 8.962 8.852 8.962
wfi_extrapolate 11 7.9 0.001 0.001 8.940 8.941
mp_alltoall_d11v 2423 14.1 7.990 8.430 7.990 8.430
mp_allgather_i34 2527 14.6 3.379 8.121 3.379 8.121
grid_integrate_task_list 129 12.3 7.831 8.069 7.831 8.069
fft_wrap_pw1pw2_140 527 13.2 0.504 0.529 8.023 8.049
fft3d_ps 1301 14.7 2.750 2.839 7.633 7.661
dbcsr_mm_accdrv_process 20926 16.1 2.618 3.467 7.014 7.653
density_rs2pw 129 9.7 0.005 0.006 7.390 7.635
dbcsr_complete_redistribute 395 12.7 2.206 2.337 7.015 7.406
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 7.194 7.202
calculate_dm_sparse 129 9.5 0.001 0.001 6.421 6.528
grid_collocate_task_list 129 9.7 5.535 5.764 5.535 5.764
copy_dbcsr_to_fm 186 11.8 0.004 0.004 5.523 5.609
rs_pw_transfer 1054 12.0 0.013 0.013 5.326 5.591
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 5.458 5.461
potential_pw2rs 129 12.3 0.027 0.028 5.245 5.272
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.185 5.223
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 4.719 4.740
qs_ot_get_derivative_taylor 41 13.0 0.001 0.001 4.432 4.471
mp_alltoall_z22v 1301 16.7 4.262 4.393 4.262 4.393
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.329 4.353
copy_fm_to_dbcsr 209 11.7 0.001 0.002 3.931 4.245
multiply_cannon_metrocomm4 7581 15.6 0.024 0.026 1.896 4.061
transfer_dbcsr_to_fm 11 10.9 0.000 0.000 3.838 4.039
mp_irecv_dv 29086 15.9 1.858 3.988 1.858 3.988
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=193.556000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1415.181818, yerr=50.266233
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1410025771008 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1930587799552 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1957877121024 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 11444718956544 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15019199483904 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15019199483904 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 92.803064E+12 0.0% 0.0% 100.0%
flops max/rank 11.607223E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6705604128 0.0% 0.0% 100.0%
number of processed stacks 1948912 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3440.7
marketing flops 143.514185E+12
-------------------------------------------------------------------------------
# multiplications 2488
max memory usage/rank 2.932457E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 99520
MPI messages size (bytes):
total size 1.127523E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.329611E+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 44888 34846277632
4194304 < size <= 16777216 43984 376564613120
16777216 < size 10032 716109036912
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4034 58688.
MP_Allreduce 11082 1513.
MP_Sync 87
MP_Alltoall 1700 36954241.
MP_ISendRecv 3556 218624.
MP_Wait 11506
MP_ISend 6360 1080477.
MP_IRecv 6360 1080477.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.038 295.306 295.307
qs_mol_dyn_low 1 2.0 0.003 0.003 294.785 294.798
qs_forces 11 3.9 0.003 0.003 294.655 294.660
qs_energies 11 4.9 0.002 0.002 285.969 285.981
scf_env_do_scf 11 5.9 0.001 0.001 263.900 263.911
velocity_verlet 10 3.0 0.001 0.002 213.330 213.342
scf_env_do_scf_inner_loop 116 6.6 0.003 0.008 136.288 136.289
init_scf_loop 11 6.9 0.000 0.000 127.340 127.343
prepare_preconditioner 11 7.9 0.000 0.000 122.592 122.614
make_preconditioner 11 8.9 0.000 0.000 122.592 122.614
make_full_inverse_cholesky 11 9.9 0.000 0.000 98.150 119.731
qs_scf_new_mos 116 7.6 0.001 0.001 94.057 94.141
qs_scf_loop_do_ot 116 8.6 0.001 0.001 94.056 94.140
ot_scf_mini 116 9.6 0.004 0.004 89.256 89.292
dbcsr_multiply_generic 2488 12.5 0.212 0.221 84.492 84.931
cp_fm_upper_to_full 105 14.8 53.107 76.578 53.107 76.578
multiply_cannon 2488 13.5 0.697 0.749 59.224 60.990
multiply_cannon_loop 2488 14.5 0.467 0.480 55.091 56.245
ot_mini 116 10.6 0.001 0.001 45.567 45.621
dbcsr_complete_redistribute 395 12.7 3.985 4.027 30.251 43.418
copy_fm_to_dbcsr 209 11.7 0.001 0.001 26.780 39.924
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 24.398 37.370
cp_fm_cholesky_invert 11 10.9 35.448 35.454 35.448 35.454
mp_alltoall_i22 714 14.1 22.217 35.424 22.217 35.424
rebuild_ks_matrix 127 8.3 0.001 0.001 33.820 33.907
qs_ks_build_kohn_sham_matrix 127 9.3 0.018 0.018 33.819 33.907
mp_waitall_1 104666 16.7 29.221 33.138 29.221 33.138
qs_ks_update_qs_env 127 7.6 0.001 0.001 31.480 31.555
qs_ot_get_p 127 10.4 0.001 0.001 28.059 28.168
qs_ot_get_derivative 116 11.6 0.001 0.002 24.666 24.697
qs_ot_p2m_diag 83 11.4 0.878 0.883 23.898 23.929
cp_dbcsr_syevd 83 12.4 0.005 0.006 22.152 22.153
make_m2s 4976 13.5 0.075 0.079 20.391 21.702
make_images 4976 14.5 3.782 3.906 19.920 21.234
ot_diis_step 116 11.6 0.022 0.022 20.850 20.850
apply_preconditioner_dbcsr 127 12.6 0.000 0.000 20.141 20.274
apply_single 127 13.6 0.001 0.001 20.141 20.274
multiply_cannon_metrocomm3 9952 15.5 0.023 0.023 19.133 20.156
cp_fm_diag_elpa 83 13.4 0.000 0.000 18.882 18.886
cp_fm_diag_elpa_base 83 14.4 14.484 16.115 18.878 18.882
multiply_cannon_multrec 9952 15.5 10.449 12.268 17.876 17.987
sum_up_and_integrate 127 10.3 0.318 0.319 15.846 15.935
integrate_v_rspace 127 11.3 0.004 0.004 15.528 15.616
multiply_cannon_sync_h2d 9952 15.5 15.540 15.556 15.540 15.556
qs_rho_update_rho_low 127 7.7 0.001 0.001 14.977 15.007
calculate_rho_elec 127 8.7 0.479 0.480 14.977 15.007
hybrid_alltoall_any 5162 16.4 1.289 3.021 11.076 13.303
make_images_data 4976 15.5 0.060 0.066 11.016 13.199
init_scf_run 11 5.9 0.000 0.001 12.001 12.001
scf_env_initial_rho_setup 11 6.9 0.007 0.008 12.001 12.001
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.704 9.738
cp_fm_cholesky_decompose 22 10.9 9.399 9.416 9.399 9.416
dbcsr_mm_accdrv_process 20602 16.0 3.802 5.799 7.192 9.056
wfi_extrapolate 11 7.9 0.001 0.001 8.914 8.914
grid_integrate_task_list 127 12.3 8.478 8.656 8.478 8.656
pw_transfer 1535 11.6 0.089 0.090 8.459 8.468
fft_wrap_pw1pw2 1281 12.7 0.011 0.011 8.227 8.235
qs_energies_init_hamiltonians 11 5.9 0.026 0.031 7.985 7.985
fft_wrap_pw1pw2_140 519 13.2 0.535 0.540 7.244 7.257
mp_alltoall_d11v 2407 14.1 7.081 7.226 7.081 7.226
calculate_dm_sparse 127 9.5 0.001 0.001 6.641 6.735
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.420 6.469
grid_collocate_task_list 127 9.7 6.291 6.324 6.291 6.324
fft3d_ps 1281 14.7 2.740 2.771 6.286 6.302
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.174 6.250
density_rs2pw 127 9.7 0.005 0.005 6.080 6.108
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=295.307000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2633.363636, yerr=163.263464
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 1.259876E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255669.
MP_Allreduce 3059 6274.
MP_Sync 4
MP_Alltoall 54
MP_ISendRecv 570 19200.
MP_Wait 1302
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.039 0.138 86.702 86.719
qs_energies 1 2.0 0.000 0.000 85.803 85.819
ls_scf 1 3.0 0.003 0.023 84.868 84.877
dbcsr_multiply_generic 111 6.7 0.014 0.015 73.592 73.723
multiply_cannon 111 7.7 0.017 0.020 56.479 57.708
multiply_cannon_loop 111 8.7 0.207 0.222 53.001 54.264
ls_scf_main 1 4.0 0.000 0.002 52.795 52.795
density_matrix_trs4 2 5.0 0.002 0.003 47.233 47.354
ls_scf_init_scf 1 4.0 0.001 0.003 28.976 28.978
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.629 27.674
mp_waitall_1 11316 10.9 23.013 26.299 23.013 26.299
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.002 25.505 25.531
multiply_cannon_multrec 2664 9.7 8.166 8.955 15.526 17.300
multiply_cannon_sync_h2d 2664 9.7 13.482 15.606 13.482 15.606
make_m2s 222 7.7 0.008 0.011 13.248 13.805
make_images 222 8.7 0.099 0.109 13.226 13.784
multiply_cannon_metrocomm1 2664 9.7 0.009 0.011 9.982 13.129
make_images_data 222 9.7 0.004 0.005 7.778 8.383
hybrid_alltoall_any 227 10.6 0.215 1.835 6.665 8.158
dbcsr_mm_accdrv_process 4760 10.4 0.510 0.619 6.980 7.975
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.517 7.822
dbcsr_mm_accdrv_process_sort 4760 11.4 6.271 7.170 6.271 7.170
calculate_norms 4752 9.8 5.527 6.142 5.527 6.142
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.132 5.271
mp_sum_l 807 5.4 3.411 4.891 3.411 4.891
multiply_cannon_metrocomm4 2442 9.7 0.012 0.015 2.064 3.885
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.492 3.870
mp_irecv_dv 6231 10.9 2.046 3.853 2.046 3.853
make_images_sizes 222 9.7 0.000 0.000 0.810 3.581
mp_alltoall_i44 222 10.7 0.810 3.581 0.810 3.581
arnoldi_extremal 4 6.8 0.000 0.001 3.298 3.324
arnoldi_normal_ev 4 7.8 0.001 0.007 3.298 3.324
build_subspace 16 8.4 0.009 0.012 3.189 3.192
ls_scf_post 1 4.0 0.001 0.008 3.095 3.103
ls_scf_store_result 1 5.0 0.000 0.000 2.893 2.958
dbcsr_special_finalize 555 9.7 0.005 0.006 2.312 2.747
dbcsr_merge_single_wm 555 10.7 0.467 0.602 2.304 2.739
make_images_pack 222 9.7 2.202 2.623 2.204 2.625
dbcsr_matrix_vector_mult 304 9.0 0.003 0.010 2.352 2.594
dbcsr_sort_data 658 11.4 2.096 2.493 2.096 2.493
dbcsr_matrix_vector_mult_local 304 10.0 2.073 2.466 2.075 2.468
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.244 2.326
buffer_matrices_ensure_size 222 8.7 1.786 2.131 1.786 2.131
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.847 1.851
rebuild_ks_matrix 3 7.3 0.000 0.000 1.837 1.841
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.009 1.837 1.841
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=86.719000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1140.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 2.092605E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 10339.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_ISendRecv 282 57600.
MP_Wait 828
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.046 0.108 103.218 103.221
qs_energies 1 2.0 0.000 0.001 102.640 102.644
ls_scf 1 3.0 0.000 0.003 101.310 101.314
dbcsr_multiply_generic 111 6.7 0.015 0.019 87.061 87.503
multiply_cannon 111 7.7 0.029 0.043 59.072 65.002
ls_scf_main 1 4.0 0.000 0.005 62.159 62.166
multiply_cannon_loop 111 8.7 0.116 0.125 55.576 59.667
density_matrix_trs4 2 5.0 0.002 0.003 56.483 56.671
mp_waitall_1 9246 10.9 31.823 40.847 31.823 40.847
ls_scf_init_scf 1 4.0 0.000 0.002 35.583 35.584
ls_scf_init_matrix_S 1 5.0 0.000 0.000 34.443 34.532
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.003 32.054 32.076
multiply_cannon_multrec 1332 9.7 13.051 16.906 22.307 27.350
multiply_cannon_metrocomm3 1332 9.7 0.006 0.007 14.007 26.797
make_m2s 222 7.7 0.006 0.008 20.681 21.609
make_images 222 8.7 1.588 1.933 20.651 21.580
make_images_data 222 9.7 0.004 0.004 14.002 15.268
hybrid_alltoall_any 227 10.6 0.522 2.431 13.054 14.707
dbcsr_mm_accdrv_process 4041 10.4 0.273 0.448 8.857 10.329
dbcsr_mm_accdrv_process_sort 4041 11.4 8.452 9.905 8.452 9.905
mp_sum_l 807 5.4 6.025 9.780 6.025 9.780
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.700 8.115
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 4.419 7.695
multiply_cannon_metrocomm4 1221 9.7 0.007 0.008 3.205 7.660
mp_irecv_dv 3311 11.0 3.184 7.602 3.184 7.602
calculate_norms 2376 9.8 5.976 6.734 5.976 6.734
multiply_cannon_sync_h2d 1332 9.7 4.754 5.867 4.754 5.867
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.064 5.279
arnoldi_extremal 4 6.8 0.000 0.001 4.755 4.785
arnoldi_normal_ev 4 7.8 0.002 0.010 4.754 4.785
build_subspace 16 8.4 0.014 0.021 4.461 4.467
mp_allgather_i34 111 8.7 1.196 3.663 1.196 3.663
ls_scf_post 1 4.0 0.000 0.001 3.568 3.573
dbcsr_matrix_vector_mult 304 9.0 0.005 0.018 3.161 3.388
ls_scf_store_result 1 5.0 0.000 0.000 3.262 3.374
dbcsr_matrix_vector_mult_local 304 10.0 2.758 3.240 2.760 3.241
make_images_sizes 222 9.7 0.000 0.000 0.730 3.232
mp_alltoall_i44 222 10.7 0.730 3.232 0.730 3.232
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.534 2.632
make_images_pack 222 9.7 2.024 2.385 2.026 2.388
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=103.221000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1695.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 2.747339E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265470.
MP_Allreduce 3058 11181.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_ISendRecv 186 57600.
MP_Wait 732
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.038 0.057 94.273 94.275
qs_energies 1 2.0 0.000 0.000 93.685 93.692
ls_scf 1 3.0 0.000 0.002 92.307 92.314
dbcsr_multiply_generic 111 6.7 0.017 0.028 77.063 77.321
ls_scf_main 1 4.0 0.000 0.001 57.405 57.410
multiply_cannon 111 7.7 0.039 0.075 53.261 57.137
multiply_cannon_loop 111 8.7 0.100 0.106 49.647 53.663
density_matrix_trs4 2 5.0 0.002 0.003 51.425 51.599
mp_waitall_1 7374 11.0 23.821 33.775 23.821 33.775
ls_scf_init_scf 1 4.0 0.003 0.005 31.375 31.378
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.256 30.317
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.007 27.852 27.868
multiply_cannon_multrec 888 9.7 12.595 15.220 21.162 24.410
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.016 22.811
make_m2s 222 7.7 0.006 0.008 17.033 18.306
make_images 222 8.7 1.978 2.313 16.995 18.266
hybrid_alltoall_any 227 10.6 0.620 2.844 9.197 10.753
make_images_data 222 9.7 0.003 0.005 9.650 10.652
dbcsr_mm_accdrv_process 3754 10.4 0.234 0.421 8.103 9.363
dbcsr_mm_accdrv_process_sort 3754 11.4 7.733 8.942 7.733 8.942
mp_sum_l 807 5.4 5.217 8.764 5.217 8.764
multiply_cannon_sync_h2d 888 9.7 6.014 7.163 6.014 7.163
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.052 6.865
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.771 6.809
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.454 6.777
mp_irecv_dv 2335 11.1 2.438 6.711 2.438 6.711
arnoldi_extremal 4 6.8 0.000 0.003 5.253 5.276
arnoldi_normal_ev 4 7.8 0.003 0.010 5.253 5.275
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.961 5.150
build_subspace 16 8.4 0.014 0.020 4.910 4.921
calculate_norms 1584 9.8 4.258 4.593 4.258 4.593
mp_allgather_i34 111 8.7 1.457 3.848 1.457 3.848
dbcsr_matrix_vector_mult 304 9.0 0.005 0.016 3.507 3.805
dbcsr_matrix_vector_mult_local 304 10.0 3.045 3.614 3.047 3.616
ls_scf_post 1 4.0 0.000 0.001 3.527 3.533
ls_scf_store_result 1 5.0 0.000 0.000 3.284 3.353
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.786 2.909
make_images_sizes 222 9.7 0.000 0.000 1.083 2.218
mp_alltoall_i44 222 10.7 1.082 2.217 1.082 2.217
make_images_pack 222 9.7 1.814 2.139 1.817 2.141
dbcsr_sort_data 325 11.1 1.871 2.111 1.871 2.111
dbcsr_data_release 9322 10.9 1.287 1.965 1.287 1.965
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=94.275000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2179.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 3.340804E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266696.
MP_Allreduce 3058 13371.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_ISendRecv 138 86400.
MP_Wait 600
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.036 0.050 107.773 107.774
qs_energies 1 2.0 0.000 0.000 107.148 107.154
ls_scf 1 3.0 0.000 0.003 104.412 104.418
dbcsr_multiply_generic 111 6.7 0.018 0.028 86.103 86.324
ls_scf_main 1 4.0 0.000 0.002 65.253 65.254
multiply_cannon 111 7.7 0.063 0.185 55.348 59.516
density_matrix_trs4 2 5.0 0.002 0.003 57.808 58.163
multiply_cannon_loop 111 8.7 0.114 0.125 50.233 53.790
ls_scf_init_scf 1 4.0 0.004 0.015 35.915 35.920
mp_waitall_1 6438 11.0 28.247 34.690 28.247 34.690
ls_scf_init_matrix_S 1 5.0 0.000 0.000 34.098 34.185
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.004 31.504 31.519
make_m2s 222 7.7 0.007 0.008 23.393 24.774
make_images 222 8.7 3.158 3.616 23.343 24.726
multiply_cannon_multrec 1332 9.7 14.134 16.939 22.022 24.640
multiply_cannon_metrocomm3 1332 9.7 0.003 0.003 10.991 20.291
make_images_data 222 9.7 0.004 0.006 13.658 15.441
hybrid_alltoall_any 227 10.6 0.796 3.768 12.516 14.536
mp_sum_l 807 5.4 5.769 9.799 5.769 9.799
dbcsr_mm_accdrv_process 3641 10.4 0.228 0.406 7.532 9.064
dbcsr_mm_accdrv_process_sort 3641 11.4 7.157 8.650 7.157 8.650
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.500 7.975
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 4.130 7.788
multiply_cannon_sync_h2d 1332 9.7 5.481 6.275 5.481 6.275
multiply_cannon_metrocomm4 1110 9.7 0.004 0.006 2.076 5.971
mp_irecv_dv 3229 10.9 2.057 5.897 2.057 5.897
arnoldi_extremal 4 6.8 0.002 0.011 5.575 5.592
arnoldi_normal_ev 4 7.8 0.007 0.033 5.573 5.592
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.858 5.408
build_subspace 16 8.4 0.016 0.023 5.162 5.170
mp_allgather_i34 111 8.7 2.219 4.591 2.219 4.591
calculate_norms 2376 9.8 4.168 4.502 4.168 4.502
dbcsr_matrix_vector_mult 304 9.0 0.006 0.017 3.629 3.950
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.775 3.809
dbcsr_matrix_vector_mult_local 304 10.0 3.220 3.715 3.222 3.717
dbcsr_sort_data 658 11.4 3.050 3.362 3.050 3.362
ls_scf_post 1 4.0 0.000 0.001 3.243 3.248
dbcsr_special_finalize 555 9.7 0.006 0.007 2.795 3.160
dbcsr_merge_single_wm 555 10.7 0.536 0.664 2.787 3.151
ls_scf_store_result 1 5.0 0.000 0.000 2.987 3.049
make_images_sizes 222 9.7 0.000 0.000 1.092 2.966
mp_alltoall_i44 222 10.7 1.092 2.966 1.092 2.966
qs_energies_init_hamiltonians 1 3.0 0.007 0.026 2.726 2.729
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.634 2.636
rebuild_ks_matrix 3 7.3 0.000 0.000 2.610 2.612
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.610 2.612
dbcsr_data_release 10477 10.7 1.570 2.371 1.570 2.371
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=107.774000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2699.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 4.636778E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265558.
MP_Allreduce 3049 15663.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_ISendRecv 90 115200.
MP_Wait 573
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.040 0.059 100.440 100.441
qs_energies 1 2.0 0.000 0.000 99.624 99.629
ls_scf 1 3.0 0.000 0.000 97.705 97.709
dbcsr_multiply_generic 111 6.7 0.017 0.018 79.141 79.386
ls_scf_main 1 4.0 0.000 0.000 62.463 62.464
multiply_cannon 111 7.7 0.072 0.115 56.156 61.656
density_matrix_trs4 2 5.0 0.002 0.003 55.412 55.514
multiply_cannon_loop 111 8.7 0.069 0.075 51.745 53.357
mp_waitall_1 5481 11.0 27.055 33.113 27.055 33.113
ls_scf_init_scf 1 4.0 0.000 0.001 31.613 31.617
ls_scf_init_matrix_S 1 5.0 0.000 0.000 30.432 30.475
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.238 28.252
multiply_cannon_multrec 444 9.7 14.031 16.593 21.107 22.852
make_m2s 222 7.7 0.004 0.005 18.125 20.513
make_images 222 8.7 3.743 4.439 18.064 20.453
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 11.327 16.494
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 6.284 15.075
make_images_data 222 9.7 0.003 0.004 10.243 12.637
hybrid_alltoall_any 227 10.6 0.790 3.775 9.993 12.547
dbcsr_mm_accdrv_process 3003 10.4 0.176 0.341 6.784 7.971
multiply_cannon_sync_h2d 444 9.7 6.520 7.596 6.520 7.596
dbcsr_mm_accdrv_process_sort 3003 11.4 6.460 7.578 6.460 7.578
mp_allgather_i34 111 8.7 2.636 7.038 2.636 7.038
arnoldi_extremal 4 6.8 0.000 0.000 6.041 6.053
arnoldi_normal_ev 4 7.8 0.002 0.007 6.041 6.053
build_subspace 16 8.4 0.015 0.020 5.628 5.639
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.653 4.787
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.694 4.591
mp_irecv_dv 1241 11.2 1.675 4.565 1.675 4.565
mp_sum_l 807 5.4 2.905 4.516 2.905 4.516
dbcsr_matrix_vector_mult 304 9.0 0.007 0.017 4.300 4.482
dbcsr_matrix_vector_mult_local 304 10.0 3.730 4.199 3.732 4.201
make_images_sizes 222 9.7 0.000 0.000 1.211 3.696
mp_alltoall_i44 222 10.7 1.210 3.696 1.210 3.696
calculate_norms 792 9.8 3.550 3.691 3.550 3.691
ls_scf_post 1 4.0 0.000 0.000 3.628 3.634
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.396 3.499
ls_scf_store_result 1 5.0 0.000 0.000 3.408 3.457
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.991 3.289
dbcsr_finalize 304 7.8 0.062 0.078 2.201 2.255
dbcsr_merge_all 275 8.9 0.475 0.532 2.046 2.087
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=100.441000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3622.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44_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 8.719491E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284111.
MP_Allreduce 3043 21950.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_ISendRecv 84 732600.
MP_Wait 309
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.089 0.125 107.970 107.978
qs_energies 1 2.0 0.000 0.000 106.162 106.177
ls_scf 1 3.0 0.000 0.000 103.283 103.298
dbcsr_multiply_generic 111 6.7 0.023 0.026 77.211 77.339
ls_scf_main 1 4.0 0.000 0.000 65.486 65.487
density_matrix_trs4 2 5.0 0.002 0.003 56.633 56.688
multiply_cannon 111 7.7 0.154 0.248 49.628 51.790
multiply_cannon_loop 111 8.7 0.067 0.069 46.179 46.764
ls_scf_init_scf 1 4.0 0.001 0.001 34.060 34.061
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.686 32.697
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.853 29.863
mp_waitall_1 4569 11.1 21.993 25.693 21.993 25.693
make_m2s 222 7.7 0.005 0.005 24.015 24.989
make_images 222 8.7 4.623 5.011 23.908 24.881
multiply_cannon_multrec 444 9.7 17.815 18.617 22.453 23.208
hybrid_alltoall_any 227 10.6 1.661 3.621 12.825 15.901
make_images_data 222 9.7 0.003 0.003 13.146 15.664
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.254 10.748
multiply_cannon_sync_h2d 444 9.7 8.829 8.872 8.829 8.872
arnoldi_extremal 4 6.8 0.000 0.000 7.584 7.597
arnoldi_normal_ev 4 7.8 0.003 0.009 7.583 7.597
build_subspace 16 8.4 0.026 0.036 7.030 7.039
dbcsr_matrix_vector_mult 304 9.0 0.009 0.026 5.585 5.729
dbcsr_matrix_vector_mult_local 304 10.0 5.102 5.409 5.105 5.411
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.013 5.281
ls_scf_dm_to_ks 2 5.0 0.000 0.000 4.788 4.879
dbcsr_mm_accdrv_process 1814 10.4 0.242 0.347 4.472 4.589
dbcsr_mm_accdrv_process_sort 1814 11.4 4.157 4.283 4.157 4.283
make_images_sizes 222 9.7 0.000 0.000 1.505 3.773
mp_alltoall_i44 222 10.7 1.505 3.773 1.505 3.773
ls_scf_post 1 4.0 0.000 0.000 3.736 3.751
ls_scf_store_result 1 5.0 0.000 0.000 3.477 3.496
mp_allgather_i34 111 8.7 1.091 3.375 1.091 3.375
calculate_norms 792 9.8 3.258 3.318 3.258 3.318
dbcsr_finalize 304 7.8 0.082 0.089 3.082 3.127
dbcsr_merge_all 275 8.9 0.893 0.916 2.868 2.907
dbcsr_complete_redistribute 5 7.6 1.435 1.474 2.769 2.866
qs_energies_init_hamiltonians 1 3.0 0.002 0.005 2.849 2.849
dbcsr_data_release 12724 10.6 2.321 2.823 2.321 2.823
matrix_ls_to_qs 2 6.0 0.000 0.000 2.414 2.527
dbcsr_sort_data 325 11.1 2.444 2.500 2.444 2.500
dbcsr_new_transposed 4 7.5 0.243 0.252 2.299 2.314
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.236 2.238
dbcsr_frobenius_norm 74 6.6 2.058 2.126 2.178 2.217
dbcsr_add_d 103 6.2 0.000 0.000 2.135 2.205
dbcsr_add_anytype 103 7.2 0.859 0.893 2.135 2.205
rebuild_ks_matrix 3 7.3 0.000 0.000 2.169 2.171
qs_ks_build_kohn_sham_matrix 3 8.3 0.001 0.001 2.169 2.171
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=107.978000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6835.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: c46dbb59861cfd0a7f9d91cd65bb6cd51d673c44
Summary: empty
Status: OK