=== 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: 183bfa028a10482cb2ce2952b7487d62cedfbee6
################# ARCHITECTURE FILE ##################
#!/bin/bash
#
# CP2K arch file for Cray-XC50 (Piz Daint, CSCS, GPU partition)
#
# Tested with: GNU 9.3.0, Cray-MPICH 7.7.18, Cray-libsci 20.09.1,
# Cray-FFTW 3.3.8.10, COSMA 2.6.6, ELPA 2023.05.001,
# HDF5 1.14.2, LIBINT 2.6.0, LIBPEXSI 1.2.0,
# LIBXC 6.2.2, LIBVORI 220621, LIBXSMM 1.17,
# PLUMED 2.9.0, SIRIUS 7.4.3, SPGLIB 1.16.2
#
# Usage: Source this arch file and then run make as instructed.
# A full toolchain installation is performed as default.
# Replace or adapt the "module add" commands below if needed.
#
# Last update: 04.10.2023
#
# \
if [ "${0}" = "${BASH_SOURCE}" ]; then \
echo "ERROR: Script ${0##*/} must be sourced"; \
echo "Usage: source ${0##*/}"; \
exit 1; \
fi; \
this_file=${BASH_SOURCE##*/}; \
if [ -n "${1}" ]; then \
gcc_version="${1}"; \
else \
gcc_version="9.3.0"; \
fi; \
module add daint-gpu; \
module rm PrgEnv-cray; \
module add PrgEnv-gnu; \
module rm gcc; \
module add gcc/${gcc_version}; \
module add cray-fftw/3.3.8.10; \
module add cudatoolkit; \
echo "Expected setup:"; \
echo " cray-mpich/7.7.18"; \
echo " craype-haswell"; \
echo " daint-gpu/21.09"; \
echo " craype/2.7.10"; \
echo " cray-libsci/20.09.1"; \
echo " PrgEnv-gnu/6.0.10"; \
echo " gcc/${gcc_version}"; \
echo " cray-fftw/3.3.8.10"; \
echo " cudatoolkit/11.0.2_3.38-8.1__g5b73779"; \
module list; \
module -f save cp2k_gpu_gnu_psmp; \
echo "To load the required modules in your batch job script, use:"; \
echo " module restore cp2k_gpu_gnu_psmp"; \
cd tools/toolchain; \
./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=P100 -j${maxtasks} --no-arch-files --with-gcc=system --with-libvdwxc --with-pexsi --with-plumed; \
cd ../..; \
printf "Sourcing ${PWD}/tools/toolchain/install/setup ... "; \
source ${PWD}/tools/toolchain/install/setup; \
printf "done\n"; \
echo "Check the output above for error messages and consistency!"; \
echo; \
echo "If everything is OK, you can build a CP2K production binary with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.}"; \
echo; \
echo "Alternatively, you can add further checks, e.g. for regression testing, with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} DO_CHECKS=yes"; \
echo "or build CP2K as a library with"; \
echo " make -j ARCH=${this_file%.*} VERSION=${this_file##*.} libcp2k"; \
echo; \
return
# Set options
DO_CHECKS := no
USE_ACC := yes
USE_COSMA := 2.6.6
USE_ELPA := 2023.05.001
USE_HDF5 := 1.14.2
USE_LIBINT := 2.6.0
USE_LIBPEXSI := 1.2.0
USE_LIBVORI := 220621
USE_LIBXC := 6.2.2
USE_LIBXSMM := 1.17
USE_PLUMED := 2.9.0
#USE_QUIP := 0.9.10
USE_SIRIUS := 7.4.3
USE_SPGLIB := 1.16.2
# Only needed for SIRIUS
LIBVDWXC_VER := 0.4.0
SPFFT_VER := 1.0.6
SPLA_VER := 1.5.5
# Only needed for LIBPEXSI
SCOTCH_VER := 6.0.0
SUPERLU_VER := 6.1.0
LMAX := 5
MAX_CONTR := 4
GPUVER := P100
OFFLOAD_TARGET := cuda
CC := cc
CXX := CC
OFFLOAD_CC := nvcc
FC := ftn
LD := ftn
AR := ar -r
# cc, CC, and ftn include already the proper -march flag
CFLAGS := -O2 -fopenmp -fopenmp-simd -ftree-vectorize -funroll-loops -g
DFLAGS := -D__parallel
DFLAGS += -D__SCALAPACK
DFLAGS += -D__FFTW3
DFLAGS += -D__MAX_CONTR=$(strip $(MAX_CONTR))
INSTALL_PATH := $(PWD)/tools/toolchain/install
ifeq ($(DO_CHECKS), yes)
DFLAGS += -D__CHECK_DIAG
endif
ifeq ($(USE_ACC), yes)
DFLAGS += -D__DBCSR_ACC
DFLAGS += -D__OFFLOAD_CUDA
# Possibly no performance gain with PW_CUDA currently
DFLAGS += -D__NO_OFFLOAD_PW
endif
ifneq ($(USE_PLUMED),)
USE_PLUMED := $(strip $(USE_PLUMED))
PLUMED_LIB := $(INSTALL_PATH)/plumed-$(USE_PLUMED)/lib
DFLAGS += -D__PLUMED2
USE_GSL := 2.7
LIBS += $(PLUMED_LIB)/libplumed.a
endif
ifneq ($(USE_ELPA),)
USE_ELPA := $(strip $(USE_ELPA))
TARGET := nvidia
ELPA_INC := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/include/elpa-$(USE_ELPA)
ELPA_LIB := $(INSTALL_PATH)/elpa-$(USE_ELPA)/$(TARGET)/lib
CFLAGS += -I$(ELPA_INC)/elpa -I$(ELPA_INC)/modules
DFLAGS += -D__ELPA
ifeq ($(TARGET), nvidia)
DFLAGS += -D__ELPA_NVIDIA_GPU
endif
LIBS += $(ELPA_LIB)/libelpa.a
endif
ifneq ($(USE_QUIP),)
USE_QUIP := $(strip $(USE_QUIP))
QUIP_INC := $(INSTALL_PATH)/quip-$(USE_QUIP)/include
QUIP_LIB := $(INSTALL_PATH)/quip-$(USE_QUIP)/lib
CFLAGS += -I$(QUIP_INC)
DFLAGS += -D__QUIP
LIBS += $(QUIP_LIB)/libquip_core.a
LIBS += $(QUIP_LIB)/libatoms.a
LIBS += $(QUIP_LIB)/libFoX_sax.a
LIBS += $(QUIP_LIB)/libFoX_common.a
LIBS += $(QUIP_LIB)/libFoX_utils.a
LIBS += $(QUIP_LIB)/libFoX_fsys.a
endif
ifneq ($(USE_LIBPEXSI),)
USE_LIBPEXSI := $(strip $(USE_LIBPEXSI))
SCOTCH_VER := $(strip $(SCOTCH_VER))
SUPERLU_VER := $(strip $(SUPERLU_VER))
LIBPEXSI_INC := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/include
LIBPEXSI_LIB := $(INSTALL_PATH)/pexsi-$(USE_LIBPEXSI)/lib
SCOTCH_INC := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/include
SCOTCH_LIB := $(INSTALL_PATH)/scotch-$(SCOTCH_VER)/lib
SUPERLU_INC := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/include
SUPERLU_LIB := $(INSTALL_PATH)/superlu_dist-$(SUPERLU_VER)/lib
CFLAGS += -I$(LIBPEXSI_INC) -I$(SCOTCH_INC) -I$(SUPERLU_INC)
DFLAGS += -D__LIBPEXSI
LIBS += $(LIBPEXSI_LIB)/libpexsi.a
LIBS += $(SUPERLU_LIB)/libsuperlu_dist.a
LIBS += $(SCOTCH_LIB)/libptscotchparmetis.a
LIBS += $(SCOTCH_LIB)/libptscotch.a
LIBS += $(SCOTCH_LIB)/libptscotcherr.a
LIBS += $(SCOTCH_LIB)/libscotchmetis.a
LIBS += $(SCOTCH_LIB)/libscotch.a
endif
ifneq ($(USE_LIBVORI),)
USE_LIBVORI := $(strip $(USE_LIBVORI))
LIBVORI_LIB := $(INSTALL_PATH)/libvori-$(USE_LIBVORI)/lib
DFLAGS += -D__LIBVORI
LIBS += $(LIBVORI_LIB)/libvori.a
endif
ifneq ($(USE_LIBXC),)
USE_LIBXC := $(strip $(USE_LIBXC))
LIBXC_INC := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/include
LIBXC_LIB := $(INSTALL_PATH)/libxc-$(USE_LIBXC)/lib
CFLAGS += -I$(LIBXC_INC)
DFLAGS += -D__LIBXC
LIBS += $(LIBXC_LIB)/libxcf03.a
LIBS += $(LIBXC_LIB)/libxc.a
endif
ifneq ($(USE_LIBINT),)
USE_LIBINT := $(strip $(USE_LIBINT))
LMAX := $(strip $(LMAX))
LIBINT_INC := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/include
LIBINT_LIB := $(INSTALL_PATH)/libint-v$(USE_LIBINT)-cp2k-lmax-$(LMAX)/lib
CFLAGS += -I$(LIBINT_INC)
DFLAGS += -D__LIBINT
LIBS += $(LIBINT_LIB)/libint2.a
endif
ifneq ($(USE_SPGLIB),)
USE_SPGLIB := $(strip $(USE_SPGLIB))
SPGLIB_INC := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/include
SPGLIB_LIB := $(INSTALL_PATH)/spglib-$(USE_SPGLIB)/lib
CFLAGS += -I$(SPGLIB_INC)
DFLAGS += -D__SPGLIB
LIBS += $(SPGLIB_LIB)/libsymspg.a
endif
ifneq ($(USE_LIBXSMM),)
USE_LIBXSMM := $(strip $(USE_LIBXSMM))
LIBXSMM_INC := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/include
LIBXSMM_LIB := $(INSTALL_PATH)/libxsmm-$(USE_LIBXSMM)/lib
CFLAGS += -I$(LIBXSMM_INC)
DFLAGS += -D__LIBXSMM
LIBS += $(LIBXSMM_LIB)/libxsmmf.a
LIBS += $(LIBXSMM_LIB)/libxsmm.a
endif
ifneq ($(USE_SIRIUS),)
USE_SIRIUS := $(strip $(USE_SIRIUS))
LIBVDWXC_VER := $(strip $(LIBVDWXC_VER))
LIBVDWXC_INC := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/include
LIBVDWXC_LIB := $(INSTALL_PATH)/libvdwxc-$(LIBVDWXC_VER)/lib
SPFFT_VER := $(strip $(SPFFT_VER))
SPFFT_INC := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/include
SPLA_VER := $(strip $(SPLA_VER))
SPLA_INC := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/include/spla
ifeq ($(USE_ACC), yes)
DFLAGS += -D__OFFLOAD_GEMM
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib/cuda
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib/cuda
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include/cuda
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib/cuda
else
SPFFT_LIB := $(INSTALL_PATH)/SpFFT-$(SPFFT_VER)/lib
SPLA_LIB := $(INSTALL_PATH)/SpLA-$(SPLA_VER)/lib
SIRIUS_INC := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/include
SIRIUS_LIB := $(INSTALL_PATH)/sirius-$(USE_SIRIUS)/lib
endif
CFLAGS += -I$(LIBVDWXC_INC)
CFLAGS += -I$(SPFFT_INC)
CFLAGS += -I$(SPLA_INC)
CFLAGS += -I$(SIRIUS_INC)
DFLAGS += -D__LIBVDWXC
DFLAGS += -D__SPFFT
DFLAGS += -D__SPLA
DFLAGS += -D__SIRIUS
LIBS += $(SIRIUS_LIB)/libsirius.a
LIBS += $(SPLA_LIB)/libspla.a
LIBS += $(SPFFT_LIB)/libspfft.a
LIBS += $(LIBVDWXC_LIB)/libvdwxc.a
endif
ifneq ($(USE_HDF5),)
USE_HDF5 := $(strip $(USE_HDF5))
HDF5_INC := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/include
HDF5_LIB := $(INSTALL_PATH)/hdf5-$(USE_HDF5)/lib
CFLAGS += -I$(HDF5_INC)
DFLAGS += -D__HDF5
LIBS += $(HDF5_LIB)/libhdf5_fortran.a
LIBS += $(HDF5_LIB)/libhdf5_hl.a
LIBS += $(HDF5_LIB)/libhdf5.a
endif
ifneq ($(USE_COSMA),)
USE_COSMA := $(strip $(USE_COSMA))
ifeq ($(USE_ACC), yes)
USE_COSMA := $(USE_COSMA)-cuda
endif
COSMA_INC := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/include
COSMA_LIB := $(INSTALL_PATH)/COSMA-$(USE_COSMA)/lib
CFLAGS += -I$(COSMA_INC)
DFLAGS += -D__COSMA
LIBS += $(COSMA_LIB)/libcosma_prefixed_pxgemm.a
LIBS += $(COSMA_LIB)/libcosma.a
LIBS += $(COSMA_LIB)/libcosta.a
LIBS += $(COSMA_LIB)/libTiled-MM.a
endif
ifneq ($(USE_GSL),)
USE_GSL := $(strip $(USE_GSL))
GSL_INC := $(INSTALL_PATH)/gsl-$(USE_GSL)/include
GSL_LIB := $(INSTALL_PATH)/gsl-$(USE_GSL)/lib
CFLAGS += -I$(GSL_INC)
DFLAGS += -D__GSL
LIBS += $(GSL_LIB)/libgsl.a
endif
CFLAGS += $(DFLAGS)
CXXFLAGS := $(CFLAGS) -std=c++11
OFFLOAD_FLAGS := $(DFLAGS) -O3 -Xcompiler="-fopenmp" -arch sm_60 --std=c++11
FCFLAGS := $(CFLAGS)
ifeq ($(shell [ $(shell gcc -dumpversion | cut -d. -f1) -gt 9 ] && echo yes), yes)
FCFLAGS += -fallow-argument-mismatch
endif
FCFLAGS += -fbacktrace
FCFLAGS += -ffree-form
FCFLAGS += -ffree-line-length-none
FCFLAGS += -fno-omit-frame-pointer
FCFLAGS += -std=f2008
ifneq ($(CUDA_HOME),)
CUDA_LIB := $(CUDA_HOME)/lib64
LDFLAGS := $(FCFLAGS) -L$(CUDA_LIB) -Wl,-rpath=$(CUDA_LIB)
else
LDFLAGS := $(FCFLAGS)
endif
LIBS += -lcusolver -lcudart -lnvrtc -lcuda -lcufft -lcublas -lrt
LIBS += -lz -ldl -lpthread -lstdc++
# End
############### END ARCHITECTURE FILE ################
===== TESTS (description) =====
~~~~~~~~~ TEST ~~~~~~~~~
description: H2O-32 RI-RPA/RI-MP2 correlation energy
input file: benchmarks/QS_mp2_rpa/32-H2O/RI-RPA.inp
required files: ['benchmarks/QS_mp2_rpa/32-H2O/BASIS_H2O', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32.xyz', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-PBE-TZ.inp', 'benchmarks/QS_mp2_rpa/32-H2O/H2O-32-RI-dRPA-TZ.inp']
output file: result.log
# nodes = 8
# ranks/node = 2
# threads/rank = 6
nrepeat = 1
time[min] = 15
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/01
job id: 50175986
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/02
job id: 50175987
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/03
job id: 50175988
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/04
job id: 50175989
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/05
job id: 50175990
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/06
job id: 50175991
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/07
job id: 50175992
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/08
job id: 50175993
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/09
job id: 50175994
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/10
job id: 50175995
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/11
job id: 50175996
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/12
job id: 50175997
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/13
job id: 50175998
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/14
job id: 50175999
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/15
job id: 50176000
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/16
job id: 50176001
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/17
job id: 50176002
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/18
job id: 50176003
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/19
job id: 50176004
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/20
job id: 50176005
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/21
job id: 50176006
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/22
job id: 50176007
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/23
job id: 50176008
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/24
job id: 50176009
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/25
job id: 50176010
--- 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/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/26
job id: 50176011
--- Point ---
name: 510
plot: h2o_32_nrep3_ls
regex: CP2K
label: (8n/1r/12t)
--- Point ---
name: 511
plot: h2o_32_nrep3_ls_mem
regex: Estimated peak process memory
label: (8n/1r/12t)
~~~~~~~ END TEST ~~~~~~~
~~~~~~~~~ TEST ~~~~~~~~~
description: 512 H2O (4 NVE MD steps on 64 nodes)
input file: benchmarks/QS/00512_H2O/H2O-512_md.inp
required files: []
output file: result.log
# nodes = 64
# ranks/node = 12
# threads/rank = 1
nrepeat = 1
time[min] = 10
run dir: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/27
job id: 50176012
--- Point ---
name: 601
plot: h2o_512_md
regex: CP2K
label: (64n/12r/1t)
--- Point ---
name: 602
plot: h2o_512_md_mem
regex: Estimated peak process memory
label: (64n/12r/1t)
~~~~~~~ END TEST ~~~~~~~
=== END TESTS (description) ===
===== PLOTS (description) =====
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_ri_rpa_mp2_mem", title="32 H2O molecules (RI-MP2, RI-RPA)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_64_md_mem", title="64 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_128_md_mem", title="128 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_256_md_mem", title="256 H2O molecules (10 MD steps)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_32_nrep3_ls_mem", title="864 H2O molecules (LS SCF)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Time [s]"
~~~~~~~~~ PLOT ~~~~~~~~~
Plot: name="h2o_512_md_mem", title="512 H2O (4 NVE MD steps on 64 nodes)", xlabel="Revision", ylabel="Est. peak process memory [MiB]"
=== END PLOTS (description) ===
============ RESULTS ============
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 424 8.
MP_Sync 3
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.017 0.033 134.384 134.384
farming_run 1 2.0 133.678 133.689 134.352 134.355
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 1.488757E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 2592
MPI messages size (bytes):
total size 1.140326E+09
min size 0.000000E+00
max size 1.663488E+06
average size 439.940750E+03
MPI breakdown and total messages size (bytes):
size <= 128 132 0
128 < size <= 8192 348 2850816
8192 < size <= 32768 0 0
32768 < size <= 131072 1536 179306496
131072 < size <= 4194304 576 958169088
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 2308 54.
MP_Alltoall 4670 822215.
MP_ISend 2604 90577.
MP_IRecv 2604 90574.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 230 1134128.
MP_Allreduce 571 1938539.
MP_Sync 25
MP_Alltoall 38 9316958.
MP_SendRecv 120 384007.
MP_ISendRecv 45 235435.
MP_Wait 191
MP_comm_split 10
MP_ISend 127 3867574.
MP_IRecv 127 3866554.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.007 0.028 115.416 115.416
qs_energies 1 2.0 0.000 0.000 115.127 115.130
mp2_main 1 3.0 0.000 0.000 113.066 113.069
mp2_gpw_main 1 4.0 0.027 0.033 112.068 112.071
mp2_ri_gpw_compute_in 1 5.0 0.173 0.174 92.742 93.339
mp2_ri_gpw_compute_in_loop 1 6.0 0.004 0.004 54.586 55.183
mp2_eri_3c_integrate_gpw 272 7.0 0.152 0.166 40.997 46.281
get_2c_integrals 1 6.0 0.008 0.009 37.050 37.982
integrate_v_rspace 273 8.0 0.436 0.451 24.594 29.587
pw_transfer 6555 10.6 0.379 0.398 26.812 27.461
fft_wrap_pw1pw2 5465 11.4 0.045 0.048 25.441 26.172
grid_integrate_task_list 273 9.0 20.509 25.959 20.509 25.959
fft_wrap_pw1pw2_100 2178 12.4 0.027 0.028 22.981 23.696
rpa_ri_compute_en 1 5.0 0.019 0.021 19.223 19.488
compute_2c_integrals 1 7.0 0.002 0.002 19.175 19.177
compute_2c_integrals_loop_lm 1 8.0 0.001 0.003 18.597 18.912
mp2_eri_2c_integrate_gpw 1 9.0 2.381 2.415 18.595 18.911
cp_fm_cholesky_decompose 12 8.2 17.783 18.732 17.783 18.732
cholesky_decomp 1 7.0 0.000 0.000 16.707 17.659
fft3d_s 5443 13.4 16.122 16.643 16.143 16.664
ao_to_mo_and_store_B_mult_1 272 7.0 10.765 15.332 10.765 15.332
calculate_wavefunction 272 8.0 5.399 5.553 12.316 13.003
rpa_num_int 1 6.0 0.001 0.011 10.590 10.590
rpa_num_int_RPA_matrix_operati 8 7.0 0.000 0.000 10.576 10.583
calc_mat_Q 8 8.0 0.000 0.000 9.452 9.537
contract_S_to_Q 8 9.0 0.000 0.000 8.874 8.962
calc_potential_gpw 544 9.5 0.004 0.005 8.192 8.639
parallel_gemm_fm 14 9.1 0.000 0.000 8.457 8.556
parallel_gemm_fm_cosma 14 10.1 8.457 8.556 8.457 8.556
mp2_eri_2c_integrate_gpw_pot_l 272 10.0 0.001 0.001 8.070 8.336
potential_pw2rs 545 10.0 0.106 0.108 7.510 8.133
create_integ_mat 1 6.0 0.022 0.029 7.945 7.945
collocate_single_gaussian 272 10.0 0.039 0.042 7.342 7.630
array2fm 1 7.0 0.000 0.000 6.747 7.338
pw_scatter_s 2720 13.7 4.332 4.488 4.332 4.488
pw_gather_s 2722 13.2 3.457 3.782 3.457 3.782
array2fm_buffer_send 1 8.0 3.023 3.187 3.023 3.187
pw_poisson_solve 545 10.5 1.119 1.181 2.194 2.347
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="10", plot="h2o_32_ri_rpa_mp2", label="RI-RPA (8n/2r/6t)", y=112.067373, yerr=0.000000
PlotPoint: name="11", plot="h2o_32_ri_rpa_mp2_mem", label="RI-RPA (8n/2r/6t)", y=2808.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 424 9.
MP_Sync 4
MP_comm_split 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.037 0.042 419.980 419.982
farming_run 1 2.0 419.175 419.182 419.925 419.928
-------------------------------------------------------------------------------
@@@@@@@@@@ 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 897827128576 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 249788821 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 1.249559E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 61440
MPI messages size (bytes):
total size 6.073508E+09
min size 0.000000E+00
max size 642.960000E+03
average size 98.852664E+03
MPI breakdown and total messages size (bytes):
size <= 128 32004 0
128 < size <= 8192 1820 14909440
8192 < size <= 32768 0 0
32768 < size <= 131072 18640 1081442304
131072 < size <= 4194304 8976 4977156096
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 1003 44.
MP_Alltoall 1797 713538.
MP_ISend 3686 54943.
MP_IRecv 3622 54292.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 12
MP_Bcast 757 478553.
MP_Allreduce 2021 21391.
MP_Sync 37
MP_Alltoall 77 14112249.
MP_SendRecv 2876 2171486.
MP_ISendRecv 1034 172620.
MP_Wait 1346
MP_comm_split 7
MP_ISend 264 362227.
MP_IRecv 264 362718.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.034 217.088 217.089
qs_energies 1 2.0 0.002 0.017 216.869 216.877
scf_env_do_scf 1 3.0 0.000 0.000 114.971 114.972
qs_ks_update_qs_env 5 5.0 0.000 0.000 114.069 114.076
rebuild_ks_matrix 4 6.0 0.000 0.000 114.068 114.075
qs_ks_build_kohn_sham_matrix 4 7.0 0.054 0.060 114.068 114.075
hfx_ks_matrix 4 8.0 0.001 0.001 113.716 113.720
integrate_four_center 4 9.0 0.152 0.463 113.715 113.720
integrate_four_center_main 4 10.0 0.115 0.519 101.719 105.634
integrate_four_center_bin 263 11.0 101.604 105.506 101.604 105.506
mp2_main 1 3.0 0.004 0.031 101.604 101.614
mp2_gpw_main 1 4.0 0.055 0.106 100.707 100.720
init_scf_loop 1 4.0 0.000 0.000 97.243 97.243
mp2_ri_gpw_compute_in 1 5.0 0.067 0.115 73.904 74.899
mp2_ri_gpw_compute_in_loop 1 6.0 0.002 0.002 53.686 54.697
mp2_eri_3c_integrate_gpw 91 7.0 0.143 0.163 41.257 46.439
integrate_v_rspace 95 8.0 0.396 0.573 27.745 32.787
pw_transfer 2240 10.6 0.144 0.165 29.525 29.963
fft_wrap_pw1pw2 1868 11.4 0.017 0.020 28.529 28.961
ao_to_mo_and_store_B_mult_1 91 7.0 10.759 28.742 10.759 28.742
mp2_ri_gpw_compute_en 1 5.0 0.054 0.065 26.597 28.614
grid_integrate_task_list 95 9.0 23.161 28.415 23.161 28.415
fft_wrap_pw1pw2_100 730 12.4 0.012 0.013 26.270 26.681
mp2_ri_gpw_compute_en_RI_loop 1 6.0 1.838 1.901 24.937 24.947
get_2c_integrals 1 6.0 0.001 0.011 20.129 20.153
compute_2c_integrals 1 7.0 0.005 0.023 19.104 19.112
compute_2c_integrals_loop_lm 1 8.0 0.001 0.002 18.681 18.944
mp2_eri_2c_integrate_gpw 1 9.0 1.727 1.841 18.680 18.943
fft3d_s 1823 13.4 18.481 18.905 18.494 18.919
scf_env_do_scf_inner_loop 4 4.0 0.000 0.001 17.727 17.727
calculate_wavefunction 91 8.0 2.011 2.041 9.616 9.838
mp2_ri_gpw_compute_en_expansio 172 7.0 0.555 0.584 8.798 9.420
potential_pw2rs 186 10.0 0.033 0.034 8.451 8.935
local_gemm 172 8.0 8.243 8.850 8.243 8.850
mp2_eri_2c_integrate_gpw_pot_l 91 10.0 0.001 0.001 8.190 8.567
mp2_ri_gpw_compute_en_comm 22 7.0 0.501 0.524 7.918 8.317
calc_potential_gpw 182 9.5 0.002 0.002 7.907 8.158
collocate_single_gaussian 91 10.0 0.017 0.034 7.776 8.010
mp_sync 37 10.5 3.973 7.960 3.973 7.960
integrate_four_center_load 4 10.0 0.000 0.000 6.760 6.763
hfx_load_balance 1 11.0 0.000 0.000 6.760 6.763
mp2_ri_gpw_compute_en_ener 172 7.0 6.339 6.407 6.339 6.407
mp_sendrecv_dm3 2068 8.0 5.945 6.343 5.945 6.343
pw_gather_s 912 13.2 4.481 4.961 4.481 4.961
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="20", plot="h2o_32_ri_rpa_mp2", label="RI-MP2 (8n/6r/2t)", y=100.693406, yerr=0.000000
PlotPoint: name="21", plot="h2o_32_ri_rpa_mp2_mem", label="RI-MP2 (8n/6r/2t)", y=1509.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/03/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 29.277748E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 5055360 0.0% 0.0% 100.0%
average stack size 0.0 0.0 29.1
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 453.496832E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 9436608
MPI messages size (bytes):
total size 333.233553E+09
min size 0.000000E+00
max size 315.840000E+03
average size 35.312852E+03
MPI breakdown and total messages size (bytes):
size <= 128 4913240 0
128 < size <= 8192 1155432 9465298944
8192 < size <= 32768 1984512 54190407680
32768 < size <= 131072 551296 42776657920
131072 < size <= 4194304 832128 226802306368
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3683 62379.
MP_Allreduce 10329 270.
MP_Sync 530
MP_Alltoall 2083 592243.
MP_SendRecv 22610 5520.
MP_ISendRecv 22610 5520.
MP_Wait 37876
MP_comm_split 50
MP_ISend 20771 42672.
MP_IRecv 20771 42672.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.008 0.027 53.949 53.950
qs_mol_dyn_low 1 2.0 0.003 0.004 53.747 53.756
qs_forces 11 3.9 0.002 0.002 53.674 53.674
qs_energies 11 4.9 0.006 0.011 52.174 52.181
scf_env_do_scf 11 5.9 0.000 0.001 45.921 45.922
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 43.874 43.874
qs_scf_new_mos 108 7.5 0.000 0.001 33.863 34.135
qs_scf_loop_do_ot 108 8.5 0.000 0.001 33.863 34.134
dbcsr_multiply_generic 2286 12.5 0.094 0.098 33.498 33.974
ot_scf_mini 108 9.5 0.002 0.002 32.164 32.394
multiply_cannon 2286 13.5 0.189 0.199 26.223 28.048
multiply_cannon_loop 2286 14.5 1.809 1.917 25.553 27.394
velocity_verlet 10 3.0 0.004 0.011 26.044 26.045
ot_mini 108 10.5 0.001 0.001 19.125 19.399
qs_ot_get_derivative 108 11.5 0.001 0.001 16.130 16.335
mp_waitall_1 245248 16.5 8.250 14.809 8.250 14.809
multiply_cannon_metrocomm3 54864 15.5 0.072 0.078 5.916 13.071
multiply_cannon_multrec 54864 15.5 3.649 5.809 7.797 11.280
qs_ot_get_p 119 10.4 0.001 0.001 8.382 8.685
rebuild_ks_matrix 119 8.3 0.000 0.000 7.857 8.011
qs_ks_build_kohn_sham_matrix 119 9.3 0.010 0.011 7.857 8.011
mp_sum_l 7287 12.8 5.324 7.074 5.324 7.074
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.900 7.036
multiply_cannon_sync_h2d 54864 15.5 5.182 6.069 5.182 6.069
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 5.576 6.042
dbcsr_mm_accdrv_process 76910 16.1 1.837 2.896 4.062 5.791
qs_ot_p2m_diag 50 11.0 0.004 0.006 5.443 5.474
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 5.316 5.431
init_scf_run 11 5.9 0.000 0.001 4.916 4.917
scf_env_initial_rho_setup 11 6.9 0.002 0.007 4.916 4.916
cp_dbcsr_syevd 50 12.0 0.003 0.003 4.594 4.594
sum_up_and_integrate 119 10.3 0.001 0.002 4.460 4.468
integrate_v_rspace 119 11.3 0.002 0.002 4.449 4.459
cp_fm_diag_elpa 50 13.0 0.000 0.000 4.380 4.380
cp_fm_redistribute_end 50 14.0 2.232 4.351 2.239 4.354
qs_rho_update_rho_low 119 7.7 0.000 0.001 4.115 4.250
cp_fm_diag_elpa_base 50 14.0 2.109 4.246 2.114 4.250
calculate_rho_elec 119 8.7 0.011 0.017 4.115 4.249
calculate_dm_sparse 119 9.5 0.000 0.000 3.036 3.193
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.928 3.126
apply_single 119 13.6 0.000 0.000 2.927 3.125
multiply_cannon_metrocomm1 54864 15.5 0.055 0.060 1.683 3.058
jit_kernel_multiply 13 15.8 2.157 2.799 2.157 2.799
acc_transpose_blocks 54864 15.5 0.223 0.245 2.252 2.792
calculate_first_density_matrix 1 7.0 0.000 0.001 2.728 2.733
ot_diis_step 108 11.5 0.006 0.006 2.719 2.720
qs_ot_get_orbitals 108 10.5 0.000 0.000 2.447 2.516
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.416 2.418
density_rs2pw 119 9.7 0.004 0.004 2.146 2.257
grid_integrate_task_list 119 12.3 2.024 2.120 2.024 2.120
wfi_extrapolate 11 7.9 0.001 0.001 2.118 2.118
mp_sum_d 4137 12.0 1.363 2.092 1.363 2.092
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.000 2.054
init_scf_loop 11 6.9 0.000 0.000 2.031 2.031
potential_pw2rs 119 12.3 0.004 0.004 1.820 1.834
pw_transfer 1439 11.6 0.051 0.056 1.681 1.749
make_m2s 4572 13.5 0.053 0.056 1.635 1.684
fft_wrap_pw1pw2 1201 12.6 0.007 0.007 1.606 1.676
make_images 4572 14.5 0.133 0.139 1.552 1.600
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.496 1.517
transfer_rs2pw 487 10.6 0.005 0.006 1.356 1.474
mp_alltoall_d11v 2130 13.8 1.311 1.454 1.311 1.454
acc_transpose_blocks_sync 164592 16.5 1.201 1.440 1.201 1.440
grid_collocate_task_list 119 9.7 1.352 1.430 1.352 1.430
mp_waitany 12084 13.8 1.237 1.391 1.237 1.391
fft3d_ps 1201 14.6 0.371 0.478 1.320 1.381
transfer_pw2rs 487 13.2 0.006 0.006 1.323 1.334
fft_wrap_pw1pw2_140 487 13.2 0.107 0.113 1.246 1.319
dbcsr_dot_sd 1205 11.9 0.049 0.059 0.740 1.138
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="100", plot="h2o_64_md", label="(8n/12r/1t)", y=53.950000, yerr=0.000000
PlotPoint: name="101", plot="h2o_64_md_mem", label="(8n/12r/1t)", y=432.181818, yerr=1.113404
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/04/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 26877100032 0.0% 0.0% 100.0%
flops 9 x 9 x 32 44168260608 0.0% 0.0% 100.0%
flops 22 x 9 x 32 53835724800 0.0% 0.0% 100.0%
flops 9 x 22 x 32 53885500416 0.0% 0.0% 100.0%
flops 32 x 32 x 9 63568871424 0.0% 0.0% 100.0%
flops 22 x 22 x 32 67007283200 0.0% 0.0% 100.0%
flops 32 x 32 x 22 77695287296 0.0% 0.0% 100.0%
flops 9 x 32 x 32 78422999040 0.0% 0.0% 100.0%
flops 22 x 32 x 32 95850332160 0.0% 0.0% 100.0%
flops 9 x 32 x 9 266263676928 0.0% 0.0% 100.0%
flops 22 x 32 x 9 326697440256 0.0% 0.0% 100.0%
flops 9 x 32 x 22 326697440256 0.0% 0.0% 100.0%
flops 22 x 32 x 22 399918497792 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 1.880888E+12 0.0% 0.0% 100.0%
flops max/rank 57.173320E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 146984760 0.0% 0.0% 100.0%
number of processed stacks 3066240 0.0% 0.0% 100.0%
average stack size 0.0 0.0 47.9
marketing flops 2.107592E+12
-------------------------------------------------------------------------------
# multiplications 2286
max memory usage/rank 488.792064E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2194560
MPI messages size (bytes):
total size 310.646604E+09
min size 0.000000E+00
max size 1.145520E+06
average size 141.553031E+03
MPI breakdown and total messages size (bytes):
size <= 128 724648 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 281952 4619501568
32768 < size <= 131072 494448 39143342080
131072 < size <= 4194304 440000 264807943488
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3672 62658.
MP_Allreduce 10306 303.
MP_Sync 54
MP_Alltoall 2060 247987.
MP_SendRecv 16779 37093.
MP_ISendRecv 16779 37093.
MP_Wait 23539
MP_comm_split 50
MP_ISend 5720 128509.
MP_IRecv 5720 128509.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.029 38.707 38.709
qs_mol_dyn_low 1 2.0 0.003 0.004 38.452 38.459
qs_forces 11 3.9 0.002 0.003 38.392 38.392
qs_energies 11 4.9 0.002 0.007 36.724 36.728
scf_env_do_scf 11 5.9 0.001 0.002 31.670 31.671
scf_env_do_scf_inner_loop 108 6.5 0.002 0.006 29.244 29.245
dbcsr_multiply_generic 2286 12.5 0.101 0.105 22.029 22.377
qs_scf_new_mos 108 7.5 0.001 0.001 20.695 20.913
qs_scf_loop_do_ot 108 8.5 0.001 0.001 20.695 20.913
ot_scf_mini 108 9.5 0.003 0.004 19.763 19.921
multiply_cannon 2286 13.5 0.209 0.217 16.820 18.302
velocity_verlet 10 3.0 0.001 0.002 18.226 18.227
multiply_cannon_loop 2286 14.5 1.194 1.254 15.662 17.188
ot_mini 108 10.5 0.001 0.001 12.121 12.347
mp_waitall_1 200699 16.5 5.491 10.652 5.491 10.652
qs_ot_get_derivative 108 11.5 0.001 0.001 9.756 9.918
multiply_cannon_metrocomm3 27432 15.5 0.072 0.075 4.005 9.418
multiply_cannon_multrec 27432 15.5 1.826 4.118 6.148 9.035
rebuild_ks_matrix 119 8.3 0.000 0.000 6.723 6.864
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.018 6.723 6.863
dbcsr_mm_accdrv_process 47894 16.0 3.537 5.815 4.242 6.328
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.913 6.038
qs_ot_get_p 119 10.4 0.001 0.003 4.727 4.940
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.638 4.550
mp_sum_l 7287 12.8 2.125 4.122 2.125 4.122
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 3.068 4.049
apply_single 119 13.6 0.000 0.000 3.068 4.048
init_scf_run 11 5.9 0.000 0.001 3.848 3.849
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.848 3.849
sum_up_and_integrate 119 10.3 0.001 0.002 3.654 3.659
integrate_v_rspace 119 11.3 0.002 0.002 3.640 3.645
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.571 3.598
calculate_rho_elec 119 8.7 0.021 0.024 3.570 3.597
qs_ot_p2m_diag 50 11.0 0.009 0.014 3.063 3.081
make_m2s 4572 13.5 0.051 0.053 2.638 2.964
make_images 4572 14.5 0.205 0.245 2.548 2.876
cp_dbcsr_syevd 50 12.0 0.003 0.003 2.665 2.665
init_scf_loop 11 6.9 0.001 0.004 2.407 2.414
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.325 2.406
calculate_first_density_matrix 1 7.0 0.000 0.002 2.340 2.343
ot_diis_step 108 11.5 0.010 0.011 2.317 2.318
cp_fm_diag_elpa 50 13.0 0.000 0.000 2.314 2.315
cp_fm_redistribute_end 50 14.0 1.171 2.285 1.174 2.287
cp_fm_diag_elpa_base 50 14.0 1.081 2.196 1.109 2.229
calculate_dm_sparse 119 9.5 0.000 0.001 2.079 2.161
multiply_cannon_sync_h2d 27432 15.5 1.674 2.144 1.674 2.144
acc_transpose_blocks 27432 15.5 0.113 0.120 1.607 1.938
density_rs2pw 119 9.7 0.004 0.004 1.868 1.933
grid_integrate_task_list 119 12.3 1.838 1.923 1.838 1.923
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.889 1.891
pw_transfer 1439 11.6 0.065 0.069 1.828 1.866
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.737 1.777
jit_kernel_multiply 9 16.4 0.644 1.735 0.644 1.735
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.692 1.732
make_images_data 4572 15.5 0.047 0.053 1.237 1.611
prepare_preconditioner 11 7.9 0.000 0.000 1.505 1.531
make_preconditioner 11 8.9 0.000 0.002 1.505 1.531
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.404 1.460
hybrid_alltoall_any 4725 16.4 0.053 0.114 1.102 1.456
wfi_extrapolate 11 7.9 0.001 0.001 1.449 1.449
fft_wrap_pw1pw2_140 487 13.2 0.127 0.131 1.392 1.433
potential_pw2rs 119 12.3 0.006 0.006 1.400 1.406
fft3d_ps 1201 14.6 0.521 0.574 1.368 1.399
grid_collocate_task_list 119 9.7 1.290 1.371 1.290 1.371
mp_alltoall_d11v 2130 13.8 1.202 1.348 1.202 1.348
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.255 1.264
qs_ot_get_orbitals 108 10.5 0.000 0.001 1.215 1.261
transfer_rs2pw 487 10.6 0.005 0.005 1.053 1.148
mp_sum_d 4137 12.0 0.559 1.044 0.559 1.044
mp_allgather_i34 2286 14.5 0.582 0.969 0.582 0.969
qs_energies_init_hamiltonians 11 5.9 0.001 0.003 0.939 0.939
acc_transpose_blocks_sync 82296 16.5 0.816 0.938 0.816 0.938
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.871 0.885
transfer_pw2rs 487 13.2 0.004 0.005 0.870 0.873
acc_transpose_blocks_kernels 27432 16.5 0.187 0.274 0.652 0.857
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="102", plot="h2o_64_md", label="(8n/6r/2t)", y=38.709000, yerr=0.000000
PlotPoint: name="103", plot="h2o_64_md_mem", label="(8n/6r/2t)", y=465.363636, yerr=1.666391
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 520.892416E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 950976
MPI messages size (bytes):
total size 203.844256E+09
min size 0.000000E+00
max size 1.638400E+06
average size 214.352688E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 253512 2076770304
8192 < size <= 32768 179424 2939682816
32768 < size <= 131072 181440 14863564800
131072 < size <= 4194304 330176 183964913216
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63490.
MP_Allreduce 10155 305.
MP_Sync 54
MP_Alltoall 1821 1607811.
MP_SendRecv 11067 57667.
MP_ISendRecv 11067 57667.
MP_Wait 21987
MP_ISend 9880 92618.
MP_IRecv 9880 92618.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.032 32.529 32.531
qs_mol_dyn_low 1 2.0 0.003 0.004 32.218 32.225
qs_forces 11 3.9 0.003 0.007 32.158 32.158
qs_energies 11 4.9 0.005 0.031 30.610 30.612
scf_env_do_scf 11 5.9 0.003 0.023 25.828 25.832
scf_env_do_scf_inner_loop 108 6.5 0.002 0.007 23.253 23.254
dbcsr_multiply_generic 2286 12.5 0.096 0.098 16.927 17.031
qs_scf_new_mos 108 7.5 0.001 0.001 15.561 15.572
qs_scf_loop_do_ot 108 8.5 0.001 0.001 15.561 15.571
velocity_verlet 10 3.0 0.001 0.001 15.219 15.221
ot_scf_mini 108 9.5 0.002 0.003 14.797 14.812
multiply_cannon 2286 13.5 0.194 0.198 13.686 14.317
multiply_cannon_loop 2286 14.5 0.869 0.911 12.924 13.664
ot_mini 108 10.5 0.001 0.001 9.112 9.128
qs_ot_get_derivative 108 11.5 0.001 0.001 7.622 7.636
multiply_cannon_multrec 18288 15.5 1.881 2.874 6.949 7.224
rebuild_ks_matrix 119 8.3 0.000 0.000 6.021 6.035
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.017 6.020 6.035
dbcsr_mm_accdrv_process 38222 16.0 4.908 5.688 4.974 5.753
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.318 5.331
mp_waitall_1 158411 16.6 2.684 3.835 2.684 3.835
sum_up_and_integrate 119 10.3 0.003 0.016 3.563 3.576
integrate_v_rspace 119 11.3 0.003 0.003 3.548 3.555
init_scf_run 11 5.9 0.000 0.001 3.538 3.538
scf_env_initial_rho_setup 11 6.9 0.001 0.002 3.538 3.538
qs_ot_get_p 119 10.4 0.001 0.001 3.443 3.468
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.905 3.446
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.346 3.356
calculate_rho_elec 119 8.7 0.031 0.031 3.346 3.355
init_scf_loop 11 6.9 0.001 0.005 2.556 2.562
multiply_cannon_metrocomm3 18288 15.5 0.048 0.050 1.572 2.543
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.952 2.360
apply_single 119 13.6 0.000 0.000 1.951 2.359
qs_ot_p2m_diag 50 11.0 0.012 0.013 2.289 2.295
calculate_first_density_matrix 1 7.0 0.007 0.055 2.247 2.291
cp_dbcsr_syevd 50 12.0 0.003 0.005 1.986 1.986
density_rs2pw 119 9.7 0.004 0.004 1.873 1.949
make_m2s 4572 13.5 0.044 0.045 1.819 1.946
grid_integrate_task_list 119 12.3 1.802 1.906 1.802 1.906
pw_transfer 1439 11.6 0.065 0.068 1.875 1.887
make_images 4572 14.5 0.191 0.204 1.732 1.858
calculate_dm_sparse 119 9.5 0.000 0.001 1.845 1.855
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.783 1.795
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.739 1.752
acc_transpose_blocks 18288 15.5 0.079 0.081 1.637 1.743
cp_fm_diag_elpa 50 13.0 0.000 0.001 1.728 1.740
cp_fm_diag_elpa_base 50 14.0 1.702 1.718 1.724 1.737
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.706 1.707
prepare_preconditioner 11 7.9 0.000 0.000 1.693 1.695
make_preconditioner 11 8.9 0.000 0.001 1.693 1.695
make_full_inverse_cholesky 11 9.9 0.000 0.001 1.551 1.635
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.511 1.516
ot_diis_step 108 11.5 0.011 0.012 1.471 1.471
mp_sum_l 7287 12.8 1.040 1.463 1.040 1.463
fft_wrap_pw1pw2_140 487 13.2 0.178 0.182 1.427 1.438
potential_pw2rs 119 12.3 0.007 0.008 1.364 1.370
fft3d_ps 1201 14.6 0.547 0.566 1.338 1.349
grid_collocate_task_list 119 9.7 1.236 1.300 1.236 1.300
multiply_cannon_sync_h2d 18288 15.5 1.025 1.208 1.025 1.208
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.177 1.183
wfi_extrapolate 11 7.9 0.001 0.001 1.172 1.172
transfer_rs2pw 487 10.6 0.005 0.005 1.047 1.153
qs_energies_init_hamiltonians 11 5.9 0.002 0.005 1.008 1.009
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.957 0.974
make_images_data 4572 15.5 0.047 0.051 0.775 0.919
acc_transpose_blocks_sync 54864 16.5 0.747 0.846 0.747 0.846
hybrid_alltoall_any 4725 16.4 0.058 0.116 0.665 0.828
transfer_pw2rs 487 13.2 0.004 0.004 0.822 0.827
mp_alltoall_d11v 2130 13.8 0.711 0.825 0.711 0.825
acc_transpose_blocks_kernels 18288 16.5 0.217 0.222 0.792 0.799
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.777 0.779
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.671 0.734
cp_fm_cholesky_invert 11 10.9 0.715 0.719 0.715 0.719
mp_alltoall_z22v 1201 16.6 0.639 0.709 0.639 0.709
mp_waitany 9880 13.7 0.575 0.695 0.575 0.695
transfer_rs2pw_140 130 11.5 0.115 0.118 0.564 0.660
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="104", plot="h2o_64_md", label="(8n/4r/3t)", y=32.531000, yerr=0.000000
PlotPoint: name="105", plot="h2o_64_md_mem", label="(8n/4r/3t)", y=496.181818, yerr=0.833196
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 557.334528E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1042416
MPI messages size (bytes):
total size 150.443262E+09
min size 0.000000E+00
max size 1.188816E+06
average size 144.321719E+03
MPI breakdown and total messages size (bytes):
size <= 128 228256 0
128 < size <= 8192 126888 1039466496
8192 < size <= 32768 191472 3137077248
32768 < size <= 131072 295800 25899827200
131072 < size <= 4194304 200000 120367247040
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63489.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 2412273.
MP_SendRecv 8211 74133.
MP_ISendRecv 8211 74133.
MP_Wait 16271
MP_ISend 7280 135929.
MP_IRecv 7280 135929.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.028 36.159 36.161
qs_mol_dyn_low 1 2.0 0.003 0.004 35.990 35.999
qs_forces 11 3.9 0.002 0.002 35.928 35.928
qs_energies 11 4.9 0.001 0.001 34.227 34.233
scf_env_do_scf 11 5.9 0.000 0.001 29.296 29.298
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 25.944 25.944
dbcsr_multiply_generic 2286 12.5 0.102 0.104 19.828 19.948
velocity_verlet 10 3.0 0.001 0.002 18.407 18.409
qs_scf_new_mos 108 7.5 0.001 0.001 17.875 17.932
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.874 17.931
ot_scf_mini 108 9.5 0.002 0.003 16.842 16.891
multiply_cannon 2286 13.5 0.218 0.225 16.118 16.413
multiply_cannon_loop 2286 14.5 1.537 1.611 15.161 15.576
ot_mini 108 10.5 0.001 0.001 10.343 10.401
multiply_cannon_multrec 27432 15.5 2.450 3.108 9.013 9.282
qs_ot_get_derivative 108 11.5 0.001 0.001 8.476 8.526
dbcsr_mm_accdrv_process 47916 15.9 6.053 7.609 6.461 7.809
rebuild_ks_matrix 119 8.3 0.000 0.000 6.304 6.355
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.012 6.304 6.354
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.586 5.631
qs_ot_get_p 119 10.4 0.001 0.001 3.639 3.711
init_scf_run 11 5.9 0.000 0.001 3.607 3.608
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.607 3.607
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 3.148 3.553
sum_up_and_integrate 119 10.3 0.001 0.001 3.475 3.483
integrate_v_rspace 119 11.3 0.002 0.003 3.464 3.471
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.361 3.404
calculate_rho_elec 119 8.7 0.040 0.046 3.360 3.403
init_scf_loop 11 6.9 0.000 0.000 3.334 3.334
acc_transpose_blocks 27432 15.5 0.119 0.122 2.420 2.565
prepare_preconditioner 11 7.9 0.000 0.000 2.507 2.516
make_preconditioner 11 8.9 0.000 0.000 2.507 2.516
make_full_inverse_cholesky 11 9.9 0.000 0.000 2.117 2.442
make_m2s 4572 13.5 0.054 0.056 2.259 2.386
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.347 2.374
make_images 4572 14.5 0.272 0.336 2.150 2.275
mp_waitall_1 137007 16.6 1.720 2.262 1.720 2.262
qs_ot_p2m_diag 50 11.0 0.015 0.023 2.225 2.239
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.102 2.227
apply_single 119 13.6 0.000 0.000 2.102 2.227
calculate_first_density_matrix 1 7.0 0.000 0.000 2.223 2.227
calculate_dm_sparse 119 9.5 0.000 0.000 2.140 2.189
pw_transfer 1439 11.6 0.065 0.068 1.991 2.025
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 1.899 1.936
density_rs2pw 119 9.7 0.003 0.004 1.816 1.921
grid_integrate_task_list 119 12.3 1.828 1.898 1.828 1.898
ot_diis_step 108 11.5 0.012 0.012 1.826 1.826
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.819 1.820
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.803 1.804
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.637 1.647
fft_wrap_pw1pw2_140 487 13.2 0.227 0.233 1.569 1.607
acc_transpose_blocks_sync 82296 16.5 1.449 1.589 1.449 1.589
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.580 1.588
cp_fm_diag_elpa_base 50 14.0 1.545 1.561 1.578 1.586
fft3d_ps 1201 14.6 0.592 0.653 1.387 1.412
grid_collocate_task_list 119 9.7 1.248 1.341 1.248 1.341
wfi_extrapolate 11 7.9 0.001 0.001 1.330 1.330
multiply_cannon_metrocomm3 27432 15.5 0.040 0.041 0.757 1.316
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.297 1.312
mp_sum_l 7287 12.8 1.020 1.293 1.020 1.293
potential_pw2rs 119 12.3 0.008 0.009 1.283 1.285
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.235 1.246
cp_fm_upper_to_full 72 14.2 0.819 1.164 0.819 1.164
jit_kernel_multiply 6 16.6 0.340 1.072 0.340 1.072
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.068 1.069
dbcsr_complete_redistribute 329 12.2 0.118 0.145 0.770 1.052
transfer_rs2pw 487 10.6 0.004 0.005 0.884 0.991
make_images_data 4572 15.5 0.048 0.053 0.830 0.959
hybrid_alltoall_any 4725 16.4 0.066 0.156 0.709 0.905
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.798 0.876
mp_alltoall_d11v 2130 13.8 0.724 0.833 0.724 0.833
copy_fm_to_dbcsr 176 11.2 0.001 0.001 0.561 0.831
acc_transpose_blocks_kernels 27432 16.5 0.269 0.276 0.823 0.830
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.806 0.811
mp_alltoall_i22 627 13.8 0.436 0.734 0.436 0.734
cp_fm_cholesky_invert 11 10.9 0.727 0.730 0.727 0.730
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="106", plot="h2o_64_md", label="(8n/3r/4t)", y=36.161000, yerr=0.000000
PlotPoint: name="107", plot="h2o_64_md_mem", label="(8n/3r/4t)", y=529.181818, yerr=3.270201
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 617.091072E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 219456
MPI messages size (bytes):
total size 97.042514E+09
min size 0.000000E+00
max size 3.276800E+06
average size 442.195750E+03
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 101892 3336634368
32768 < size <= 131072 0 0
131072 < size <= 4194304 116112 93705670464
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8156 20.
MP_Alltoall 8655 64935.
MP_ISend 36532 168375.
MP_IRecv 36532 168349.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63488.
MP_Allreduce 10154 346.
MP_Sync 54
MP_Alltoall 1582 3682667.
MP_SendRecv 5355 94533.
MP_ISendRecv 5355 94533.
MP_Wait 11335
MP_ISend 5200 225425.
MP_IRecv 5200 225425.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.034 28.630 28.631
qs_mol_dyn_low 1 2.0 0.003 0.004 28.387 28.409
qs_forces 11 3.9 0.002 0.003 28.312 28.313
qs_energies 11 4.9 0.001 0.001 26.612 26.615
scf_env_do_scf 11 5.9 0.000 0.001 21.957 21.957
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 19.510 19.510
velocity_verlet 10 3.0 0.010 0.012 14.653 14.657
dbcsr_multiply_generic 2286 12.5 0.094 0.097 13.122 13.253
qs_scf_new_mos 108 7.5 0.001 0.001 11.866 11.893
qs_scf_loop_do_ot 108 8.5 0.001 0.001 11.866 11.892
ot_scf_mini 108 9.5 0.002 0.002 11.162 11.189
multiply_cannon 2286 13.5 0.223 0.228 10.539 11.124
multiply_cannon_loop 2286 14.5 0.642 0.660 9.618 9.856
ot_mini 108 10.5 0.001 0.001 6.528 6.562
multiply_cannon_multrec 9144 15.5 1.677 1.924 5.943 6.235
rebuild_ks_matrix 119 8.3 0.000 0.000 5.778 5.798
qs_ks_build_kohn_sham_matrix 119 9.3 0.012 0.013 5.778 5.798
qs_ot_get_derivative 108 11.5 0.001 0.001 5.225 5.252
qs_ks_update_qs_env 119 7.6 0.001 0.001 5.140 5.159
dbcsr_mm_accdrv_process 12550 15.8 3.625 4.191 4.157 4.232
sum_up_and_integrate 119 10.3 0.001 0.001 3.422 3.426
integrate_v_rspace 119 11.3 0.003 0.003 3.412 3.416
qs_rho_update_rho_low 119 7.7 0.001 0.001 3.339 3.345
calculate_rho_elec 119 8.7 0.060 0.061 3.339 3.344
init_scf_run 11 5.9 0.000 0.001 3.219 3.219
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.219 3.219
qs_ot_get_p 119 10.4 0.001 0.001 2.747 2.777
init_scf_loop 11 6.9 0.000 0.000 2.427 2.428
pw_transfer 1439 11.6 0.065 0.068 2.108 2.116
calculate_first_density_matrix 1 7.0 0.000 0.000 2.063 2.064
mp_waitall_1 115863 16.7 1.501 2.058 1.501 2.058
fft_wrap_pw1pw2 1201 12.6 0.008 0.008 2.014 2.025
make_m2s 4572 13.5 0.033 0.034 1.793 1.951
grid_integrate_task_list 119 12.3 1.873 1.936 1.873 1.936
make_images 4572 14.5 0.268 0.298 1.703 1.860
density_rs2pw 119 9.7 0.003 0.003 1.756 1.816
qs_ot_p2m_diag 50 11.0 0.022 0.023 1.779 1.781
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 1.724 1.749
calculate_dm_sparse 119 9.5 0.000 0.000 1.731 1.748
prepare_preconditioner 11 7.9 0.000 0.000 1.694 1.699
make_preconditioner 11 8.9 0.000 0.000 1.694 1.699
fft_wrap_pw1pw2_140 487 13.2 0.324 0.332 1.669 1.681
make_full_inverse_cholesky 11 9.9 0.000 0.000 1.585 1.610
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.539 1.539
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.537 1.538
acc_transpose_blocks 9144 15.5 0.043 0.044 1.425 1.475
fft3d_ps 1201 14.6 0.649 0.662 1.368 1.377
grid_collocate_task_list 119 9.7 1.294 1.374 1.294 1.374
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.362 1.373
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 1.337 1.351
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.308 1.316
cp_fm_diag_elpa_base 50 14.0 1.281 1.297 1.306 1.314
ot_diis_step 108 11.5 0.013 0.013 1.289 1.289
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 1.238 1.271
apply_single 119 13.6 0.000 0.000 1.238 1.271
potential_pw2rs 119 12.3 0.010 0.011 1.235 1.238
qs_energies_init_hamiltonians 11 5.9 0.002 0.002 1.226 1.227
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.123 1.127
wfi_extrapolate 11 7.9 0.001 0.001 1.105 1.105
jit_kernel_multiply 6 15.5 0.492 1.027 0.492 1.027
make_images_data 4572 15.5 0.042 0.046 0.784 0.948
hybrid_alltoall_any 4725 16.4 0.065 0.177 0.720 0.940
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.000 0.873 0.923
multiply_cannon_metrocomm3 9144 15.5 0.020 0.020 0.458 0.872
cp_fm_cholesky_invert 11 10.9 0.849 0.852 0.849 0.852
transfer_rs2pw 487 10.6 0.004 0.004 0.768 0.838
qs_ot_get_orbitals 108 10.5 0.000 0.000 0.814 0.820
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.757 0.759
mp_alltoall_d11v 2130 13.8 0.693 0.751 0.693 0.751
acc_transpose_blocks_sync 27432 16.5 0.716 0.740 0.716 0.740
qs_env_update_s_mstruct 11 6.9 0.001 0.001 0.675 0.728
acc_transpose_blocks_kernels 9144 16.5 0.117 0.120 0.651 0.675
mp_allgather_i34 2286 14.5 0.237 0.673 0.237 0.673
transfer_pw2rs 487 13.2 0.003 0.004 0.634 0.636
mp_alltoall_z22v 1201 16.6 0.593 0.629 0.593 0.629
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="108", plot="h2o_64_md", label="(8n/2r/6t)", y=28.631000, yerr=0.000000
PlotPoint: name="109", plot="h2o_64_md_mem", label="(8n/2r/6t)", y=586.636364, yerr=4.051324
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 804.106240E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 91440
MPI messages size (bytes):
total size 85.748679E+09
min size 0.000000E+00
max size 6.553600E+06
average size 937.758938E+03
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 21148 692256768
32768 < size <= 131072 19224 1259864064
131072 < size <= 4194304 41040 21941452800
4194304 < size <= 16777216 9456 61855174464
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3622 63723.
MP_Allreduce 10154 429.
MP_Sync 54
MP_Alltoall 1582 7383731.
MP_SendRecv 2499 189067.
MP_ISendRecv 2499 189067.
MP_Wait 6399
MP_ISend 3120 546875.
MP_IRecv 3120 546875.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.030 42.682 42.683
qs_mol_dyn_low 1 2.0 0.003 0.004 42.478 42.487
qs_forces 11 3.9 0.002 0.002 42.416 42.418
qs_energies 11 4.9 0.001 0.001 40.403 40.408
scf_env_do_scf 11 5.9 0.001 0.001 34.653 34.653
scf_env_do_scf_inner_loop 108 6.5 0.003 0.007 26.842 26.843
velocity_verlet 10 3.0 0.002 0.002 24.133 24.139
dbcsr_multiply_generic 2286 12.5 0.102 0.104 18.893 19.065
qs_scf_new_mos 108 7.5 0.001 0.001 17.116 17.214
qs_scf_loop_do_ot 108 8.5 0.001 0.001 17.115 17.213
ot_scf_mini 108 9.5 0.002 0.002 15.954 16.054
multiply_cannon 2286 13.5 0.299 0.306 15.000 15.873
multiply_cannon_loop 2286 14.5 0.845 0.866 13.729 14.642
ot_mini 108 10.5 0.001 0.001 9.706 9.820
multiply_cannon_multrec 9144 15.5 3.400 4.684 8.776 8.892
init_scf_loop 11 6.9 0.000 0.000 7.785 7.789
qs_ot_get_derivative 108 11.5 0.001 0.001 7.640 7.739
rebuild_ks_matrix 119 8.3 0.000 0.000 7.288 7.430
qs_ks_build_kohn_sham_matrix 119 9.3 0.013 0.013 7.288 7.429
prepare_preconditioner 11 7.9 0.000 0.000 6.796 6.809
make_preconditioner 11 8.9 0.000 0.000 6.796 6.809
qs_ks_update_qs_env 119 7.6 0.001 0.001 6.567 6.694
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.374 6.689
dbcsr_mm_accdrv_process 12550 15.8 4.515 6.470 5.241 6.564
cp_fm_upper_to_full 72 14.2 3.160 4.526 3.160 4.526
qs_rho_update_rho_low 119 7.7 0.001 0.001 4.228 4.233
calculate_rho_elec 119 8.7 0.117 0.120 4.227 4.233
sum_up_and_integrate 119 10.3 0.001 0.001 3.858 3.864
integrate_v_rspace 119 11.3 0.003 0.003 3.847 3.854
init_scf_run 11 5.9 0.000 0.001 3.675 3.675
scf_env_initial_rho_setup 11 6.9 0.001 0.001 3.675 3.675
qs_ot_get_p 119 10.4 0.001 0.001 3.333 3.465
mp_waitall_1 94719 16.7 2.299 3.284 2.299 3.284
qs_ot_get_derivative_taylor 59 13.0 0.001 0.001 2.821 3.215
pw_transfer 1439 11.6 0.068 0.069 3.035 3.043
fft_wrap_pw1pw2 1201 12.6 0.009 0.009 2.937 2.945
dbcsr_complete_redistribute 329 12.2 0.282 0.286 1.957 2.761
make_m2s 4572 13.5 0.037 0.038 2.369 2.540
apply_preconditioner_dbcsr 119 12.6 0.000 0.000 2.214 2.528
apply_single 119 13.6 0.000 0.000 2.214 2.528
fft_wrap_pw1pw2_140 487 13.2 0.649 0.653 2.505 2.514
copy_fm_to_dbcsr 176 11.2 0.001 0.001 1.653 2.461
make_images 4572 14.5 0.352 0.384 2.248 2.420
density_rs2pw 119 9.7 0.003 0.003 2.250 2.271
calculate_dm_sparse 119 9.5 0.000 0.000 2.234 2.248
mp_alltoall_i22 627 13.8 1.450 2.240 1.450 2.240
multiply_cannon_metrocomm3 9144 15.5 0.021 0.021 1.346 2.225
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.417 2.218
calculate_first_density_matrix 1 7.0 0.000 0.000 2.204 2.206
grid_integrate_task_list 119 12.3 2.085 2.133 2.085 2.133
qs_ot_get_derivative_diag 49 12.0 0.001 0.001 2.038 2.089
ot_diis_step 108 11.5 0.014 0.014 2.041 2.041
qs_ot_p2m_diag 50 11.0 0.043 0.044 1.959 1.961
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 1.935 1.936
fft3d_ps 1201 14.6 0.868 0.888 1.858 1.864
acc_transpose_blocks 9144 15.5 0.044 0.045 1.787 1.811
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.776 1.778
mp_sum_l 7287 12.8 0.977 1.694 0.977 1.694
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 1.630 1.651
cp_dbcsr_syevd 50 12.0 0.003 0.003 1.606 1.607
grid_collocate_task_list 119 9.7 1.525 1.554 1.525 1.554
cp_fm_cholesky_invert 11 10.9 1.443 1.446 1.443 1.446
wfi_extrapolate 11 7.9 0.001 0.001 1.408 1.408
potential_pw2rs 119 12.3 0.014 0.014 1.357 1.359
hybrid_alltoall_any 4725 16.4 0.091 0.150 1.086 1.347
cp_fm_diag_elpa 50 13.0 0.000 0.000 1.320 1.320
cp_fm_diag_elpa_base 50 14.0 1.176 1.225 1.318 1.318
qs_ot_get_orbitals 108 10.5 0.000 0.000 1.300 1.317
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.000 0.000 1.240 1.245
make_images_data 4572 15.5 0.046 0.049 1.027 1.244
mp_alltoall_d11v 2130 13.8 1.182 1.218 1.182 1.218
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.108 1.126
acc_transpose_blocks_sync 27432 16.5 1.075 1.101 1.075 1.101
build_core_hamiltonian_matrix_ 11 4.9 0.000 0.001 0.993 1.026
jit_kernel_multiply 6 15.5 0.696 1.014 0.696 1.014
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 0.931 0.944
qs_create_task_list 11 7.9 0.001 0.001 0.927 0.938
generate_qs_task_list 11 8.9 0.368 0.387 0.926 0.938
mp_alltoall_z22v 1201 16.6 0.854 0.883 0.854 0.883
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="110", plot="h2o_64_md", label="(8n/1r/12t)", y=42.683000, yerr=0.000000
PlotPoint: name="111", plot="h2o_64_md_mem", label="(8n/1r/12t)", y=743.454545, yerr=20.951735
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/09/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 198.287135E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 8410880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 117.0
marketing flops 15.646302E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 500.514816E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 8483040
MPI messages size (bytes):
total size 1.160510E+12
min size 0.000000E+00
max size 1.161504E+06
average size 136.803609E+03
MPI breakdown and total messages size (bytes):
size <= 128 1836752 0
128 < size <= 8192 1040592 8524529664
8192 < size <= 32768 1486976 24362614784
32768 < size <= 131072 2491776 216971345920
131072 < size <= 4194304 1626944 910632720448
4194304 < size <= 16777216 0 0
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66212.
MP_Allreduce 9776 488.
MP_Sync 52
MP_Alltoall 1938 1384653.
MP_SendRecv 20900 9096.
MP_ISendRecv 20900 9096.
MP_Wait 37268
MP_ISend 14300 82312.
MP_IRecv 14300 82312.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.030 83.550 83.552
qs_mol_dyn_low 1 2.0 0.003 0.004 83.282 83.312
qs_forces 11 3.9 0.003 0.003 83.214 83.215
qs_energies 11 4.9 0.001 0.002 80.305 80.325
scf_env_do_scf 11 5.9 0.000 0.001 71.365 71.368
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 65.666 65.667
dbcsr_multiply_generic 2055 12.4 0.105 0.108 51.342 51.537
qs_scf_new_mos 99 7.5 0.000 0.001 48.153 48.292
qs_scf_loop_do_ot 99 8.5 0.001 0.001 48.152 48.292
ot_scf_mini 99 9.5 0.002 0.002 45.765 45.860
velocity_verlet 10 3.0 0.001 0.002 43.697 43.698
multiply_cannon 2055 13.4 0.184 0.190 42.562 43.401
multiply_cannon_loop 2055 14.4 1.789 1.834 41.471 42.340
ot_mini 99 10.5 0.001 0.001 26.913 27.005
qs_ot_get_derivative 99 11.5 0.002 0.009 20.061 20.144
multiply_cannon_multrec 49320 15.4 11.311 12.102 17.399 18.178
rebuild_ks_matrix 110 8.3 0.000 0.000 14.782 14.917
qs_ks_build_kohn_sham_matrix 110 9.3 0.011 0.012 14.782 14.917
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.966 13.086
mp_waitall_1 220248 16.4 11.020 11.897 11.020 11.897
qs_ot_get_p 110 10.4 0.001 0.001 10.320 10.419
multiply_cannon_sync_h2d 49320 15.4 9.536 10.096 9.536 10.096
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 7.590 8.155
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 7.265 7.706
apply_single 110 13.6 0.000 0.000 7.265 7.706
multiply_cannon_metrocomm3 49320 15.4 0.081 0.084 6.438 7.528
sum_up_and_integrate 110 10.3 0.001 0.003 7.298 7.313
integrate_v_rspace 110 11.3 0.002 0.003 7.273 7.294
qs_ot_p2m_diag 48 11.0 0.012 0.018 7.023 7.065
init_scf_run 11 5.9 0.000 0.001 6.862 6.863
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.862 6.862
qs_rho_update_rho_low 110 7.6 0.000 0.001 6.563 6.699
calculate_rho_elec 110 8.6 0.020 0.024 6.563 6.699
ot_diis_step 99 11.5 0.006 0.006 6.630 6.630
dbcsr_mm_accdrv_process 87628 16.1 3.091 3.189 5.957 6.239
cp_dbcsr_syevd 48 12.0 0.003 0.003 6.101 6.102
init_scf_loop 11 6.9 0.000 0.000 5.665 5.666
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 5.518 5.575
cp_fm_diag_elpa 48 13.0 0.000 0.000 5.473 5.498
cp_fm_diag_elpa_base 48 14.0 5.457 5.485 5.471 5.496
mp_sum_l 6594 12.7 4.185 4.871 4.185 4.871
make_m2s 4110 13.4 0.061 0.065 4.136 4.254
make_images 4110 14.4 0.177 0.190 4.040 4.159
wfi_extrapolate 11 7.9 0.001 0.001 4.101 4.101
calculate_dm_sparse 110 9.5 0.001 0.001 3.756 3.860
density_rs2pw 110 9.6 0.004 0.004 3.601 3.824
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.590 3.594
multiply_cannon_metrocomm1 49320 15.4 0.063 0.066 2.439 3.586
prepare_preconditioner 11 7.9 0.000 0.000 3.491 3.510
make_preconditioner 11 8.9 0.000 0.000 3.491 3.510
grid_integrate_task_list 110 12.3 3.267 3.419 3.267 3.419
make_full_inverse_cholesky 11 9.9 0.000 0.000 3.274 3.322
pw_transfer 1331 11.6 0.055 0.067 3.210 3.309
qs_ot_get_orbitals 99 10.5 0.001 0.001 3.241 3.295
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 3.203 3.236
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.121 3.224
potential_pw2rs 110 12.3 0.005 0.006 2.863 2.910
fft_wrap_pw1pw2_140 451 13.1 0.309 0.332 2.621 2.730
calculate_first_density_matrix 1 7.0 0.000 0.001 2.657 2.662
acc_transpose_blocks 49320 15.4 0.211 0.223 2.563 2.636
jit_kernel_multiply 13 15.9 2.580 2.615 2.580 2.615
mp_alltoall_d11v 2046 13.8 2.140 2.550 2.140 2.550
fft3d_ps 1111 14.6 0.800 0.890 2.416 2.504
transfer_rs2pw 451 10.6 0.005 0.006 2.082 2.405
grid_collocate_task_list 110 9.6 2.157 2.283 2.157 2.283
mp_waitany 14300 13.8 1.839 2.269 1.839 2.269
transfer_pw2rs 451 13.1 0.006 0.007 1.982 2.017
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.946 1.979
make_images_data 4110 15.4 0.043 0.047 1.833 1.976
cp_fm_cholesky_invert 11 10.9 1.899 1.904 1.899 1.904
hybrid_alltoall_any 4261 16.3 0.083 0.480 1.587 1.836
mp_sum_d 3891 11.9 1.384 1.823 1.384 1.823
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.655 1.693
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="200", plot="h2o_128_md", label="(8n/12r/1t)", y=83.552000, yerr=0.000000
PlotPoint: name="201", plot="h2o_128_md_mem", label="(8n/12r/1t)", y=475.909091, yerr=2.108621
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 594.669568E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1972800
MPI messages size (bytes):
total size 1.077520E+12
min size 0.000000E+00
max size 4.537280E+06
average size 546.188250E+03
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 222984 1826684928
8192 < size <= 32768 520356 13399818240
32768 < size <= 131072 372336 35386294272
131072 < size <= 4194304 787758 788321309808
4194304 < size <= 16777216 54450 238588003280
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66430.
MP_Allreduce 9775 566.
MP_Sync 52
MP_Alltoall 1717 3558822.
MP_SendRecv 10340 26400.
MP_ISendRecv 10340 26400.
MP_Wait 22352
MP_ISend 10164 155761.
MP_IRecv 10164 155761.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.009 0.029 68.906 68.909
qs_mol_dyn_low 1 2.0 0.003 0.004 68.682 68.692
qs_forces 11 3.9 0.003 0.003 68.614 68.615
qs_energies 11 4.9 0.001 0.002 65.304 65.307
scf_env_do_scf 11 5.9 0.000 0.001 56.813 56.816
scf_env_do_scf_inner_loop 99 6.5 0.002 0.007 49.225 49.226
dbcsr_multiply_generic 2055 12.4 0.114 0.118 37.920 38.103
velocity_verlet 10 3.0 0.001 0.002 36.050 36.052
qs_scf_new_mos 99 7.5 0.001 0.001 33.357 33.481
qs_scf_loop_do_ot 99 8.5 0.001 0.001 33.356 33.480
multiply_cannon 2055 13.4 0.225 0.243 31.224 32.250
ot_scf_mini 99 9.5 0.003 0.003 31.682 31.831
multiply_cannon_loop 2055 14.4 1.166 1.193 29.936 30.741
ot_mini 99 10.5 0.001 0.001 18.409 18.560
multiply_cannon_multrec 24660 15.4 6.970 8.755 14.029 15.745
rebuild_ks_matrix 110 8.3 0.000 0.001 13.535 13.633
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.017 13.535 13.632
qs_ot_get_derivative 99 11.5 0.001 0.001 12.639 12.785
qs_ks_update_qs_env 110 7.6 0.001 0.001 11.914 12.011
mp_waitall_1 176588 16.5 7.456 9.898 7.456 9.898
multiply_cannon_metrocomm3 24660 15.4 0.074 0.076 5.060 8.012
init_scf_loop 11 6.9 0.000 0.000 7.553 7.553
multiply_cannon_sync_h2d 24660 15.4 6.350 7.502 6.350 7.502
dbcsr_mm_accdrv_process 52282 16.1 5.434 6.287 6.888 7.203
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 6.445 7.055
apply_single 110 13.6 0.000 0.001 6.445 7.055
qs_ot_get_p 110 10.4 0.001 0.001 6.629 6.823
sum_up_and_integrate 110 10.3 0.002 0.003 6.387 6.400
integrate_v_rspace 110 11.3 0.002 0.003 6.360 6.372
init_scf_run 11 5.9 0.000 0.001 6.126 6.127
scf_env_initial_rho_setup 11 6.9 0.001 0.001 6.126 6.126
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.906 5.915
calculate_rho_elec 110 8.6 0.039 0.047 5.905 5.915
ot_diis_step 99 11.5 0.010 0.010 5.722 5.723
prepare_preconditioner 11 7.9 0.000 0.000 5.537 5.561
make_preconditioner 11 8.9 0.000 0.000 5.537 5.561
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 4.744 5.484
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.145 5.294
make_m2s 4110 13.4 0.057 0.059 4.284 4.745
qs_ot_p2m_diag 48 11.0 0.029 0.044 4.680 4.700
make_images 4110 14.4 0.404 0.462 4.171 4.630
cp_dbcsr_syevd 48 12.0 0.003 0.003 4.212 4.212
pw_transfer 1331 11.6 0.066 0.074 3.582 3.729
cp_fm_diag_elpa 48 13.0 0.000 0.000 3.633 3.647
cp_fm_diag_elpa_base 48 14.0 3.584 3.600 3.630 3.644
fft_wrap_pw1pw2 1111 12.6 0.008 0.009 3.475 3.623
wfi_extrapolate 11 7.9 0.001 0.001 3.523 3.523
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.328 3.404
density_rs2pw 110 9.6 0.004 0.004 3.148 3.327
grid_integrate_task_list 110 12.3 3.163 3.320 3.163 3.320
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.002 3.230 3.231
fft_wrap_pw1pw2_140 451 13.1 0.358 0.379 2.959 3.109
calculate_dm_sparse 110 9.5 0.001 0.001 2.978 3.006
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.887 2.924
hybrid_alltoall_any 4261 16.3 0.105 0.451 2.012 2.797
make_images_data 4110 15.4 0.048 0.052 2.301 2.787
fft3d_ps 1111 14.6 1.116 1.345 2.568 2.730
cp_fm_cholesky_invert 11 10.9 2.600 2.607 2.600 2.607
calculate_first_density_matrix 1 7.0 0.000 0.000 2.500 2.503
mp_sum_l 6594 12.7 1.780 2.500 1.780 2.500
grid_collocate_task_list 110 9.6 2.170 2.331 2.170 2.331
potential_pw2rs 110 12.3 0.008 0.009 2.256 2.266
mp_alltoall_d11v 2046 13.8 1.755 2.000 1.755 2.000
acc_transpose_blocks 24660 15.4 0.115 0.119 1.966 1.995
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.972 1.991
jit_kernel_multiply 10 16.2 1.085 1.897 1.085 1.897
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 1.848 1.849
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.777 1.790
multiply_cannon_metrocomm4 22605 15.4 0.074 0.078 0.783 1.690
mp_allgather_i34 2055 14.4 0.598 1.628 0.598 1.628
transfer_rs2pw 451 10.6 0.006 0.007 1.425 1.623
mp_irecv_dv 57340 16.2 0.659 1.572 0.659 1.572
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.540 1.548
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.002 1.384 1.487
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="202", plot="h2o_128_md", label="(8n/6r/2t)", y=68.909000, yerr=0.000000
PlotPoint: name="203", plot="h2o_128_md_mem", label="(8n/6r/2t)", y=561.727273, yerr=7.312392
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/11/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 184415158272 0.0% 0.0% 100.0%
flops 9 x 9 x 32 269180485632 0.0% 0.0% 100.0%
flops 9 x 22 x 32 349395425280 0.0% 0.0% 100.0%
flops 22 x 9 x 32 350042406912 0.0% 0.0% 100.0%
flops 22 x 22 x 32 453581815808 0.0% 0.0% 100.0%
flops 32 x 32 x 9 465064427520 0.0% 0.0% 100.0%
flops 32 x 32 x 22 568412078080 0.0% 0.0% 100.0%
flops 9 x 32 x 32 572195340288 0.0% 0.0% 100.0%
flops 22 x 32 x 32 699349860352 0.0% 0.0% 100.0%
flops 9 x 32 x 9 1735942275072 0.0% 0.0% 100.0%
flops 22 x 32 x 9 2216407818240 0.0% 0.0% 100.0%
flops 9 x 32 x 22 2216407818240 0.0% 0.0% 100.0%
flops 22 x 32 x 22 2803661053952 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 12.884056E+12 0.0% 0.0% 100.0%
flops max/rank 404.681598E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 984178160 0.0% 0.0% 100.0%
number of processed stacks 3346752 0.0% 0.0% 100.0%
average stack size 0.0 0.0 294.1
marketing flops 15.646297E+12
-------------------------------------------------------------------------------
# multiplications 2055
max memory usage/rank 666.173440E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 854880
MPI messages size (bytes):
total size 708.322787E+09
min size 0.000000E+00
max size 6.553600E+06
average size 828.564000E+03
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 222984 7302414336
32768 < size <= 131072 153888 10085203968
131072 < size <= 4194304 389376 200257044480
4194304 < size <= 16777216 82208 490679162176
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66421.
MP_Allreduce 9774 562.
MP_Sync 52
MP_Alltoall 1496 4511006.
MP_SendRecv 6820 27424.
MP_ISendRecv 6820 27424.
MP_Wait 25498
MP_ISend 17072 115022.
MP_IRecv 17072 115022.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.032 59.226 59.227
qs_mol_dyn_low 1 2.0 0.003 0.004 58.959 58.972
qs_forces 11 3.9 0.003 0.003 58.888 58.889
qs_energies 11 4.9 0.001 0.001 55.743 55.746
scf_env_do_scf 11 5.9 0.000 0.001 47.925 47.925
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 39.513 39.513
velocity_verlet 10 3.0 0.001 0.002 32.190 32.192
dbcsr_multiply_generic 2055 12.4 0.109 0.112 28.523 28.741
qs_scf_new_mos 99 7.5 0.001 0.001 24.979 25.073
qs_scf_loop_do_ot 99 8.5 0.001 0.001 24.979 25.073
ot_scf_mini 99 9.5 0.002 0.003 23.733 23.842
multiply_cannon 2055 13.4 0.212 0.221 22.390 23.556
multiply_cannon_loop 2055 14.4 0.822 0.849 21.232 22.225
ot_mini 99 10.5 0.001 0.001 13.569 13.676
rebuild_ks_matrix 110 8.3 0.000 0.001 12.066 12.186
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.065 12.186
multiply_cannon_multrec 16440 15.4 3.701 4.744 9.925 10.905
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.616 10.724
mp_waitall_1 139946 16.5 6.639 9.778 6.639 9.778
qs_ot_get_derivative 99 11.5 0.005 0.013 9.186 9.293
init_scf_loop 11 6.9 0.000 0.000 8.380 8.380
multiply_cannon_metrocomm3 16440 15.4 0.045 0.047 4.146 7.133
prepare_preconditioner 11 7.9 0.000 0.000 6.642 6.657
make_preconditioner 11 8.9 0.000 0.000 6.642 6.657
make_full_inverse_cholesky 11 9.9 0.000 0.000 6.009 6.362
sum_up_and_integrate 110 10.3 0.001 0.002 6.246 6.259
integrate_v_rspace 110 11.3 0.003 0.003 6.220 6.234
dbcsr_mm_accdrv_process 34862 16.1 5.367 5.730 6.067 6.153
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.699 5.710
calculate_rho_elec 110 8.6 0.059 0.059 5.698 5.709
qs_ot_get_p 110 10.4 0.001 0.001 5.357 5.489
init_scf_run 11 5.9 0.000 0.001 5.441 5.441
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.440 5.440
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.902 5.312
apply_single 110 13.6 0.000 0.000 4.901 5.312
make_m2s 4110 13.4 0.050 0.051 4.106 4.485
make_images 4110 14.4 0.397 0.521 3.987 4.363
ot_diis_step 99 11.5 0.011 0.011 4.352 4.353
multiply_cannon_sync_h2d 16440 15.4 3.277 3.829 3.277 3.829
qs_ot_p2m_diag 48 11.0 0.042 0.044 3.757 3.761
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.124 3.751
pw_transfer 1331 11.6 0.066 0.073 3.581 3.592
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.474 3.487
grid_integrate_task_list 110 12.3 3.200 3.439 3.200 3.439
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.392 3.392
density_rs2pw 110 9.6 0.004 0.004 2.941 3.196
fft_wrap_pw1pw2_140 451 13.1 0.461 0.470 3.012 3.027
wfi_extrapolate 11 7.9 0.001 0.001 2.942 2.942
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.868 2.878
cp_fm_diag_elpa_base 48 14.0 2.802 2.832 2.866 2.877
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.776 2.777
make_images_data 4110 15.4 0.046 0.050 2.294 2.776
hybrid_alltoall_any 4261 16.3 0.108 0.383 2.077 2.764
cp_fm_cholesky_invert 11 10.9 2.600 2.606 2.600 2.606
calculate_dm_sparse 110 9.5 0.001 0.001 2.533 2.565
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.462 2.499
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.414 2.476
calculate_first_density_matrix 1 7.0 0.000 0.001 2.408 2.409
fft3d_ps 1111 14.6 1.119 1.131 2.391 2.401
grid_collocate_task_list 110 9.6 2.220 2.390 2.220 2.390
multiply_cannon_metrocomm4 14385 15.4 0.047 0.050 0.872 2.307
mp_alltoall_d11v 2046 13.8 1.734 2.201 1.734 2.201
mp_irecv_dv 48980 15.7 0.798 2.173 0.798 2.173
potential_pw2rs 110 12.3 0.010 0.011 2.091 2.096
mp_sum_l 6594 12.7 1.400 1.982 1.400 1.982
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 1.946 1.947
dbcsr_complete_redistribute 325 12.2 0.322 0.356 1.392 1.852
cp_fm_upper_to_full 70 14.2 1.396 1.757 1.396 1.757
acc_transpose_blocks 16440 15.4 0.079 0.082 1.594 1.679
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.640 1.653
cp_fm_cholesky_decompose 22 10.9 1.540 1.560 1.540 1.560
mp_allgather_i34 2055 14.4 0.473 1.517 0.473 1.517
transfer_rs2pw 451 10.6 0.005 0.006 1.249 1.496
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.471 1.483
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.359 1.451
mp_waitany 17072 13.8 1.114 1.379 1.114 1.379
copy_fm_to_dbcsr 174 11.2 0.001 0.001 0.919 1.370
rs_gather_matrices 110 12.3 0.137 0.150 0.877 1.350
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.268 1.276
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="204", plot="h2o_128_md", label="(8n/4r/3t)", y=59.227000, yerr=0.000000
PlotPoint: name="205", plot="h2o_128_md_mem", label="(8n/4r/3t)", y=630.000000, yerr=8.301150
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 740.179968E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 937080
MPI messages size (bytes):
total size 523.723932E+09
min size 0.000000E+00
max size 4.537280E+06
average size 558.889250E+03
MPI breakdown and total messages size (bytes):
size <= 128 6996 0
128 < size <= 8192 264 2162688
8192 < size <= 32768 304932 8165326848
32768 < size <= 131072 110640 6338641920
131072 < size <= 4194304 489498 400769458320
4194304 < size <= 16777216 24750 108449092400
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66419.
MP_Allreduce 9774 603.
MP_Sync 52
MP_Alltoall 1496 5863162.
MP_SendRecv 5060 43184.
MP_ISendRecv 5060 43184.
MP_Wait 20042
MP_ISend 13376 163145.
MP_IRecv 13376 163145.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.010 0.030 65.762 65.764
qs_mol_dyn_low 1 2.0 0.003 0.004 65.543 65.553
qs_forces 11 3.9 0.003 0.003 65.469 65.477
qs_energies 11 4.9 0.001 0.001 62.081 62.091
scf_env_do_scf 11 5.9 0.000 0.001 53.766 53.769
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 42.200 42.200
velocity_verlet 10 3.0 0.002 0.002 37.221 37.223
dbcsr_multiply_generic 2055 12.4 0.116 0.121 30.674 30.880
qs_scf_new_mos 99 7.5 0.001 0.001 27.498 27.621
qs_scf_loop_do_ot 99 8.5 0.001 0.001 27.497 27.620
ot_scf_mini 99 9.5 0.002 0.003 25.809 25.904
multiply_cannon 2055 13.4 0.237 0.254 23.428 24.689
multiply_cannon_loop 2055 14.4 1.412 1.451 22.063 22.720
ot_mini 99 10.5 0.001 0.001 14.760 14.880
multiply_cannon_multrec 24660 15.4 4.066 6.924 13.005 14.176
rebuild_ks_matrix 110 8.3 0.000 0.000 12.028 12.144
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 12.027 12.144
init_scf_loop 11 6.9 0.000 0.000 11.526 11.527
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.628 10.730
qs_ot_get_derivative 99 11.5 0.001 0.001 10.569 10.669
prepare_preconditioner 11 7.9 0.000 0.000 9.792 9.811
make_preconditioner 11 8.9 0.000 0.000 9.792 9.811
dbcsr_mm_accdrv_process 52304 16.0 7.756 9.159 8.782 9.780
make_full_inverse_cholesky 11 9.9 0.000 0.000 8.002 9.477
sum_up_and_integrate 110 10.3 0.001 0.002 6.261 6.274
integrate_v_rspace 110 11.3 0.003 0.003 6.235 6.247
mp_waitall_1 121746 16.5 4.303 6.172 4.303 6.172
qs_ot_get_p 110 10.4 0.001 0.001 5.897 6.043
make_m2s 4110 13.4 0.059 0.061 5.623 5.925
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.880 5.889
calculate_rho_elec 110 8.6 0.077 0.081 5.880 5.889
make_images 4110 14.4 0.571 0.695 5.481 5.778
init_scf_run 11 5.9 0.000 0.001 5.586 5.586
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.585 5.586
cp_fm_upper_to_full 70 14.2 3.367 4.883 3.367 4.883
ot_diis_step 99 11.5 0.011 0.011 4.136 4.137
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.045 4.123
apply_single 110 13.6 0.000 0.000 4.045 4.123
qs_ot_p2m_diag 48 11.0 0.055 0.064 4.023 4.039
pw_transfer 1331 11.6 0.065 0.072 3.797 3.830
dbcsr_complete_redistribute 325 12.2 0.413 0.467 2.649 3.789
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.691 3.726
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 3.485 3.533
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.459 3.459
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 3.357 3.412
grid_integrate_task_list 110 12.3 3.301 3.403 3.301 3.403
copy_fm_to_dbcsr 174 11.2 0.001 0.001 2.132 3.265
density_rs2pw 110 9.6 0.004 0.004 2.996 3.211
fft_wrap_pw1pw2_140 451 13.1 0.530 0.543 3.164 3.202
make_images_data 4110 15.4 0.048 0.052 2.821 3.166
hybrid_alltoall_any 4261 16.3 0.122 0.455 2.344 3.094
multiply_cannon_metrocomm3 24660 15.4 0.037 0.038 1.330 3.031
calculate_dm_sparse 110 9.5 0.001 0.001 2.990 3.025
wfi_extrapolate 11 7.9 0.001 0.001 3.010 3.010
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.919 2.928
cp_fm_diag_elpa_base 48 14.0 2.771 2.833 2.916 2.927
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 1.782 2.899
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.869 2.870
mp_alltoall_i22 605 13.7 1.666 2.856 1.666 2.856
cp_fm_cholesky_invert 11 10.9 2.598 2.606 2.598 2.606
multiply_cannon_sync_h2d 24660 15.4 2.376 2.548 2.376 2.548
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.502 2.539
fft3d_ps 1111 14.6 1.151 1.191 2.518 2.538
calculate_first_density_matrix 1 7.0 0.000 0.000 2.478 2.482
acc_transpose_blocks 24660 15.4 0.111 0.114 2.393 2.475
grid_collocate_task_list 110 9.6 2.264 2.395 2.264 2.395
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.235 2.243
potential_pw2rs 110 12.3 0.012 0.013 2.072 2.075
mp_alltoall_d11v 2046 13.8 1.764 2.039 1.764 2.039
qs_ot_get_orbitals 99 10.5 0.001 0.001 1.791 1.825
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.610 1.706
cp_fm_cholesky_decompose 22 10.9 1.639 1.683 1.639 1.683
mp_allgather_i34 2055 14.4 0.455 1.665 0.455 1.665
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.639 1.652
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.545 1.561
mp_sum_l 6594 12.7 0.933 1.550 0.933 1.550
multiply_cannon_metrocomm4 20550 15.4 0.059 0.061 0.863 1.496
acc_transpose_blocks_sync 73980 16.4 1.386 1.470 1.386 1.470
jit_kernel_multiply 8 15.4 0.679 1.431 0.679 1.431
transfer_rs2pw 451 10.6 0.005 0.006 1.185 1.416
mp_irecv_dv 62702 16.1 0.760 1.413 0.760 1.413
mp_waitany 13376 13.8 1.029 1.319 1.029 1.319
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="206", plot="h2o_128_md", label="(8n/3r/4t)", y=65.764000, yerr=0.000000
PlotPoint: name="207", plot="h2o_128_md_mem", label="(8n/3r/4t)", y=700.727273, yerr=11.160734
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 851.464192E+06
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 197280
MPI messages size (bytes):
total size 339.125567E+09
min size 0.000000E+00
max size 13.107200E+06
average size 1.719006E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 132 4325376
32768 < size <= 131072 88656 11620319232
131072 < size <= 4194304 89424 117209825280
4194304 < size <= 16777216 17616 210291069504
16777216 < size 0 0
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 7346 33.
MP_Alltoall 8043 263767.
MP_ISend 32836 654203.
MP_IRecv 32836 654587.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3473 66417.
MP_Allreduce 9774 644.
MP_Sync 52
MP_Alltoall 1496 8504061.
MP_SendRecv 3300 54848.
MP_ISendRecv 3300 54848.
MP_Wait 13926
MP_ISend 9240 278857.
MP_IRecv 9240 278857.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.011 0.029 54.688 54.690
qs_mol_dyn_low 1 2.0 0.003 0.004 54.464 54.474
qs_forces 11 3.9 0.003 0.003 54.399 54.399
qs_energies 11 4.9 0.001 0.001 50.765 50.772
scf_env_do_scf 11 5.9 0.000 0.001 42.636 42.636
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 34.966 34.967
velocity_verlet 10 3.0 0.002 0.002 30.909 30.912
dbcsr_multiply_generic 2055 12.4 0.107 0.109 23.092 23.228
qs_scf_new_mos 99 7.5 0.001 0.001 20.448 20.501
qs_scf_loop_do_ot 99 8.5 0.001 0.001 20.447 20.500
ot_scf_mini 99 9.5 0.002 0.002 19.197 19.218
multiply_cannon 2055 13.4 0.238 0.249 17.607 19.005
multiply_cannon_loop 2055 14.4 0.605 0.626 16.335 16.654
rebuild_ks_matrix 110 8.3 0.000 0.000 11.620 11.641
qs_ks_build_kohn_sham_matrix 110 9.3 0.012 0.013 11.620 11.641
ot_mini 99 10.5 0.001 0.001 10.653 10.667
qs_ks_update_qs_env 110 7.6 0.001 0.001 10.339 10.358
multiply_cannon_multrec 8220 15.4 3.189 4.345 7.586 8.534
init_scf_loop 11 6.9 0.000 0.000 7.626 7.629
mp_waitall_1 103326 16.6 5.771 7.266 5.771 7.266
qs_ot_get_derivative 99 11.5 0.001 0.001 6.984 7.003
sum_up_and_integrate 110 10.3 0.001 0.002 6.184 6.195
integrate_v_rspace 110 11.3 0.003 0.003 6.157 6.168
prepare_preconditioner 11 7.9 0.000 0.000 5.983 5.988
make_preconditioner 11 8.9 0.000 0.000 5.983 5.988
qs_rho_update_rho_low 110 7.6 0.001 0.001 5.951 5.968
calculate_rho_elec 110 8.6 0.112 0.113 5.950 5.967
make_full_inverse_cholesky 11 9.9 0.000 0.000 5.570 5.643
dbcsr_mm_accdrv_process 17442 15.9 3.179 4.151 4.258 5.184
init_scf_run 11 5.9 0.000 0.001 5.024 5.024
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.023 5.023
qs_ot_get_p 110 10.4 0.001 0.001 4.605 4.630
multiply_cannon_metrocomm3 8220 15.4 0.018 0.019 2.980 4.575
make_m2s 4110 13.4 0.038 0.039 4.155 4.450
make_images 4110 14.4 0.644 0.701 4.025 4.321
pw_transfer 1331 11.6 0.066 0.070 4.049 4.057
fft_wrap_pw1pw2 1111 12.6 0.008 0.008 3.941 3.952
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 3.709 3.746
apply_single 110 13.6 0.000 0.000 3.709 3.746
ot_diis_step 99 11.5 0.012 0.012 3.648 3.648
grid_integrate_task_list 110 12.3 3.374 3.525 3.374 3.525
fft_wrap_pw1pw2_140 451 13.1 0.719 0.731 3.427 3.440
qs_ot_p2m_diag 48 11.0 0.081 0.084 3.264 3.267
density_rs2pw 110 9.6 0.004 0.004 2.992 3.147
cp_dbcsr_syevd 48 12.0 0.003 0.003 2.928 2.928
cp_fm_cholesky_invert 11 10.9 2.862 2.866 2.862 2.866
wfi_extrapolate 11 7.9 0.001 0.001 2.687 2.687
hybrid_alltoall_any 4261 16.3 0.198 0.845 2.194 2.657
qs_energies_init_hamiltonians 11 5.9 0.001 0.001 2.655 2.656
make_images_data 4110 15.4 0.040 0.046 2.234 2.602
grid_collocate_task_list 110 9.6 2.364 2.525 2.364 2.525
calculate_dm_sparse 110 9.5 0.001 0.001 2.475 2.524
fft3d_ps 1111 14.6 1.288 1.297 2.484 2.503
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.482 2.483
multiply_cannon_sync_h2d 8220 15.4 2.381 2.449 2.381 2.449
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.426 2.434
cp_fm_diag_elpa_base 48 14.0 2.370 2.397 2.424 2.432
calculate_first_density_matrix 1 7.0 0.000 0.000 2.241 2.242
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.129 2.143
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 1.986 2.012
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 1.782 1.994
potential_pw2rs 110 12.3 0.015 0.015 1.991 1.994
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 1.896 1.904
mp_alltoall_d11v 2046 13.8 1.588 1.778 1.588 1.778
cp_fm_cholesky_decompose 22 10.9 1.682 1.697 1.682 1.697
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.609 1.612
qs_env_update_s_mstruct 11 6.9 0.001 0.001 1.489 1.604
mp_allgather_i34 2055 14.4 0.442 1.534 0.442 1.534
dbcsr_complete_redistribute 325 12.2 0.556 0.587 1.408 1.497
cp_dbcsr_plus_fm_fm_t_native 22 8.9 0.001 0.001 1.425 1.439
acc_transpose_blocks 8220 15.4 0.041 0.042 1.361 1.389
qs_create_task_list 11 7.9 0.001 0.001 1.202 1.298
generate_qs_task_list 11 8.9 0.373 0.440 1.201 1.297
transfer_rs2pw 451 10.6 0.005 0.005 1.094 1.274
mp_waitany 9240 13.8 1.048 1.241 1.048 1.241
multiply_cannon_metrocomm1 8220 15.4 0.022 0.022 0.747 1.182
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.127 1.153
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="208", plot="h2o_128_md", label="(8n/2r/6t)", y=54.690000, yerr=0.000000
PlotPoint: name="209", plot="h2o_128_md_mem", label="(8n/2r/6t)", y=805.727273, yerr=11.070772
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 1.387241E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 82200
MPI messages size (bytes):
total size 297.640985E+09
min size 0.000000E+00
max size 26.214400E+06
average size 3.620936E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 44 1441792
32768 < size <= 131072 18560 2432696320
131072 < size <= 4194304 54216 84915781632
4194304 < size <= 16777216 0 0
16777216 < size 8808 210291069504
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3462 67098.
MP_Allreduce 9752 812.
MP_Sync 52
MP_Alltoall 1474 16505187.
MP_SendRecv 2310 360267.
MP_ISendRecv 2310 360267.
MP_Wait 5214
MP_ISend 2420 1187840.
MP_IRecv 2420 1187840.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.035 87.404 87.405
qs_mol_dyn_low 1 2.0 0.003 0.004 87.076 87.087
qs_forces 11 3.9 0.003 0.003 87.005 87.006
qs_energies 11 4.9 0.001 0.002 82.893 82.896
scf_env_do_scf 11 5.9 0.000 0.001 72.877 72.877
velocity_verlet 10 3.0 0.002 0.005 55.721 55.727
scf_env_do_scf_inner_loop 99 6.5 0.003 0.007 44.502 44.503
dbcsr_multiply_generic 2055 12.4 0.134 0.148 29.282 29.385
init_scf_loop 11 6.9 0.000 0.000 28.305 28.307
qs_scf_new_mos 99 7.5 0.001 0.001 26.633 26.668
qs_scf_loop_do_ot 99 8.5 0.001 0.001 26.632 26.667
prepare_preconditioner 11 7.9 0.000 0.000 26.171 26.179
make_preconditioner 11 8.9 0.000 0.000 26.171 26.179
make_full_inverse_cholesky 11 9.9 0.000 0.000 20.334 25.659
ot_scf_mini 99 9.5 0.002 0.002 24.841 24.866
multiply_cannon 2055 13.4 0.331 0.349 22.296 23.082
multiply_cannon_loop 2055 14.4 0.812 0.833 20.531 20.826
cp_fm_upper_to_full 70 14.2 12.755 18.336 12.755 18.336
rebuild_ks_matrix 110 8.3 0.000 0.001 14.034 14.069
qs_ks_build_kohn_sham_matrix 110 9.3 0.013 0.013 14.033 14.068
ot_mini 99 10.5 0.001 0.001 13.981 14.009
qs_ks_update_qs_env 110 7.6 0.001 0.001 12.735 12.766
dbcsr_complete_redistribute 325 12.2 1.001 1.030 7.376 10.603
multiply_cannon_multrec 8220 15.4 4.055 4.215 9.753 9.892
copy_fm_to_dbcsr 174 11.2 0.001 0.001 6.379 9.599
qs_ot_get_derivative 99 11.5 0.001 0.001 9.423 9.451
transfer_fm_to_dbcsr 11 9.9 0.000 0.000 5.821 9.003
mp_alltoall_i22 605 13.7 5.457 8.695 5.457 8.695
mp_waitall_1 84994 16.7 7.347 8.164 7.347 8.164
qs_rho_update_rho_low 110 7.6 0.001 0.001 7.464 7.500
calculate_rho_elec 110 8.6 0.222 0.223 7.464 7.500
sum_up_and_integrate 110 10.3 0.002 0.002 6.903 6.916
integrate_v_rspace 110 11.3 0.004 0.004 6.875 6.887
pw_transfer 1331 11.6 0.075 0.075 5.798 5.805
dbcsr_mm_accdrv_process 11614 15.7 3.887 4.121 5.550 5.767
init_scf_run 11 5.9 0.000 0.001 5.745 5.745
scf_env_initial_rho_setup 11 6.9 0.001 0.001 5.745 5.745
make_m2s 4110 13.4 0.043 0.043 5.273 5.736
fft_wrap_pw1pw2 1111 12.6 0.009 0.009 5.682 5.689
qs_ot_get_p 110 10.4 0.001 0.001 5.596 5.631
make_images 4110 14.4 0.880 0.933 5.083 5.544
apply_preconditioner_dbcsr 110 12.6 0.000 0.000 4.704 5.174
apply_single 110 13.6 0.000 0.000 4.704 5.174
cp_fm_cholesky_invert 11 10.9 5.105 5.109 5.105 5.109
multiply_cannon_metrocomm3 8220 15.4 0.019 0.019 4.763 5.056
fft_wrap_pw1pw2_140 451 13.1 1.367 1.372 4.992 5.002
ot_diis_step 99 11.5 0.015 0.015 4.542 4.542
qs_ot_p2m_diag 48 11.0 0.151 0.156 4.046 4.051
density_rs2pw 110 9.6 0.004 0.004 3.934 3.953
grid_integrate_task_list 110 12.3 3.673 3.749 3.673 3.749
qs_energies_init_hamiltonians 11 5.9 0.003 0.003 3.663 3.664
cp_dbcsr_syevd 48 12.0 0.003 0.003 3.525 3.526
hybrid_alltoall_any 4261 16.3 0.263 0.563 2.784 3.517
qs_ot_get_derivative_taylor 52 13.0 0.001 0.001 2.997 3.458
fft3d_ps 1111 14.6 1.880 1.890 3.387 3.391
make_images_data 4110 15.4 0.045 0.048 2.749 3.349
wfi_extrapolate 11 7.9 0.001 0.001 3.330 3.330
multiply_cannon_sync_h2d 8220 15.4 3.148 3.178 3.148 3.178
calculate_dm_sparse 110 9.5 0.001 0.001 3.153 3.175
cp_fm_diag_elpa 48 13.0 0.000 0.000 2.936 2.936
cp_fm_diag_elpa_base 48 14.0 2.393 2.600 2.933 2.933
cp_dbcsr_sm_fm_multiply 37 9.5 0.001 0.001 2.914 2.917
grid_collocate_task_list 110 9.6 2.665 2.688 2.665 2.688
qs_ot_get_derivative_diag 47 12.0 0.001 0.001 2.630 2.646
potential_pw2rs 110 12.3 0.021 0.021 2.436 2.439
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 2.423 2.436
calculate_first_density_matrix 1 7.0 0.000 0.000 2.300 2.300
qs_env_update_s_mstruct 11 6.9 0.001 0.001 2.189 2.237
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 2.091 2.140
mp_alltoall_d11v 2046 13.8 2.026 2.087 2.026 2.087
cp_fm_cholesky_decompose 22 10.9 1.966 1.994 1.966 1.994
qs_create_task_list 11 7.9 0.001 0.001 1.879 1.924
generate_qs_task_list 11 8.9 0.731 0.786 1.879 1.924
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 1.896 1.899
copy_dbcsr_to_fm 151 11.3 0.003 0.003 1.732 1.767
jit_kernel_multiply 10 15.1 1.467 1.761 1.467 1.761
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="210", plot="h2o_128_md", label="(8n/1r/12t)", y=87.405000, yerr=0.000000
PlotPoint: name="211", plot="h2o_128_md_mem", label="(8n/1r/12t)", y=1258.000000, yerr=52.483764
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/15/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420241154048 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 11528896499712 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 93.514757E+12 0.0% 0.0% 100.0%
flops max/rank 1.094965E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755939872 0.0% 0.0% 100.0%
number of processed stacks 11950464 0.0% 0.0% 100.0%
average stack size 0.0 0.0 565.3
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 629.936128E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 10348896
MPI messages size (bytes):
total size 4.491514E+12
min size 0.000000E+00
max size 4.537280E+06
average size 434.009000E+03
MPI breakdown and total messages size (bytes):
size <= 128 65736 0
128 < size <= 8192 1232 10092544
8192 < size <= 32768 3576680 95640223744
32768 < size <= 131072 1294784 74079797248
131072 < size <= 4194304 5148576 3175955098256
4194304 < size <= 16777216 261888 1145794321408
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 57905.
MP_Allreduce 11059 797.
MP_Sync 87
MP_Alltoall 2226 2686109.
MP_SendRecv 24320 18752.
MP_ISendRecv 24320 18752.
MP_Wait 42476
MP_ISend 16020 108028.
MP_IRecv 16020 108028.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.018 0.037 207.149 207.155
qs_mol_dyn_low 1 2.0 0.004 0.006 206.517 206.531
qs_forces 11 3.9 0.004 0.005 206.420 206.422
qs_energies 11 4.9 0.002 0.002 200.837 200.852
scf_env_do_scf 11 5.9 0.001 0.001 184.217 184.221
scf_env_do_scf_inner_loop 117 6.6 0.005 0.012 163.817 163.819
dbcsr_multiply_generic 2507 12.6 0.178 0.182 125.553 126.124
qs_scf_new_mos 117 7.6 0.001 0.001 124.386 124.664
qs_scf_loop_do_ot 117 8.6 0.001 0.001 124.385 124.664
velocity_verlet 10 3.0 0.001 0.002 123.466 123.469
ot_scf_mini 117 9.6 0.003 0.003 117.615 117.848
multiply_cannon 2507 13.6 0.237 0.247 101.391 102.632
multiply_cannon_loop 2507 14.6 2.381 2.454 99.268 100.316
ot_mini 117 10.6 0.001 0.001 66.097 66.322
multiply_cannon_multrec 60168 15.6 31.742 33.437 41.508 43.380
qs_ot_get_derivative 117 11.6 0.001 0.002 41.441 41.680
rebuild_ks_matrix 128 8.3 0.001 0.001 33.497 33.736
qs_ks_build_kohn_sham_matrix 128 9.3 0.015 0.017 33.496 33.735
mp_waitall_1 267128 16.5 28.746 31.593 28.746 31.593
qs_ks_update_qs_env 128 7.6 0.001 0.001 30.089 30.295
qs_ot_get_p 128 10.4 0.001 0.001 29.295 29.551
multiply_cannon_sync_h2d 60168 15.6 26.488 28.408 26.488 28.408
apply_preconditioner_dbcsr 128 12.6 0.000 0.001 24.252 25.131
apply_single 128 13.6 0.001 0.001 24.252 25.131
ot_diis_step 117 11.6 0.008 0.008 24.423 24.424
qs_ot_p2m_diag 83 11.4 0.079 0.092 22.708 22.802
init_scf_loop 11 6.9 0.000 0.001 20.327 20.329
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 19.972 20.152
cp_dbcsr_syevd 83 12.4 0.005 0.005 19.911 19.912
multiply_cannon_metrocomm3 60168 15.6 0.116 0.123 16.021 17.879
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.949 16.981
cp_fm_diag_elpa_base 83 14.4 16.876 16.909 16.944 16.978
prepare_preconditioner 11 7.9 0.000 0.000 15.742 15.784
make_preconditioner 11 8.9 0.000 0.000 15.742 15.784
make_full_inverse_cholesky 11 9.9 0.000 0.000 15.001 15.181
make_m2s 5014 13.6 0.105 0.113 13.904 14.349
make_images 5014 14.6 0.396 0.414 13.717 14.173
sum_up_and_integrate 128 10.3 0.002 0.004 14.010 14.025
integrate_v_rspace 128 11.3 0.003 0.004 13.951 13.969
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.189 13.318
calculate_rho_elec 128 8.7 0.045 0.064 13.188 13.317
init_scf_run 11 5.9 0.000 0.001 12.485 12.486
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.485 12.486
mp_sum_l 7950 12.9 8.998 10.166 8.998 10.166
dbcsr_mm_accdrv_process 124484 16.2 4.714 4.943 9.331 9.952
wfi_extrapolate 11 7.9 0.001 0.001 9.033 9.033
cp_fm_cholesky_invert 11 10.9 8.988 8.996 8.988 8.996
calculate_dm_sparse 128 9.5 0.001 0.001 8.668 8.781
qs_ot_get_orbitals 117 10.6 0.001 0.001 8.177 8.249
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 8.077 8.207
pw_transfer 1547 11.6 0.074 0.087 7.652 7.837
make_images_data 5014 15.6 0.066 0.073 6.801 7.758
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 7.450 7.632
multiply_cannon_metrocomm1 60168 15.6 0.092 0.096 6.178 7.547
grid_integrate_task_list 128 12.3 7.091 7.482 7.091 7.482
density_rs2pw 128 9.7 0.006 0.006 6.818 7.315
hybrid_alltoall_any 5200 16.5 0.292 2.259 5.977 7.260
fft_wrap_pw1pw2_140 523 13.2 0.861 0.909 6.527 6.699
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.004 6.623 6.634
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.763 5.868
fft3d_ps 1291 14.7 2.205 2.875 5.450 5.769
mp_alltoall_d11v 2415 14.1 4.280 5.449 4.280 5.449
grid_collocate_task_list 128 9.7 4.838 5.249 4.838 5.249
cp_fm_cholesky_decompose 22 10.9 4.613 4.627 4.613 4.627
potential_pw2rs 128 12.3 0.009 0.010 4.513 4.540
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="400", plot="h2o_256_md", label="(8n/12r/1t)", y=207.155000, yerr=0.000000
PlotPoint: name="401", plot="h2o_256_md_mem", label="(8n/12r/1t)", y=597.272727, yerr=5.395131
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/16/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 2.183246E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 5975232 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1130.7
marketing flops 144.580175E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 843.816960E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 2406720
MPI messages size (bytes):
total size 4.100942E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.703955E+06
MPI breakdown and total messages size (bytes):
size <= 128 14916 0
128 < size <= 8192 0 0
8192 < size <= 32768 70860 2317615104
32768 < size <= 131072 722992 55511613440
131072 < size <= 4194304 1375664 1398181724160
4194304 < size <= 16777216 154704 1463834332048
16777216 < size 67584 1181116006400
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58357.
MP_Allreduce 11058 960.
MP_Sync 87
MP_Alltoall 1969 4906255.
MP_SendRecv 12032 47072.
MP_ISendRecv 12032 47072.
MP_Wait 25916
MP_ISend 11748 212467.
MP_IRecv 11748 212467.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.024 0.055 189.248 189.249
qs_mol_dyn_low 1 2.0 0.003 0.004 188.866 188.896
qs_forces 11 3.9 0.004 0.005 188.784 188.786
qs_energies 11 4.9 0.001 0.002 182.064 182.074
scf_env_do_scf 11 5.9 0.001 0.001 165.618 165.628
scf_env_do_scf_inner_loop 117 6.6 0.003 0.009 133.254 133.257
velocity_verlet 10 3.0 0.001 0.002 118.587 118.588
dbcsr_multiply_generic 2507 12.6 0.252 0.259 97.740 98.966
qs_scf_new_mos 117 7.6 0.001 0.001 94.652 95.224
qs_scf_loop_do_ot 117 8.6 0.001 0.001 94.651 95.223
ot_scf_mini 117 9.6 0.004 0.004 89.841 90.564
multiply_cannon 2507 13.6 0.502 0.558 77.610 82.587
multiply_cannon_loop 2507 14.6 1.573 1.642 74.242 76.622
ot_mini 117 10.6 0.001 0.001 50.490 51.136
mp_waitall_1 214728 16.6 24.010 39.175 24.010 39.175
multiply_cannon_multrec 30084 15.6 21.103 26.038 31.750 37.070
rebuild_ks_matrix 128 8.3 0.001 0.001 32.281 32.995
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.019 32.280 32.995
init_scf_loop 11 6.9 0.000 0.000 32.275 32.276
qs_ks_update_qs_env 128 7.6 0.001 0.001 29.039 29.683
qs_ot_get_derivative 117 11.6 0.001 0.002 28.601 29.307
multiply_cannon_metrocomm3 30084 15.6 0.099 0.106 15.609 29.068
prepare_preconditioner 11 7.9 0.000 0.000 27.979 28.042
make_preconditioner 11 8.9 0.000 0.000 27.979 28.042
make_full_inverse_cholesky 11 9.9 0.000 0.000 26.662 27.182
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 22.007 23.231
apply_single 128 13.6 0.001 0.001 22.007 23.230
qs_ot_get_p 128 10.4 0.001 0.001 21.371 22.187
ot_diis_step 117 11.6 0.014 0.015 21.714 21.716
multiply_cannon_sync_h2d 30084 15.6 18.177 21.217 18.177 21.217
qs_ot_p2m_diag 83 11.4 0.188 0.217 16.434 16.468
cp_fm_cholesky_invert 11 10.9 16.304 16.316 16.304 16.316
make_m2s 5014 13.6 0.086 0.092 14.081 15.762
make_images 5014 14.6 1.155 1.362 13.864 15.547
cp_dbcsr_syevd 83 12.4 0.005 0.006 15.232 15.232
sum_up_and_integrate 128 10.3 0.002 0.004 13.993 14.023
integrate_v_rspace 128 11.3 0.003 0.004 13.933 13.965
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.384 13.416
calculate_rho_elec 128 8.7 0.087 0.103 13.383 13.415
cp_fm_diag_elpa 83 13.4 0.000 0.000 12.094 12.128
cp_fm_diag_elpa_base 83 14.4 11.841 11.937 12.089 12.118
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 11.232 11.737
init_scf_run 11 5.9 0.000 0.001 11.665 11.666
scf_env_initial_rho_setup 11 6.9 0.001 0.001 11.664 11.666
multiply_cannon_metrocomm4 27577 15.6 0.100 0.115 3.818 11.005
dbcsr_mm_accdrv_process 62242 16.2 5.496 6.270 10.094 10.604
mp_irecv_dv 69486 16.3 3.618 10.600 3.618 10.600
make_images_data 5014 15.6 0.066 0.076 8.272 10.313
hybrid_alltoall_any 5200 16.5 0.346 1.571 7.042 9.917
pw_transfer 1547 11.6 0.084 0.095 8.586 8.645
fft_wrap_pw1pw2 1291 12.7 0.010 0.011 8.364 8.423
wfi_extrapolate 11 7.9 0.001 0.001 8.337 8.337
density_rs2pw 128 9.7 0.006 0.006 7.110 7.963
grid_integrate_task_list 128 12.3 7.145 7.584 7.145 7.584
fft_wrap_pw1pw2_140 523 13.2 0.935 0.956 7.391 7.463
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 6.283 7.142
cp_fm_cholesky_decompose 22 10.9 6.838 6.911 6.838 6.911
calculate_dm_sparse 128 9.5 0.001 0.001 6.466 6.629
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.325 6.333
mp_sum_l 7950 12.9 4.188 6.199 4.188 6.199
fft3d_ps 1291 14.7 2.827 3.003 5.920 5.971
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 5.295 5.473
qs_ot_get_orbitals 117 10.6 0.001 0.001 5.361 5.439
grid_collocate_task_list 128 9.7 4.996 5.373 4.996 5.373
mp_alltoall_d11v 2415 14.1 4.157 5.284 4.157 5.284
mp_allgather_i34 2507 14.6 1.895 4.859 1.895 4.859
potential_pw2rs 128 12.3 0.015 0.017 4.476 4.489
dbcsr_complete_redistribute 395 12.7 0.858 0.984 3.253 4.106
mp_sum_d 4467 12.1 2.607 3.963 2.607 3.963
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="402", plot="h2o_256_md", label="(8n/6r/2t)", y=189.249000, yerr=0.000000
PlotPoint: name="403", plot="h2o_256_md_mem", label="(8n/6r/2t)", y=803.909091, yerr=2.353123
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/17/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 2.928533E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 3984192 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1695.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 954.236928E+06
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 1042912
MPI messages size (bytes):
total size 2.716210E+12
min size 0.000000E+00
max size 26.214400E+06
average size 2.604448E+06
MPI breakdown and total messages size (bytes):
size <= 128 6424 0
128 < size <= 8192 0 0
8192 < size <= 32768 264 8650752
32768 < size <= 131072 281856 36943429632
131072 < size <= 4194304 660064 996105256960
4194304 < size <= 16777216 65632 931530938576
16777216 < size 28672 751619276800
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58351.
MP_Allreduce 11057 1000.
MP_Sync 87
MP_Alltoall 1712 9388896.
MP_SendRecv 7936 75008.
MP_ISendRecv 7936 75008.
MP_Wait 21820
MP_ISend 11748 275205.
MP_IRecv 11748 275205.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.019 0.042 176.434 176.435
qs_mol_dyn_low 1 2.0 0.004 0.004 175.969 175.981
qs_forces 11 3.9 0.005 0.013 175.871 175.873
qs_energies 11 4.9 0.002 0.007 169.340 169.350
scf_env_do_scf 11 5.9 0.001 0.002 153.810 153.810
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 118.734 118.735
velocity_verlet 10 3.0 0.004 0.006 112.050 112.052
dbcsr_multiply_generic 2507 12.6 0.185 0.189 81.871 83.042
qs_scf_new_mos 117 7.6 0.001 0.001 81.479 81.818
qs_scf_loop_do_ot 117 8.6 0.001 0.002 81.478 81.818
ot_scf_mini 117 9.6 0.004 0.005 77.246 77.640
multiply_cannon 2507 13.6 0.504 0.521 62.165 66.726
multiply_cannon_loop 2507 14.6 1.134 1.201 59.048 62.010
ot_mini 117 10.6 0.001 0.001 42.398 42.796
init_scf_loop 11 6.9 0.001 0.006 34.979 34.980
mp_waitall_1 170520 16.6 24.562 33.857 24.562 33.857
prepare_preconditioner 11 7.9 0.000 0.000 30.959 31.009
make_preconditioner 11 8.9 0.000 0.002 30.959 31.009
rebuild_ks_matrix 128 8.3 0.001 0.001 30.294 30.824
qs_ks_build_kohn_sham_matrix 128 9.3 0.016 0.023 30.293 30.823
make_full_inverse_cholesky 11 9.9 0.000 0.000 28.619 30.000
qs_ks_update_qs_env 128 7.6 0.001 0.001 27.309 27.792
multiply_cannon_multrec 20056 15.6 13.065 16.877 22.405 26.228
multiply_cannon_metrocomm3 20056 15.6 0.061 0.065 15.113 24.892
qs_ot_get_derivative 117 11.6 0.002 0.002 22.922 23.317
qs_ot_get_p 128 10.4 0.001 0.002 20.147 20.661
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.488 20.538
apply_single 128 13.6 0.001 0.001 19.488 20.537
ot_diis_step 117 11.6 0.018 0.019 19.369 19.370
make_m2s 5014 13.6 0.078 0.084 14.691 16.083
make_images 5014 14.6 1.137 1.221 14.453 15.835
qs_ot_p2m_diag 83 11.4 0.265 0.272 15.704 15.713
multiply_cannon_sync_h2d 20056 15.6 13.642 15.223 13.642 15.223
cp_dbcsr_syevd 83 12.4 0.005 0.005 14.637 14.638
cp_fm_cholesky_invert 11 10.9 14.419 14.428 14.419 14.428
sum_up_and_integrate 128 10.3 0.002 0.004 13.947 13.976
integrate_v_rspace 128 11.3 0.003 0.004 13.888 13.910
qs_rho_update_rho_low 128 7.7 0.001 0.001 13.664 13.701
calculate_rho_elec 128 8.7 0.128 0.143 13.664 13.700
cp_fm_diag_elpa 83 13.4 0.000 0.000 11.381 11.404
cp_fm_diag_elpa_base 83 14.4 10.969 11.123 11.377 11.400
make_images_data 5014 15.6 0.063 0.073 8.824 10.603
init_scf_run 11 5.9 0.000 0.001 10.504 10.504
scf_env_initial_rho_setup 11 6.9 0.001 0.003 10.504 10.504
hybrid_alltoall_any 5200 16.5 0.445 2.014 7.689 10.213
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.104 9.395
multiply_cannon_metrocomm4 17549 15.6 0.065 0.073 3.473 9.262
mp_irecv_dv 50230 16.2 3.345 9.008 3.345 9.008
pw_transfer 1547 11.6 0.084 0.105 8.885 8.985
dbcsr_mm_accdrv_process 41502 16.2 5.650 5.985 8.797 8.929
fft_wrap_pw1pw2 1291 12.7 0.010 0.010 8.662 8.767
grid_integrate_task_list 128 12.3 7.326 7.796 7.326 7.796
fft_wrap_pw1pw2_140 523 13.2 1.060 1.096 7.640 7.749
density_rs2pw 128 9.7 0.006 0.006 7.036 7.641
wfi_extrapolate 11 7.9 0.001 0.001 7.419 7.419
cp_fm_cholesky_decompose 22 10.9 7.364 7.412 7.364 7.412
cp_fm_upper_to_full 105 14.8 5.762 7.274 5.762 7.274
dbcsr_complete_redistribute 395 12.7 1.161 1.190 4.585 6.348
fft3d_ps 1291 14.7 2.820 3.050 5.980 6.066
calculate_dm_sparse 128 9.5 0.001 0.001 5.842 5.933
grid_collocate_task_list 128 9.7 5.156 5.666 5.156 5.666
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.438 5.443
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.615 5.253
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.402 5.163
mp_alltoall_d11v 2415 14.1 4.281 4.938 4.281 4.938
mp_allgather_i34 2507 14.6 1.622 4.872 1.622 4.872
mp_sum_l 7950 12.9 3.188 4.743 3.188 4.743
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.551 4.721
potential_pw2rs 128 12.3 0.020 0.021 4.384 4.398
transfer_fm_to_dbcsr 11 9.9 0.017 0.022 2.321 4.068
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.919 3.945
mp_alltoall_i22 716 14.1 1.938 3.883 1.938 3.883
qs_energies_init_hamiltonians 11 5.9 0.002 0.005 3.745 3.745
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.484 3.534
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="404", plot="h2o_256_md", label="(8n/4r/3t)", y=176.435000, yerr=0.000000
PlotPoint: name="405", plot="h2o_256_md_mem", label="(8n/4r/3t)", y=906.545455, yerr=9.296760
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/18/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1430456039424 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1962800054272 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1986255912960 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1992003932160 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2753958699008 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4454954827776 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5444944789504 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5492290093056 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6712799002624 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11613072052224 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15239176077312 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15239176077312 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19911132921856 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 94.233020E+12 0.0% 0.0% 100.0%
flops max/rank 4.387242E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6806383904 0.0% 0.0% 100.0%
number of processed stacks 6026880 0.0% 0.0% 100.0%
average stack size 0.0 0.0 1129.3
marketing flops 145.651870E+12
-------------------------------------------------------------------------------
# multiplications 2529
max memory usage/rank 1.151717E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 1153224
MPI messages size (bytes):
total size 2.039489E+12
min size 0.000000E+00
max size 17.653760E+06
average size 1.768511E+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 322096 36390305792
131072 < size <= 4194304 721976 792118951936
4194304 < size <= 16777216 70800 669922227920
16777216 < size 30960 541065216000
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4039 57682.
MP_Allreduce 11174 1079.
MP_Sync 88
MP_Alltoall 1724 12509605.
MP_SendRecv 5934 75008.
MP_ISendRecv 5934 75008.
MP_Wait 22612
MP_ISend 15064 244788.
MP_IRecv 15064 244788.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.012 0.031 190.673 190.675
qs_mol_dyn_low 1 2.0 0.003 0.004 190.309 190.322
qs_forces 11 3.9 0.004 0.004 190.165 190.179
qs_energies 11 4.9 0.002 0.002 183.088 183.096
scf_env_do_scf 11 5.9 0.001 0.001 166.482 166.495
velocity_verlet 10 3.0 0.001 0.002 125.392 125.396
scf_env_do_scf_inner_loop 118 6.6 0.003 0.008 120.157 120.158
qs_scf_new_mos 118 7.6 0.001 0.001 83.264 83.556
qs_scf_loop_do_ot 118 8.6 0.001 0.001 83.263 83.555
dbcsr_multiply_generic 2529 12.6 0.194 0.199 81.094 81.825
ot_scf_mini 118 9.6 0.003 0.004 78.605 78.936
multiply_cannon 2529 13.6 0.555 0.588 56.115 59.116
multiply_cannon_loop 2529 14.6 1.854 1.932 52.392 54.146
init_scf_loop 11 6.9 0.000 0.000 46.206 46.207
ot_mini 118 10.6 0.001 0.001 43.365 43.688
prepare_preconditioner 11 7.9 0.000 0.000 42.140 42.166
make_preconditioner 11 8.9 0.000 0.000 42.140 42.166
make_full_inverse_cholesky 11 9.9 0.010 0.019 35.766 40.828
multiply_cannon_multrec 30348 15.6 13.691 19.098 26.693 31.730
rebuild_ks_matrix 129 8.3 0.001 0.001 29.619 29.900
qs_ks_build_kohn_sham_matrix 129 9.3 0.016 0.019 29.619 29.900
mp_waitall_1 149172 16.7 17.206 27.206 17.206 27.206
qs_ks_update_qs_env 129 7.6 0.001 0.001 26.767 27.015
qs_ot_get_derivative 118 11.6 0.002 0.002 23.651 23.982
qs_ot_get_p 129 10.4 0.001 0.001 21.255 21.626
make_m2s 5058 13.6 0.093 0.098 20.526 21.443
make_images 5058 14.6 1.933 2.250 20.213 21.126
apply_preconditioner_dbcsr 129 12.6 0.000 0.001 19.081 19.585
apply_single 129 13.6 0.001 0.001 19.080 19.584
ot_diis_step 118 11.6 0.018 0.018 19.582 19.584
cp_fm_upper_to_full 106 14.8 11.559 17.042 11.559 17.042
qs_ot_p2m_diag 84 11.4 0.347 0.394 16.939 16.990
cp_fm_cholesky_invert 11 10.9 16.009 16.018 16.009 16.018
cp_dbcsr_syevd 84 12.4 0.005 0.006 15.505 15.506
multiply_cannon_metrocomm3 30348 15.6 0.049 0.053 6.282 15.057
sum_up_and_integrate 129 10.3 0.002 0.003 14.175 14.203
integrate_v_rspace 129 11.3 0.003 0.006 14.115 14.144
qs_rho_update_rho_low 129 7.7 0.001 0.001 14.018 14.079
calculate_rho_elec 129 8.7 0.171 0.186 14.018 14.079
dbcsr_mm_accdrv_process 62780 16.2 8.547 9.284 12.572 13.098
dbcsr_complete_redistribute 397 12.7 1.490 1.598 9.148 13.012
make_images_data 5058 15.6 0.065 0.072 10.976 12.585
cp_fm_diag_elpa 84 13.4 0.000 0.000 12.154 12.168
cp_fm_diag_elpa_base 84 14.4 11.167 11.516 12.146 12.159
multiply_cannon_sync_h2d 30348 15.6 10.625 11.678 10.625 11.678
copy_fm_to_dbcsr 210 11.7 0.001 0.002 7.739 11.556
hybrid_alltoall_any 5245 16.5 0.522 2.140 9.796 11.444
init_scf_run 11 5.9 0.000 0.001 10.704 10.705
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.703 10.705
qs_ot_get_derivative_diag 78 12.4 0.002 0.003 10.098 10.349
transfer_fm_to_dbcsr 11 9.9 0.002 0.007 6.353 10.105
mp_alltoall_i22 720 14.1 5.638 9.503 5.638 9.503
pw_transfer 1559 11.6 0.086 0.103 9.344 9.437
fft_wrap_pw1pw2 1301 12.7 0.010 0.011 9.117 9.216
fft_wrap_pw1pw2_140 527 13.2 1.257 1.291 8.081 8.197
grid_integrate_task_list 129 12.3 7.595 7.930 7.595 7.930
cp_fm_cholesky_decompose 22 10.9 7.476 7.585 7.476 7.585
wfi_extrapolate 11 7.9 0.001 0.001 7.530 7.530
density_rs2pw 129 9.7 0.005 0.006 7.003 7.459
multiply_cannon_metrocomm4 25290 15.6 0.080 0.089 2.830 7.219
mp_irecv_dv 76751 16.2 2.677 6.941 2.677 6.941
calculate_dm_sparse 129 9.5 0.001 0.001 6.336 6.426
fft3d_ps 1301 14.7 3.011 3.094 6.146 6.210
grid_collocate_task_list 129 9.7 5.323 5.710 5.323 5.710
mp_alltoall_d11v 2429 14.1 5.068 5.651 5.068 5.651
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.425 5.465
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 4.509 4.603
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.348 4.450
potential_pw2rs 129 12.3 0.023 0.024 4.407 4.417
qs_energies_init_hamiltonians 11 5.9 0.001 0.002 4.401 4.402
qs_ot_get_orbitals 118 10.6 0.001 0.001 4.302 4.358
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="406", plot="h2o_256_md", label="(8n/3r/4t)", y=190.675000, yerr=0.000000
PlotPoint: name="407", plot="h2o_256_md_mem", label="(8n/3r/4t)", y=1082.727273, yerr=24.491523
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/19/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 5.865088E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 1960712 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3445.7
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 1.572782E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 240672
MPI messages size (bytes):
total size 1.331455E+12
min size 0.000000E+00
max size 52.428800E+06
average size 5.532237E+06
MPI breakdown and total messages size (bytes):
size <= 128 1452 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 132 8650752
131072 < size <= 4194304 113904 59718500352
4194304 < size <= 16777216 104976 550376570880
16777216 < size 20208 721350092304
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 14 12.
MP_Allreduce 8931 51.
MP_Alltoall 9654 799394.
MP_ISend 40068 2102572.
MP_IRecv 40068 2101675.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 58346.
MP_Allreduce 11057 1167.
MP_Sync 87
MP_Alltoall 1712 18838186.
MP_SendRecv 3840 122880.
MP_ISendRecv 3840 122880.
MP_Wait 16122
MP_ISend 10680 423556.
MP_IRecv 10680 423556.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.015 0.052 168.985 169.003
qs_mol_dyn_low 1 2.0 0.003 0.004 168.599 168.612
qs_forces 11 3.9 0.004 0.004 168.506 168.508
qs_energies 11 4.9 0.027 0.032 161.138 161.147
scf_env_do_scf 11 5.9 0.001 0.001 143.954 143.965
velocity_verlet 10 3.0 0.002 0.002 110.849 110.853
scf_env_do_scf_inner_loop 117 6.6 0.003 0.008 108.834 108.836
dbcsr_multiply_generic 2507 12.6 0.187 0.193 71.984 72.479
qs_scf_new_mos 117 7.6 0.001 0.001 71.775 71.878
qs_scf_loop_do_ot 117 8.6 0.001 0.001 71.774 71.877
ot_scf_mini 117 9.6 0.003 0.004 67.363 67.465
multiply_cannon 2507 13.6 0.558 0.578 53.540 56.501
multiply_cannon_loop 2507 14.6 0.807 0.833 50.432 51.534
ot_mini 117 10.6 0.001 0.001 37.695 37.792
init_scf_loop 11 6.9 0.000 0.000 34.967 34.969
mp_waitall_1 125778 16.7 24.788 31.311 24.788 31.311
prepare_preconditioner 11 7.9 0.000 0.000 31.131 31.151
make_preconditioner 11 8.9 0.000 0.000 31.131 31.151
rebuild_ks_matrix 128 8.3 0.001 0.001 29.169 29.317
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.018 29.169 29.316
make_full_inverse_cholesky 11 9.9 0.012 0.027 28.957 29.215
qs_ks_update_qs_env 128 7.6 0.001 0.001 26.613 26.754
multiply_cannon_multrec 10028 15.6 10.308 15.845 17.879 22.412
qs_ot_get_derivative 117 11.6 0.002 0.002 20.447 20.545
multiply_cannon_metrocomm3 10028 15.6 0.024 0.026 12.904 20.447
cp_fm_cholesky_invert 11 10.9 17.708 17.713 17.708 17.713
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 17.299 17.479
apply_single 128 13.6 0.001 0.001 17.299 17.478
ot_diis_step 117 11.6 0.020 0.020 17.177 17.178
qs_ot_get_p 128 10.4 0.001 0.001 16.967 17.069
make_m2s 5014 13.6 0.064 0.068 14.595 15.504
make_images 5014 14.6 2.151 2.583 14.286 15.189
qs_rho_update_rho_low 128 7.7 0.001 0.001 14.233 14.278
calculate_rho_elec 128 8.7 0.253 0.264 14.232 14.278
sum_up_and_integrate 128 10.3 0.002 0.002 14.192 14.245
integrate_v_rspace 128 11.3 0.004 0.006 14.132 14.184
qs_ot_p2m_diag 83 11.4 0.496 0.501 13.254 13.270
cp_dbcsr_syevd 83 12.4 0.005 0.006 12.083 12.084
multiply_cannon_sync_h2d 10028 15.6 10.724 11.216 10.724 11.216
init_scf_run 11 5.9 0.000 0.001 10.269 10.269
scf_env_initial_rho_setup 11 6.9 0.001 0.001 10.269 10.269
pw_transfer 1547 11.6 0.085 0.092 9.870 9.902
make_images_data 5014 15.6 0.054 0.065 8.358 9.708
fft_wrap_pw1pw2 1291 12.7 0.010 0.010 9.648 9.689
hybrid_alltoall_any 5200 16.5 0.840 3.769 8.221 9.364
cp_fm_diag_elpa 83 13.4 0.000 0.000 9.230 9.241
cp_fm_diag_elpa_base 83 14.4 8.983 9.065 9.221 9.232
fft_wrap_pw1pw2_140 523 13.2 1.677 1.714 8.544 8.589
qs_ot_get_derivative_diag 77 12.4 0.002 0.003 8.029 8.100
grid_integrate_task_list 128 12.3 7.797 8.032 7.797 8.032
dbcsr_mm_accdrv_process 20762 16.1 3.297 4.543 7.225 7.907
cp_fm_cholesky_decompose 22 10.9 7.707 7.798 7.707 7.798
multiply_cannon_metrocomm1 10028 15.6 0.030 0.031 4.295 7.520
density_rs2pw 128 9.7 0.005 0.005 6.957 7.305
wfi_extrapolate 11 7.9 0.001 0.001 7.271 7.271
calculate_dm_sparse 128 9.5 0.001 0.001 6.048 6.127
fft3d_ps 1291 14.7 3.148 3.262 6.096 6.121
grid_collocate_task_list 128 9.7 5.571 5.824 5.571 5.824
multiply_cannon_metrocomm4 7521 15.6 0.026 0.029 1.872 5.689
dbcsr_complete_redistribute 395 12.7 2.112 2.186 5.283 5.660
mp_irecv_dv 28860 15.9 1.831 5.598 1.831 5.598
qs_energies_init_hamiltonians 11 5.9 0.009 0.011 5.328 5.329
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 5.119 5.133
mp_alltoall_d11v 2415 14.1 4.597 4.802 4.597 4.802
mp_allgather_i34 2507 14.6 1.307 4.559 1.307 4.559
potential_pw2rs 128 12.3 0.026 0.026 4.376 4.384
cp_dbcsr_sm_fm_multiply_core 37 10.5 0.000 0.000 4.021 4.070
build_core_hamiltonian_matrix_ 11 4.9 0.001 0.001 3.572 3.861
copy_fm_to_dbcsr 209 11.7 0.002 0.002 3.443 3.774
qs_ot_get_orbitals 117 10.6 0.001 0.001 3.637 3.692
qs_ot_get_derivative_taylor 40 13.0 0.001 0.001 3.588 3.679
copy_dbcsr_to_fm 186 11.8 0.004 0.004 3.423 3.491
qs_ks_update_qs_env_forces 11 4.9 0.000 0.000 3.459 3.472
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="408", plot="h2o_256_md", label="(8n/2r/6t)", y=169.003000, yerr=0.000000
PlotPoint: name="409", plot="h2o_256_md_mem", label="(8n/2r/6t)", y=1458.090909, yerr=33.255087
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/20/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 9 x 9 x 32 1420239992832 0.0% 0.0% 100.0%
flops 32 x 32 x 32 1943472701440 0.0% 0.0% 100.0%
flops 22 x 9 x 32 1972057190400 0.0% 0.0% 100.0%
flops 9 x 22 x 32 1977770336256 0.0% 0.0% 100.0%
flops 22 x 22 x 32 2734287699968 0.0% 0.0% 100.0%
flops 32 x 32 x 9 4416300122112 0.0% 0.0% 100.0%
flops 32 x 32 x 22 5397700149248 0.0% 0.0% 100.0%
flops 9 x 32 x 32 5443971710976 0.0% 0.0% 100.0%
flops 22 x 32 x 32 6653743202304 0.0% 0.0% 100.0%
flops 9 x 32 x 9 11528891191296 0.0% 0.0% 100.0%
flops 22 x 32 x 9 15129160814592 0.0% 0.0% 100.0%
flops 9 x 32 x 22 15129160814592 0.0% 0.0% 100.0%
flops 22 x 32 x 22 19767995056128 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 93.514751E+12 0.0% 0.0% 100.0%
flops max/rank 11.696233E+12 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 6755938624 0.0% 0.0% 100.0%
number of processed stacks 1964048 0.0% 0.0% 100.0%
average stack size 0.0 0.0 3439.8
marketing flops 144.579337E+12
-------------------------------------------------------------------------------
# multiplications 2507
max memory usage/rank 3.124339E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 100280
MPI messages size (bytes):
total size 1.136195E+12
min size 0.000000E+00
max size 104.857600E+06
average size 11.330227E+06
MPI breakdown and total messages size (bytes):
size <= 128 572 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 44 2883584
131072 < size <= 4194304 45208 35089547264
4194304 < size <= 16777216 44352 379752284160
16777216 < size 10104 721350092304
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 3992 59279.
MP_Allreduce 11057 1504.
MP_Sync 87
MP_Alltoall 1712 36974159.
MP_SendRecv 1792 218624.
MP_ISendRecv 1792 218624.
MP_Wait 9802
MP_ISend 6408 1080322.
MP_IRecv 6408 1080322.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.020 0.039 289.640 289.642
qs_mol_dyn_low 1 2.0 0.003 0.004 289.115 289.128
qs_forces 11 3.9 0.004 0.005 289.024 289.030
qs_energies 11 4.9 0.002 0.002 280.076 280.079
scf_env_do_scf 11 5.9 0.001 0.001 257.650 257.657
velocity_verlet 10 3.0 0.002 0.002 209.445 209.452
scf_env_do_scf_inner_loop 117 6.6 0.004 0.009 134.282 134.284
init_scf_loop 11 6.9 0.000 0.000 123.098 123.102
prepare_preconditioner 11 7.9 0.000 0.000 118.252 118.272
make_preconditioner 11 8.9 0.000 0.000 118.252 118.272
make_full_inverse_cholesky 11 9.9 0.038 0.039 94.695 115.398
qs_scf_new_mos 117 7.6 0.001 0.001 89.532 89.649
qs_scf_loop_do_ot 117 8.6 0.001 0.001 89.531 89.648
ot_scf_mini 117 9.6 0.004 0.004 84.679 84.723
dbcsr_multiply_generic 2507 12.6 0.216 0.226 81.729 82.271
cp_fm_upper_to_full 105 14.8 52.566 75.038 52.566 75.038
multiply_cannon 2507 13.6 0.686 0.730 58.625 59.832
multiply_cannon_loop 2507 14.6 1.043 1.077 54.594 56.010
ot_mini 117 10.6 0.001 0.001 44.451 44.520
dbcsr_complete_redistribute 395 12.7 4.000 4.082 29.548 42.070
copy_fm_to_dbcsr 209 11.7 0.001 0.002 26.049 38.600
transfer_fm_to_dbcsr 11 9.9 0.030 0.030 23.515 35.862
rebuild_ks_matrix 128 8.3 0.001 0.001 34.718 34.770
qs_ks_build_kohn_sham_matrix 128 9.3 0.017 0.017 34.717 34.770
mp_alltoall_i22 716 14.1 21.268 33.963 21.268 33.963
cp_fm_cholesky_invert 11 10.9 33.094 33.100 33.094 33.100
qs_ks_update_qs_env 128 7.6 0.001 0.001 32.198 32.248
mp_waitall_1 103674 16.8 27.914 32.107 27.914 32.107
qs_ot_get_p 128 10.4 0.001 0.001 25.398 25.494
qs_ot_get_derivative 117 11.6 0.002 0.002 24.447 24.489
qs_ot_p2m_diag 83 11.4 0.879 0.884 21.351 21.381
multiply_cannon_metrocomm3 10028 15.6 0.025 0.026 18.689 20.141
make_m2s 5014 13.6 0.074 0.077 18.550 20.050
ot_diis_step 117 11.6 0.022 0.022 19.967 19.967
make_images 5014 14.6 3.056 3.264 18.066 19.568
cp_dbcsr_syevd 83 12.4 0.006 0.006 19.556 19.558
apply_preconditioner_dbcsr 128 12.6 0.000 0.000 19.302 19.450
apply_single 128 13.6 0.001 0.001 19.302 19.450
multiply_cannon_multrec 10028 15.6 10.242 12.078 18.007 18.083
qs_rho_update_rho_low 128 7.7 0.001 0.001 17.299 17.328
calculate_rho_elec 128 8.7 0.478 0.478 17.298 17.327
sum_up_and_integrate 128 10.3 0.002 0.002 16.430 16.515
integrate_v_rspace 128 11.3 0.004 0.004 16.367 16.451
cp_fm_diag_elpa 83 13.4 0.000 0.000 16.385 16.386
cp_fm_diag_elpa_base 83 14.4 12.014 13.572 16.382 16.382
multiply_cannon_sync_h2d 10028 15.6 14.311 14.359 14.311 14.359
pw_transfer 1547 11.6 0.095 0.096 13.159 13.171
fft_wrap_pw1pw2 1291 12.7 0.011 0.012 12.921 12.934
hybrid_alltoall_any 5200 16.5 1.314 3.046 10.742 12.867
make_images_data 5014 15.6 0.063 0.068 10.538 12.718
init_scf_run 11 5.9 0.000 0.001 12.073 12.073
scf_env_initial_rho_setup 11 6.9 0.001 0.001 12.072 12.073
fft_wrap_pw1pw2_140 523 13.2 3.227 3.268 11.510 11.524
qs_ot_get_derivative_diag 77 12.4 0.002 0.002 9.665 9.694
dbcsr_mm_accdrv_process 20762 16.1 4.227 5.980 7.514 9.370
wfi_extrapolate 11 7.9 0.001 0.001 8.958 8.958
cp_fm_cholesky_decompose 22 10.9 8.828 8.845 8.828 8.845
grid_integrate_task_list 128 12.3 8.552 8.745 8.552 8.745
density_rs2pw 128 9.7 0.005 0.005 8.335 8.390
qs_energies_init_hamiltonians 11 5.9 0.002 0.002 8.169 8.170
fft3d_ps 1291 14.7 3.980 3.995 7.370 7.435
mp_alltoall_d11v 2415 14.1 6.871 7.029 6.871 7.029
calculate_dm_sparse 128 9.5 0.001 0.001 6.728 6.773
grid_collocate_task_list 128 9.7 6.415 6.491 6.415 6.491
cp_dbcsr_sm_fm_multiply 37 9.5 0.002 0.002 6.300 6.352
copy_dbcsr_to_fm 186 11.8 0.004 0.004 6.156 6.222
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="410", plot="h2o_256_md", label="(8n/1r/12t)", y=289.642000, yerr=0.000000
PlotPoint: name="411", plot="h2o_256_md_mem", label="(8n/1r/12t)", y=2862.181818, yerr=135.724198
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 1.260769E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 458208
MPI messages size (bytes):
total size 3.456111E+12
min size 0.000000E+00
max size 18.735064E+06
average size 7.542668E+06
MPI breakdown and total messages size (bytes):
size <= 128 112896 0
128 < size <= 8192 0 0
8192 < size <= 32768 224 5687808
32768 < size <= 131072 10528 813356544
131072 < size <= 4194304 36422 76284728544
4194304 < size <= 16777216 294266 3312457683808
16777216 < size 3872 66548597808
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 255646.
MP_Allreduce 3139 6114.
MP_Sync 4
MP_Alltoall 54 27994378.
MP_SendRecv 285 19200.
MP_ISendRecv 285 19200.
MP_Wait 1017
MP_ISend 642 197829.
MP_IRecv 642 197607.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.014 0.033 84.980 84.981
qs_energies 1 2.0 0.000 0.000 84.499 84.505
ls_scf 1 3.0 0.000 0.000 83.585 83.592
dbcsr_multiply_generic 111 6.7 0.018 0.029 72.534 72.742
multiply_cannon 111 7.7 0.017 0.020 55.816 57.051
multiply_cannon_loop 111 8.7 0.228 0.239 52.401 53.961
ls_scf_main 1 4.0 0.000 0.000 52.162 52.162
density_matrix_trs4 2 5.0 0.002 0.003 46.652 46.749
ls_scf_init_scf 1 4.0 0.000 0.000 28.386 28.388
ls_scf_init_matrix_S 1 5.0 0.000 0.000 27.259 27.316
mp_waitall_1 11031 10.9 22.476 25.239 22.476 25.239
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.080 25.099
multiply_cannon_multrec 2664 9.7 8.131 8.805 15.536 17.249
multiply_cannon_sync_h2d 2664 9.7 13.376 15.415 13.376 15.415
make_m2s 222 7.7 0.009 0.011 13.047 13.604
make_images 222 8.7 0.098 0.107 13.025 13.586
multiply_cannon_metrocomm1 2664 9.7 0.010 0.011 9.690 11.852
multiply_cannon_metrocomm3 2664 9.7 0.009 0.010 5.531 8.191
make_images_data 222 9.7 0.004 0.005 7.583 8.097
dbcsr_mm_accdrv_process 4760 10.4 0.592 0.707 7.024 8.027
hybrid_alltoall_any 227 10.6 0.215 1.830 6.548 7.658
dbcsr_mm_accdrv_process_sort 4760 11.4 6.233 7.133 6.233 7.133
calculate_norms 4752 9.8 5.508 6.101 5.508 6.101
apply_matrix_preconditioner 6 5.3 0.000 0.000 5.030 5.163
mp_sum_l 887 5.1 3.127 4.565 3.127 4.565
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 2.356 3.441
multiply_cannon_metrocomm4 2442 9.7 0.012 0.014 2.069 3.436
make_images_sizes 222 9.7 0.000 0.000 0.726 3.431
mp_alltoall_i44 222 10.7 0.726 3.431 0.726 3.431
mp_irecv_dv 6231 10.9 2.053 3.409 2.053 3.409
arnoldi_extremal 4 6.8 0.000 0.000 3.256 3.291
arnoldi_normal_ev 4 7.8 0.005 0.010 3.256 3.290
build_subspace 16 8.4 0.009 0.013 3.163 3.166
ls_scf_post 1 4.0 0.000 0.000 3.036 3.043
ls_scf_store_result 1 5.0 0.000 0.000 2.854 2.898
dbcsr_special_finalize 555 9.7 0.005 0.006 2.383 2.823
dbcsr_merge_single_wm 555 10.7 0.462 0.595 2.375 2.814
make_images_pack 222 9.7 2.208 2.585 2.210 2.587
dbcsr_matrix_vector_mult 304 9.0 0.006 0.014 2.330 2.570
dbcsr_sort_data 658 11.4 2.171 2.536 2.171 2.536
dbcsr_matrix_vector_mult_local 304 10.0 2.070 2.458 2.072 2.460
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.278 2.378
buffer_matrices_ensure_size 222 8.7 1.753 2.103 1.753 2.103
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.760 1.761
rebuild_ks_matrix 3 7.3 0.000 0.000 1.750 1.752
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.750 1.752
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="500", plot="h2o_32_nrep3_ls", label="(8n/12r/1t)", y=84.981000, yerr=0.000000
PlotPoint: name="501", plot="h2o_32_nrep3_ls_mem", label="(8n/12r/1t)", y=1132.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 2.175656E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 106560
MPI messages size (bytes):
total size 2.699093E+12
min size 0.000000E+00
max size 72.286792E+06
average size 25.329324E+06
MPI breakdown and total messages size (bytes):
size <= 128 23040 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 3264 325830144
131072 < size <= 4194304 5280 3328561104
4194304 < size <= 16777216 12709 156766962056
16777216 < size 62267 2538670978840
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 10075.
MP_Sync 4
MP_Alltoall 47 15335933.
MP_SendRecv 141 57600.
MP_ISendRecv 141 57600.
MP_Wait 687
MP_ISend 462 414589.
MP_IRecv 462 413870.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.023 0.042 89.385 89.386
qs_energies 1 2.0 0.000 0.000 88.954 88.963
ls_scf 1 3.0 0.000 0.000 87.622 87.631
dbcsr_multiply_generic 111 6.7 0.015 0.015 73.829 74.181
multiply_cannon 111 7.7 0.027 0.037 52.741 56.602
ls_scf_main 1 4.0 0.000 0.000 54.176 54.181
multiply_cannon_loop 111 8.7 0.135 0.147 49.929 52.656
density_matrix_trs4 2 5.0 0.002 0.003 48.547 48.761
ls_scf_init_scf 1 4.0 0.000 0.000 29.740 29.742
mp_waitall_1 9105 10.9 20.511 29.736 20.511 29.736
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.572 28.655
multiply_cannon_multrec 1332 9.7 13.375 17.007 22.686 27.436
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.218 26.231
multiply_cannon_metrocomm3 1332 9.7 0.007 0.008 11.483 20.204
make_m2s 222 7.7 0.006 0.007 14.885 15.472
make_images 222 8.7 1.364 1.691 14.854 15.442
dbcsr_mm_accdrv_process 4041 10.4 0.356 0.533 8.906 10.471
dbcsr_mm_accdrv_process_sort 4041 11.4 8.423 9.937 8.423 9.937
make_images_data 222 9.7 0.004 0.005 8.578 9.452
hybrid_alltoall_any 227 10.6 0.541 2.532 7.962 9.379
multiply_cannon_metrocomm4 1221 9.7 0.007 0.009 3.200 7.713
mp_irecv_dv 3311 11.0 3.180 7.653 3.180 7.653
mp_sum_l 887 5.1 4.716 7.074 4.716 7.074
calculate_norms 2376 9.8 6.067 6.704 6.067 6.704
multiply_cannon_sync_h2d 1332 9.7 4.902 6.237 4.902 6.237
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.547 5.676
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.867 5.080
arnoldi_extremal 4 6.8 0.000 0.000 4.618 4.636
arnoldi_normal_ev 4 7.8 0.001 0.005 4.618 4.635
build_subspace 16 8.4 0.014 0.021 4.364 4.368
ls_scf_post 1 4.0 0.000 0.000 3.705 3.714
ls_scf_store_result 1 5.0 0.000 0.000 3.410 3.530
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.164 3.388
dbcsr_matrix_vector_mult_local 304 10.0 2.782 3.258 2.784 3.259
mp_allgather_i34 111 8.7 0.809 2.985 0.809 2.985
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.586 2.666
multiply_cannon_metrocomm1 1332 9.7 0.003 0.004 1.098 2.535
dbcsr_data_new 4174 10.1 2.111 2.398 2.111 2.398
dbcsr_sort_data 436 11.2 1.839 2.126 1.839 2.126
make_images_pack 222 9.7 1.817 2.117 1.820 2.119
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.842 1.845
rebuild_ks_matrix 3 7.3 0.000 0.000 1.830 1.832
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.830 1.832
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="502", plot="h2o_32_nrep3_ls", label="(8n/6r/2t)", y=89.386000, yerr=0.000000
PlotPoint: name="503", plot="h2o_32_nrep3_ls_mem", label="(8n/6r/2t)", y=1779.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 2.852221E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 46176
MPI messages size (bytes):
total size 1.924064E+12
min size 0.000000E+00
max size 108.059888E+06
average size 41.668048E+06
MPI breakdown and total messages size (bytes):
size <= 128 9984 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 3328 1170063360
4194304 < size <= 16777216 1870 19378539600
16777216 < size 30994 1903514987232
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265448.
MP_Allreduce 3138 10896.
MP_Sync 4
MP_Alltoall 47 23526250.
MP_SendRecv 93 57600.
MP_ISendRecv 93 57600.
MP_Wait 639
MP_ISend 462 560046.
MP_IRecv 462 560662.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.027 0.049 92.663 92.663
qs_energies 1 2.0 0.000 0.000 92.115 92.120
ls_scf 1 3.0 0.000 0.000 90.712 90.717
dbcsr_multiply_generic 111 6.7 0.016 0.021 75.315 75.578
ls_scf_main 1 4.0 0.000 0.000 56.649 56.653
multiply_cannon 111 7.7 0.029 0.075 51.971 55.617
multiply_cannon_loop 111 8.7 0.117 0.130 49.216 53.754
density_matrix_trs4 2 5.0 0.002 0.003 50.661 50.891
mp_waitall_1 7281 11.0 23.505 33.870 23.505 33.870
ls_scf_init_scf 1 4.0 0.000 0.000 30.387 30.390
ls_scf_init_matrix_S 1 5.0 0.000 0.000 29.174 29.269
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 26.836 26.849
multiply_cannon_multrec 888 9.7 12.651 15.569 21.244 24.759
multiply_cannon_metrocomm3 888 9.7 0.004 0.004 11.086 23.432
make_m2s 222 7.7 0.006 0.007 16.160 16.936
make_images 222 8.7 1.581 1.857 16.121 16.896
make_images_data 222 9.7 0.004 0.005 9.535 10.518
hybrid_alltoall_any 227 10.6 0.639 2.942 9.086 10.182
dbcsr_mm_accdrv_process 3754 10.4 0.300 0.490 8.121 9.376
mp_sum_l 887 5.1 5.484 8.903 5.484 8.903
dbcsr_mm_accdrv_process_sort 3754 11.4 7.686 8.886 7.686 8.886
multiply_cannon_sync_h2d 888 9.7 6.077 7.481 6.077 7.481
multiply_cannon_metrocomm4 777 9.7 0.004 0.005 2.496 7.043
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 4.296 7.004
mp_irecv_dv 2335 11.1 2.479 6.998 2.479 6.998
multiply_cannon_metrocomm1 888 9.7 0.003 0.003 3.524 6.960
arnoldi_extremal 4 6.8 0.000 0.000 5.160 5.175
arnoldi_normal_ev 4 7.8 0.001 0.005 5.160 5.175
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.813 5.097
build_subspace 16 8.4 0.014 0.020 4.862 4.869
calculate_norms 1584 9.8 4.327 4.750 4.327 4.750
dbcsr_matrix_vector_mult 304 9.0 0.010 0.021 3.507 3.834
mp_allgather_i34 111 8.7 0.880 3.716 0.880 3.716
ls_scf_post 1 4.0 0.000 0.000 3.677 3.681
dbcsr_matrix_vector_mult_local 304 10.0 3.072 3.661 3.074 3.663
ls_scf_store_result 1 5.0 0.000 0.000 3.435 3.504
ls_scf_dm_to_ks 2 5.0 0.000 0.000 2.881 3.012
dbcsr_data_new 4116 9.9 2.113 2.457 2.113 2.457
make_images_sizes 222 9.7 0.000 0.000 0.811 2.371
mp_alltoall_i44 222 10.7 0.811 2.370 0.811 2.370
dbcsr_sort_data 325 11.1 1.863 2.106 1.863 2.106
qs_ks_update_qs_env 3 6.3 0.000 0.000 1.887 1.889
make_images_pack 222 9.7 1.620 1.880 1.622 1.883
rebuild_ks_matrix 3 7.3 0.000 0.000 1.870 1.871
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 1.870 1.871
dbcsr_finalize 304 7.8 0.026 0.032 1.603 1.861
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="504", plot="h2o_32_nrep3_ls", label="(8n/4r/3t)", y=92.663000, yerr=0.000000
PlotPoint: name="505", plot="h2o_32_nrep3_ls_mem", label="(8n/4r/3t)", y=2216.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 3.362456E+09
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 50616
MPI messages size (bytes):
total size 1.536549E+12
min size 0.000000E+00
max size 72.286792E+06
average size 30.356986E+06
MPI breakdown and total messages size (bytes):
size <= 128 10368 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 1056 104411904
131072 < size <= 4194304 3168 831638784
4194304 < size <= 16777216 3103 33613273640
16777216 < size 32921 1501999894888
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 266673.
MP_Allreduce 3138 13030.
MP_Sync 4
MP_Alltoall 47 30278988.
MP_SendRecv 69 86400.
MP_ISendRecv 69 86400.
MP_Wait 531
MP_ISend 378 823502.
MP_IRecv 378 823753.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.030 0.047 96.816 96.817
qs_energies 1 2.0 0.000 0.000 96.220 96.225
ls_scf 1 3.0 0.000 0.000 94.564 94.569
dbcsr_multiply_generic 111 6.7 0.016 0.017 78.220 78.466
ls_scf_main 1 4.0 0.000 0.000 58.838 58.839
multiply_cannon 111 7.7 0.042 0.081 51.667 56.184
density_matrix_trs4 2 5.0 0.002 0.003 52.690 52.834
multiply_cannon_loop 111 8.7 0.153 0.168 46.576 49.278
ls_scf_init_scf 1 4.0 0.000 0.000 32.484 32.487
ls_scf_init_matrix_S 1 5.0 0.000 0.000 31.269 31.346
mp_waitall_1 6369 11.0 22.515 29.086 22.515 29.086
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 28.809 28.821
multiply_cannon_multrec 1332 9.7 14.194 17.488 22.122 24.858
make_m2s 222 7.7 0.007 0.008 20.978 22.457
make_images 222 8.7 3.131 3.595 20.928 22.409
multiply_cannon_metrocomm3 1332 9.7 0.003 0.004 9.137 16.644
make_images_data 222 9.7 0.004 0.004 11.715 13.297
hybrid_alltoall_any 227 10.6 0.798 3.775 11.046 13.132
dbcsr_mm_accdrv_process 3641 10.4 0.297 0.475 7.573 9.104
dbcsr_mm_accdrv_process_sort 3641 11.4 7.131 8.606 7.131 8.606
mp_sum_l 887 5.1 4.002 7.615 4.002 7.615
multiply_cannon_metrocomm4 1110 9.7 0.005 0.007 2.084 6.063
multiply_cannon_sync_h2d 1332 9.7 5.472 6.005 5.472 6.005
mp_irecv_dv 3229 10.9 2.059 5.986 2.059 5.986
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 3.046 5.961
arnoldi_extremal 4 6.8 0.000 0.000 5.309 5.321
arnoldi_normal_ev 4 7.8 0.001 0.005 5.309 5.321
multiply_cannon_metrocomm1 1332 9.7 0.003 0.003 2.494 5.066
build_subspace 16 8.4 0.014 0.021 4.967 4.977
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.565 4.746
mp_allgather_i34 111 8.7 2.209 4.729 2.209 4.729
calculate_norms 2376 9.8 4.167 4.513 4.167 4.513
dbcsr_matrix_vector_mult 304 9.0 0.010 0.022 3.650 3.969
dbcsr_matrix_vector_mult_local 304 10.0 3.250 3.747 3.252 3.749
dbcsr_sort_data 658 11.4 3.089 3.454 3.089 3.454
ls_scf_post 1 4.0 0.000 0.000 3.241 3.248
dbcsr_special_finalize 555 9.7 0.006 0.007 2.847 3.222
dbcsr_merge_single_wm 555 10.7 0.544 0.665 2.838 3.214
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.022 3.092
ls_scf_store_result 1 5.0 0.000 0.000 2.985 3.045
dbcsr_data_release 10477 10.7 1.566 2.380 1.566 2.380
dbcsr_finalize 304 7.8 0.049 0.061 1.792 1.979
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="506", plot="h2o_32_nrep3_ls", label="(8n/3r/4t)", y=96.817000, yerr=0.000000
PlotPoint: name="507", plot="h2o_32_nrep3_ls_mem", label="(8n/3r/4t)", y=2762.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 4.751692E+09
# max total images/rank 1
# max 3D layers 1
# MPI messages exchanged 10656
MPI messages size (bytes):
total size 1.149035E+12
min size 0.000000E+00
max size 203.538048E+06
average size 107.829832E+06
MPI breakdown and total messages size (bytes):
size <= 128 2304 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 768 702038016
4194304 < size <= 16777216 0 0
16777216 < size 7584 1148332810224
-------------------------------------------------------------------------------
- -
- DBCSR MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Bcast 2 12.
MP_Allreduce 705 128.
MP_Alltoall 310 12920694.
MP_ISend 1776 40180424.
MP_IRecv 1776 40465030.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 265536.
MP_Allreduce 3129 15263.
MP_Sync 4
MP_Alltoall 47 46208988.
MP_SendRecv 45 115200.
MP_ISendRecv 45 115200.
MP_Wait 528
MP_ISend 420 924980.
MP_IRecv 420 924528.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.038 0.057 91.598 91.598
qs_energies 1 2.0 0.000 0.000 90.849 90.861
ls_scf 1 3.0 0.000 0.000 88.902 88.915
dbcsr_multiply_generic 111 6.7 0.017 0.018 70.385 70.547
ls_scf_main 1 4.0 0.000 0.000 56.193 56.193
multiply_cannon 111 7.7 0.081 0.141 52.388 55.265
multiply_cannon_loop 111 8.7 0.088 0.093 49.806 50.881
density_matrix_trs4 2 5.0 0.002 0.003 49.192 49.293
ls_scf_init_scf 1 4.0 0.000 0.000 29.324 29.327
mp_waitall_1 5436 11.0 24.010 28.623 24.010 28.623
ls_scf_init_matrix_S 1 5.0 0.000 0.000 28.065 28.092
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 25.984 25.995
multiply_cannon_multrec 444 9.7 13.758 16.494 20.861 22.337
multiply_cannon_metrocomm1 444 9.7 0.002 0.002 10.435 15.073
make_m2s 222 7.7 0.005 0.005 13.368 14.313
make_images 222 8.7 2.039 2.480 13.301 14.246
multiply_cannon_metrocomm3 444 9.7 0.001 0.002 6.279 13.986
make_images_data 222 9.7 0.003 0.004 8.114 9.670
hybrid_alltoall_any 227 10.6 0.803 3.841 8.047 9.457
dbcsr_mm_accdrv_process 3003 10.4 0.363 0.408 6.792 7.897
multiply_cannon_sync_h2d 444 9.7 6.777 7.737 6.777 7.737
dbcsr_mm_accdrv_process_sort 3003 11.4 6.429 7.502 6.429 7.502
arnoldi_extremal 4 6.8 0.000 0.000 5.897 5.911
arnoldi_normal_ev 4 7.8 0.002 0.005 5.897 5.911
build_subspace 16 8.4 0.015 0.020 5.499 5.507
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.415 4.698
dbcsr_matrix_vector_mult 304 9.0 0.011 0.021 4.238 4.453
dbcsr_matrix_vector_mult_local 304 10.0 3.777 4.268 3.779 4.270
mp_sum_l 887 5.1 2.619 4.179 2.619 4.179
multiply_cannon_metrocomm4 333 9.7 0.001 0.002 1.500 3.780
mp_irecv_dv 1241 11.2 1.485 3.737 1.485 3.737
calculate_norms 792 9.8 3.617 3.708 3.617 3.708
ls_scf_dm_to_ks 2 5.0 0.000 0.000 3.551 3.618
mp_allgather_i34 111 8.7 1.157 3.525 1.157 3.525
ls_scf_post 1 4.0 0.000 0.000 3.385 3.397
dbcsr_multiply_generic_mpsum_f 86 7.8 0.000 0.000 1.830 3.312
make_images_sizes 222 9.7 0.000 0.000 0.868 3.227
mp_alltoall_i44 222 10.7 0.868 3.227 0.868 3.227
ls_scf_store_result 1 5.0 0.000 0.000 3.172 3.203
dbcsr_finalize 304 7.8 0.062 0.077 2.204 2.303
dbcsr_data_new 4608 9.7 1.786 2.215 1.786 2.215
dbcsr_merge_all 275 8.9 0.480 0.526 2.061 2.134
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.046 2.046
rebuild_ks_matrix 3 7.3 0.000 0.000 2.013 2.014
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.001 2.013 2.014
qs_energies_init_hamiltonians 1 3.0 0.001 0.002 1.931 1.931
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="508", plot="h2o_32_nrep3_ls", label="(8n/2r/6t)", y=91.598000, yerr=0.000000
PlotPoint: name="509", plot="h2o_32_nrep3_ls_mem", label="(8n/2r/6t)", y=3735.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_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 8.836071E+09
# max total images/rank 2
# max 3D layers 1
# MPI messages exchanged 4440
MPI messages size (bytes):
total size 770.525954E+09
min size 0.000000E+00
max size 399.069120E+06
average size 173.541888E+06
MPI breakdown and total messages size (bytes):
size <= 128 640 0
128 < size <= 8192 0 0
8192 < size <= 32768 0 0
32768 < size <= 131072 0 0
131072 < size <= 4194304 640 468025344
4194304 < size <= 16777216 0 0
16777216 < size 3160 770057961712
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 1026 284089.
MP_Allreduce 3123 21388.
MP_Sync 4
MP_Alltoall 47 88727262.
MP_SendRecv 42 732600.
MP_ISendRecv 42 732600.
MP_Wait 267
MP_ISend 180 3337386.
MP_IRecv 180 3339494.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.072 0.089 105.614 105.614
qs_energies 1 2.0 0.000 0.000 104.192 104.200
ls_scf 1 3.0 0.000 0.000 101.215 101.222
dbcsr_multiply_generic 111 6.7 0.023 0.026 74.678 74.830
ls_scf_main 1 4.0 0.000 0.000 63.826 63.826
density_matrix_trs4 2 5.0 0.002 0.003 54.795 54.853
multiply_cannon 111 7.7 0.141 0.194 48.249 50.372
multiply_cannon_loop 111 8.7 0.097 0.100 45.295 45.863
ls_scf_init_scf 1 4.0 0.000 0.000 33.600 33.601
ls_scf_init_matrix_S 1 5.0 0.000 0.000 32.040 32.063
matrix_sqrt_Newton_Schulz 2 6.5 0.001 0.001 29.352 29.361
mp_waitall_1 4527 11.1 21.800 25.256 21.800 25.256
make_m2s 222 7.7 0.005 0.005 22.703 23.768
make_images 222 8.7 3.565 3.859 22.595 23.658
multiply_cannon_multrec 444 9.7 17.820 18.467 22.482 23.120
hybrid_alltoall_any 227 10.6 1.653 3.640 12.892 15.812
make_images_data 222 9.7 0.003 0.004 13.120 15.186
multiply_cannon_metrocomm3 444 9.7 0.001 0.001 10.116 10.503
multiply_cannon_sync_h2d 444 9.7 8.791 8.856 8.791 8.856
arnoldi_extremal 4 6.8 0.000 0.000 7.502 7.514
arnoldi_normal_ev 4 7.8 0.003 0.009 7.502 7.514
build_subspace 16 8.4 0.026 0.036 6.953 6.964
dbcsr_matrix_vector_mult 304 9.0 0.017 0.034 5.605 5.771
dbcsr_matrix_vector_mult_local 304 10.0 5.167 5.487 5.170 5.489
ls_scf_dm_to_ks 2 5.0 0.000 0.000 5.186 5.283
apply_matrix_preconditioner 6 5.3 0.000 0.000 4.826 5.084
dbcsr_mm_accdrv_process 1814 10.4 0.291 0.355 4.475 4.598
dbcsr_mm_accdrv_process_sort 1814 11.4 4.138 4.268 4.138 4.268
ls_scf_post 1 4.0 0.000 0.000 3.789 3.797
ls_scf_store_result 1 5.0 0.000 0.000 3.511 3.549
make_images_sizes 222 9.7 0.000 0.000 1.468 3.529
mp_alltoall_i44 222 10.7 1.468 3.529 1.468 3.529
mp_allgather_i34 111 8.7 1.071 3.399 1.071 3.399
calculate_norms 792 9.8 3.231 3.268 3.231 3.268
dbcsr_finalize 304 7.8 0.082 0.089 3.073 3.157
qs_energies_init_hamiltonians 1 3.0 0.001 0.002 2.947 2.947
dbcsr_merge_all 275 8.9 0.884 0.916 2.858 2.937
dbcsr_complete_redistribute 5 7.6 1.443 1.484 2.755 2.878
matrix_ls_to_qs 2 6.0 0.000 0.000 2.402 2.534
dbcsr_sort_data 325 11.1 2.437 2.490 2.437 2.490
qs_ks_update_qs_env 3 6.3 0.000 0.000 2.465 2.467
rebuild_ks_matrix 3 7.3 0.000 0.000 2.399 2.401
qs_ks_build_kohn_sham_matrix 3 8.3 0.000 0.000 2.399 2.401
dbcsr_data_new 6591 9.6 1.859 2.370 1.859 2.370
dbcsr_new_transposed 4 7.5 0.243 0.253 2.281 2.298
dbcsr_frobenius_norm 74 6.6 2.059 2.138 2.205 2.238
dbcsr_add_d 103 6.2 0.000 0.000 2.125 2.202
dbcsr_add_anytype 103 7.2 0.859 0.890 2.124 2.202
dbcsr_data_release 12724 10.6 1.971 2.175 1.971 2.175
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="510", plot="h2o_32_nrep3_ls", label="(8n/1r/12t)", y=105.614000, yerr=0.000000
PlotPoint: name="511", plot="h2o_32_nrep3_ls_mem", label="(8n/1r/12t)", y=6944.000000, yerr=0.000000
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
~~~~~~~~~ RESULT ~~~~~~~~~
RESULT file: /scratch/snx3000/mkrack/rt/../rt/CRAY-XC50-gnu/183bfa028a10482cb2ce2952b7487d62cedfbee6_performance_tests/27/result.log
@@@@@@@@@@ Run number: 1 @@@@@@@@@@
-------------------------------------------------------------------------------
- -
- DBCSR STATISTICS -
- -
-------------------------------------------------------------------------------
COUNTER TOTAL BLAS SMM ACC
flops 32 x 32 x 32 7009386627072 0.0% 0.0% 100.0%
flops 9 x 9 x 32 7335108845568 0.0% 0.0% 100.0%
flops 9 x 22 x 32 9866241589248 0.0% 0.0% 100.0%
flops 22 x 9 x 32 9884108906496 0.0% 0.0% 100.0%
flops 22 x 22 x 32 13354440523776 0.0% 0.0% 100.0%
flops 32 x 32 x 9 20607185977344 0.0% 0.0% 100.0%
flops 32 x 32 x 22 25186560638976 0.0% 0.0% 100.0%
flops 9 x 32 x 32 28458319085568 0.0% 0.0% 100.0%
flops 22 x 32 x 32 34782389993472 0.0% 0.0% 100.0%
flops 9 x 32 x 9 42881542373376 0.0% 0.0% 100.0%
flops 22 x 32 x 9 55680402235392 0.0% 0.0% 100.0%
flops 9 x 32 x 22 55680402235392 0.0% 0.0% 100.0%
flops 22 x 32 x 22 72328573419520 0.0% 0.0% 100.0%
flops inhomo. stacks 0 0.0% 0.0% 0.0%
flops total 383.054662E+12 0.0% 0.0% 100.0%
flops max/rank 733.641090E+09 0.0% 0.0% 100.0%
matmuls inhomo. stacks 0 0.0% 0.0% 0.0%
matmuls total 26899403712 0.0% 0.0% 100.0%
number of processed stacks 118860288 0.0% 0.0% 100.0%
average stack size 0.0 0.0 226.3
marketing flops 780.439111E+12
-------------------------------------------------------------------------------
# multiplications 1445
max memory usage/rank 593.256448E+06
# max total images/rank 3
# max 3D layers 1
# MPI messages exchanged 102097920
MPI messages size (bytes):
total size 37.227590E+12
min size 0.000000E+00
max size 4.551360E+06
average size 364.626312E+03
MPI breakdown and total messages size (bytes):
size <= 128 731472 0
128 < size <= 8192 11922720 97670922240
8192 < size <= 32768 24718992 614677610496
32768 < size <= 131072 20000256 1970081366016
131072 < size <= 4194304 42515668 24886801223040
4194304 < size <= 16777216 2208812 9656099886720
16777216 < size 0 0
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- MESSAGE PASSING PERFORMANCE -
- -
-------------------------------------------------------------------------------
ROUTINE CALLS AVE VOLUME [Bytes]
MP_Group 4
MP_Bcast 4640 78072.
MP_Allreduce 13232 2081.
MP_Sync 1064
MP_Alltoall 2588 4222134.
MP_SendRecv 168740 11136.
MP_ISendRecv 92040 11136.
MP_Wait 102830
MP_comm_split 40
MP_ISend 26090 85106.
MP_IRecv 37890 59644.
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
- -
- T I M I N G -
- -
-------------------------------------------------------------------------------
SUBROUTINE CALLS ASD SELF TIME TOTAL TIME
MAXIMUM AVERAGE MAXIMUM AVERAGE MAXIMUM
CP2K 1 1.0 0.042 0.116 222.713 222.715
qs_mol_dyn_low 1 2.0 0.005 0.017 221.461 221.513
qs_forces 5 3.8 0.006 0.058 221.323 221.338
qs_energies 5 4.8 0.003 0.031 218.275 218.301
scf_env_do_scf 5 5.8 0.001 0.009 203.176 203.179
scf_env_do_scf_inner_loop 105 6.6 0.002 0.006 176.963 176.968
qs_scf_new_mos 105 7.6 0.000 0.001 138.411 138.569
qs_scf_loop_do_ot 105 8.6 0.001 0.001 138.411 138.569
dbcsr_multiply_generic 1445 12.2 0.129 0.139 132.397 132.849
ot_scf_mini 105 9.6 0.003 0.004 128.533 128.698
multiply_cannon 1445 13.2 0.275 0.284 114.044 116.275
multiply_cannon_loop 1445 14.2 2.844 3.026 112.458 113.858
velocity_verlet 4 3.0 0.004 0.018 104.656 104.658
ot_mini 105 10.6 0.001 0.003 59.253 59.371
multiply_cannon_multrec 69360 15.2 29.729 34.307 39.707 44.621
mp_waitall_1 488190 16.1 33.717 40.652 33.717 40.652
qs_ot_get_p 112 10.4 0.001 0.003 39.529 39.798
qs_ot_get_derivative 55 11.6 0.001 0.001 37.621 37.739
multiply_cannon_sync_h2d 69360 15.2 29.218 33.574 29.218 33.574
multiply_cannon_metrocomm3 69360 15.2 0.200 0.211 25.251 32.262
qs_ot_p2m_diag 40 11.0 0.020 0.033 28.659 28.724
rebuild_ks_matrix 110 8.4 0.000 0.000 28.315 28.478
qs_ks_build_kohn_sham_matrix 110 9.4 0.013 0.040 28.315 28.478
init_scf_loop 7 6.6 0.000 0.005 26.184 26.185
qs_ks_update_qs_env 112 7.6 0.001 0.001 25.875 26.029
cp_dbcsr_syevd 40 12.0 0.002 0.003 25.464 25.465
apply_preconditioner_dbcsr 62 12.6 0.000 0.001 22.919 23.144
apply_single 62 13.6 0.000 0.000 22.919 23.144
prepare_preconditioner 7 7.6 0.000 0.000 21.413 21.441
make_preconditioner 7 8.6 0.000 0.003 21.413 21.441
ot_new_cg_direction 55 11.6 0.001 0.002 20.947 20.947
cp_fm_syevd 40 13.0 0.000 0.002 20.508 20.659
qs_rho_update_rho_low 110 7.6 0.001 0.001 15.944 16.330
calculate_rho_elec 110 8.6 0.030 0.032 15.943 16.329
cp_fm_redistribute_end 40 14.0 7.954 15.870 7.958 15.872
cp_fm_syevd_base 40 14.0 7.908 15.827 7.908 15.827
qs_ot_get_orbitals 105 10.6 0.001 0.001 14.613 14.771
make_full_inverse_cholesky 7 9.6 0.000 0.001 14.390 14.458
qs_ot_get_derivative_taylor 37 12.8 0.001 0.001 13.607 13.722
mp_sum_l 4764 12.2 11.827 12.847 11.827 12.847
init_scf_run 5 5.8 0.000 0.001 12.430 12.431
scf_env_initial_rho_setup 5 6.8 0.002 0.007 12.430 12.431
calculate_dm_sparse 110 9.5 0.000 0.001 11.499 11.652
pw_transfer 1645 12.4 0.079 0.095 11.279 11.452
fft_wrap_pw1pw2 1425 13.5 0.012 0.014 11.141 11.318
density_rs2pw 110 9.6 0.005 0.006 10.353 10.926
dbcsr_mm_accdrv_process 154766 15.8 6.267 6.486 9.846 10.661
qs_vxc_create 110 10.4 0.002 0.008 10.150 10.184
cp_fm_cholesky_invert 7 10.6 10.161 10.168 10.161 10.168
qs_ot_get_derivative_diag 18 12.0 0.000 0.001 10.082 10.158
fft_wrap_pw1pw2_240 915 15.0 0.839 0.917 9.782 10.000
check_diag 80 13.5 8.595 8.886 9.274 9.424
acc_transpose_blocks 69360 15.2 0.359 0.373 7.976 8.376
sum_up_and_integrate 60 10.3 0.001 0.003 8.218 8.251
integrate_v_rspace 60 11.3 0.001 0.002 8.201 8.234
fft3d_pb 915 16.0 2.402 2.576 7.952 8.121
transfer_rs2pw 445 10.6 0.007 0.008 7.219 7.787
cp_dbcsr_sm_fm_multiply 15 9.3 0.001 0.002 7.697 7.713
calculate_first_density_matrix 1 7.0 0.000 0.004 7.667 7.682
xc_rho_set_and_dset_create 110 12.4 0.076 0.097 7.168 7.425
cp_dbcsr_sm_fm_multiply_core 15 10.3 0.000 0.000 7.303 7.350
multiply_cannon_metrocomm1 69360 15.2 0.093 0.100 4.372 7.020
make_full_single_inverse 7 9.6 0.001 0.004 6.760 6.789
xc_vxc_pw_create 60 11.3 0.039 0.049 6.751 6.784
make_m2s 2890 13.2 0.078 0.089 6.057 6.611
make_images 2890 14.2 0.236 0.257 5.949 6.502
xc_pw_derive 510 13.4 0.006 0.007 5.935 5.999
acc_transpose_blocks_kernels 69360 16.2 0.837 0.886 5.214 5.491
mp_alltoall_z22v 2340 17.7 4.918 5.231 4.918 5.231
mp_waitany 7680 13.5 4.511 5.097 4.511 5.097
jit_kernel_transpose 5 15.0 4.378 4.619 4.378 4.619
potential_pw2rs 60 12.3 0.002 0.005 4.511 4.549
multiply_cannon_metrocomm4 67915 15.2 0.186 0.201 2.007 4.535
-------------------------------------------------------------------------------
~ ~ ~ ~ DATA POINTS ~ ~ ~ ~
PlotPoint: name="601", plot="h2o_512_md", label="(64n/12r/1t)", y=222.715000, yerr=0.000000
PlotPoint: name="602", plot="h2o_512_md_mem", label="(64n/12r/1t)", y=563.800000, yerr=1.833030
~ ~ ~ ~ END DATA POINTS ~ ~ ~
~~~~~~ END RESULT ~~~~~~~~
========= END RESULTS ===========
CommitSHA: 183bfa028a10482cb2ce2952b7487d62cedfbee6
Summary: empty
Status: OK