# Invocation command line:
# /home/HPC2021v1.1.8/bin/harness/runhpc --reportable --config=2Ns_icx_I204_IMPI204.cfg --tune=base --pmodel=OMP --define model=omp --threads=32 --rank=24 --define NRNK=12 --size=ref --iterations=3 tiny
# output_root was not used for this run
############################################################################

# The following setting was inserted automatically as a result of
# post-run basepeak application.
basepeak = 1

build_in_build_dir=0        # build in run dir

strict_rundir_verify = 1
%ifndef %{label}         # IF acctype is not set use mpi
%   define label icx_I204_IMPI204
%endif

%ifndef %{model}         # IF acctype is not set use mpi
%   define model mpi
%endif

teeout = yes
makeflags=-j 40

flagsurl000=http://www.spec.org/hpc2021/flags/Intel_compiler_flags.2024-12-31.00.xml

######################################################################
# The header section of the config file.  Must appear
# before any instances of "section markers" (see below)
#
# ext = how the binaries you generated will be identified
# tune = specify "base" or "peak" or "all"

label         = %{label}_%{model}
tune          = base
output_format = text
use_submit_for_speed = 1

default:
AR           = ar
ARFLAGS      = cr
CC           = mpiicc -cc=icx
CXX          = mpiicpc -cxx=icx
FC           = mpiifort  -fc=ifx
system_class = Homogenous
sw_compiler  = Intel C/C++/Fortran Compiler 2022.1.0
sw_mpi_library = Intel MPI Library for Linux OS, Build 20240701

test_sponsor = Lenovo Global Technology
license_num = 28
showtimer = 0
tester = Lenovo Global Technology
hw_avail     = Feb-2025
sw_avail     = Feb-2025
prepared_by = Lenovo Global Technology
system_vendor = Lenovo Global Technology
system_name = ThinkSystem SD535 V3 (AMD EPYC 9965)
sw_os_list = Ubuntu 24.04 LTSKernel 6.8.0-39-generic
hw_vendor_list = Lenovo Global Technology
hw_total_accel = 0
hw_model_list = ThinkSystem SD535 V3
hw_cpu_name_list = AMD EPYC 9965
node_compute_syslbl = ThinkSystem SD535 V3
node_compute_sw_state = Multi-user, run level 3
node_compute_sw_sharedfile = None
node_compute_sw_other = None
node_compute_sw_localfile = xfs
node_compute_purpose = Compute
node_compute_order = 1
node_compute_hw_vendor = Lenovo Global Technology
node_compute_hw_scache = 1 MB I+D on chip per core
node_compute_hw_pcache = 32 KB I + 48 KB D on chip per core
node_compute_hw_other = None
node_compute_hw_ocache = None
node_compute_hw_nthreadspercore = 2
node_compute_hw_ncpuorder = 1 chips
node_compute_hw_ncoresperchip = 192
node_compute_hw_ncores = 192
node_compute_hw_nchips = 1
node_compute_hw_model = ThinkSystem SD535 V3
node_compute_hw_disk = 1x ThinkSystem 2.5" 5300 480GB SSD
node_compute_hw_cpu_name = AMD EPYC 9965
node_compute_hw_cpu_mhz = 2250
node_compute_hw_cpu_char = Max Boost Clock up to 3.7 GHz
node_compute_hw_adapter_fs_slot_type = PCI-Express 5.0 x16
node_compute_hw_adapter_fs_ports_used = 1
node_compute_hw_adapter_fs_model = ConnectX-7
node_compute_hw_adapter_fs_interconnect = ConnectX-7 NDR
node_compute_hw_adapter_fs_firmware = 28.41.1000
node_compute_hw_adapter_fs_driver = 24.07-0.6.1
node_compute_hw_adapter_fs_data_rate = 400 Gb/s
node_compute_hw_adapter_fs_count = 1
node_compute_count = 2
interconnect_fs_syslbl = -
interconnect_fs_purpose = MPI Traffic
interconnect_fs_order = 0
interconnect_fs_hw_vendor = Nvidia
interconnect_fs_hw_topo = Direct Connect
interconnect_fs_hw_switch_fs_ports = 0
interconnect_fs_hw_switch_fs_model = None
interconnect_fs_hw_switch_fs_firmware = N/A
interconnect_fs_hw_switch_fs_data_rate = N/A
interconnect_fs_hw_switch_fs_count = 0
interconnect_fs_hw_model = Mellanox NDR

CC_VERSION_OPTION  = -V -c
CXX_VERSION_OPTION = -V -c
FC_VERSION_OPTION  = -V -c


%if %{VEC} eq 'novec'
    vec=-no-vec
%elif %{VEC} eq 'avx2'
    vec=-xCORE-AVX2
%elif %{VEC} eq 'avx512'
    vec=-xCORE-AVX512
%elif %{VEC} eq 'avx512_high'
    vec=-xCORE-AVX512 -qopt-zmm-usage=high
%else
    vec=-xCORE-AVX512
%endif


default=base,peak=default:
#OPTIMIZE      = -Ofast -march=x86-64-v4 -ipo -mprefer-vector-width=512 #-no-prec-div
OPTIMIZE      = -Ofast -march=skylake-avx512 -ipo -mprefer-vector-width=512
COPTIMIZE     = -ansi-alias
CXXOPTIMIZE   = -ansi-alias
FOPTIMIZE     = -nostandard-realloc-lhs -align array64byte
PORTABILITY   = -lstdc++
%if %{model} eq 'mpi'
        submit = mpirun -hosts 10.241.70.23 -np $ranks $command
%else %{model} eq 'omp'
#       submit = mpiexec --allow-run-as-root -host 192.168.99.15:24,192.168.99.16:24 -np $ranks $command
        submit = mpiexec -hosts 192.168.99.14:12,192.168.99.15:12 -np $ranks -genv OMP_NUM_THREADS=$threads  -ppn %{NRNK} $command
%endif

%if %{model} eq 'omp'
  pmodel=OMP
  OPTIMIZE += -fiopenmp
%endif

default=base=default:
    ranks   = %{RANKS}
    threads = %{THREADS}
    ppn     = %{NRNK}

#default=peak=default:
basepeak=0

613.soma_s=default=default:
%if %{model} eq 'omp'
        PORTABILITY += -DSPEC_NO_VAR_ARRAY_REDUCE
%endif

513.soma_t=default=default:
%if %{model} eq 'omp'
        PORTABILITY += -DSPEC_NO_VAR_ARRAY_REDUCE
%endif

505.lbm_t=peak:
    basepeak=1

513.soma_t=peak:
    basepeak=1

518.tealeaf_t=peak:
    ranks=24
    threads=8
    pmodel=OMP

519.clvleaf_t=peak:
    ranks=24
    threads=8
    pmodel=OMP

521.miniswp_t=peak:
    ranks=4
    threads=96
    pmodel=OMP

528.pot3d_t=peak:
    basepeak=1

532.sph_exa_t=peak:
    basepeak=1

534.hpgmgfv_t=peak:
    basepeak=1

535.weather_t=peak:
     basepeak=1



# The following section was added automatically, and contains settings that
# did not appear in the original configuration file, but were added to the
# raw file after the run.
default:
node_compute_hw_memory000 = 768 GB (12 x 64 GB 2Rx4 PC5-6400B-R), running at
node_compute_hw_memory001 = 6000
node_compute_hw_tcache000 = 384 MB I+D on chip per chip
node_compute_hw_tcache001 = 32 MB shared / 16 cores
node_compute_sw_os000 = Ubuntu 24.04 LTS
node_compute_sw_os001 = Kernel 6.8.0-39-generic