# Invocation command line: # /home/HPC2021/bin/harness/runhpc --reportable --config=1N_Pro8S_I204_IMPI204.cfg --tune=base --pmodel=OMP --define model=omp --threads=8 --rank=16 --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 Pro8S_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.2021-10-20.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 CXX = mpiicpc FC = mpiifort system_class = Homogenous sw_compiler = Intel C/C++/Fortran Compiler 2021.5.0 sw_mpi_library = Intel MPI 2021.5 test_sponsor = Lenovo Global Technology license_num = 28 tester = Lenovo Global Technology hw_avail = May-2022 sw_avail = Oct-2020 prepared_by = Lenovo Global Technology system_vendor = Lenovo Global Technology system_name = ThinkSystem SR645 (AMD EPYC 7773X) node_fileserver_syslbl = ThinkSystem SR645 node_fileserver_sw_state = Multi-User, run level 3 node_fileserver_sw_sharedfile = N/A node_fileserver_sw_other = None node_fileserver_sw_os = Red Hat Enterprise Linux Server release 8.3 node_fileserver_sw_localfile = xfs node_fileserver_purpose = Fileserver node_fileserver_order = 1 node_fileserver_hw_vendor = Lenovo Global Technology node_fileserver_hw_scache = 512 KB I+D on chip per core node_fileserver_hw_pcache = 32 KB I + 32 KB D on chip per core node_fileserver_hw_other = None node_fileserver_hw_ocache = None node_fileserver_hw_nthreadspercore = 1 node_fileserver_hw_ncpuorder = 1,2 chips node_fileserver_hw_ncoresperchip = 64 node_fileserver_hw_ncores = 128 node_fileserver_hw_nchips = 2 node_fileserver_hw_model = ThinkSystem SR645 node_fileserver_hw_memory = 1 TB (16 x 64 GB 2Rx4 PC4-3200A-R) node_fileserver_hw_disk = 1 x 7.68 TB 2.5" SSD node_fileserver_hw_cpu_name = AMD EPYC 7773X node_fileserver_hw_cpu_mhz = 2200 node_fileserver_hw_cpu_char = Max Boost Clock up to 3.5 GHz node_fileserver_hw_adapter_fs_slot_type = PCI-Express 4.0 x16 node_fileserver_hw_adapter_fs_ports_used = 1 node_fileserver_hw_adapter_fs_model = Mellanox ConnectX-6 HDR node_fileserver_hw_adapter_fs_interconnect = Nvidia Mellanox ConnectX-6 HDR node_fileserver_hw_adapter_fs_firmware = 20.28.1002 node_fileserver_hw_adapter_fs_driver = 5.2-1.0.4 node_fileserver_hw_adapter_fs_data_rate = 200 Gb/s node_fileserver_hw_adapter_fs_count = 1 node_fileserver_count = 1 node_compute_syslbl = ThinkSystem SR645 node_compute_sw_state = Multi-user, run level 3 node_compute_sw_sharedfile = NFS 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 = 512 KB I+D on chip per core node_compute_hw_pcache = 32 KB I + 32 KB D on chip per core node_compute_hw_other = None node_compute_hw_ocache = None node_compute_hw_nthreadspercore = 1 node_compute_hw_ncpuorder = 1,2 chips node_compute_hw_ncoresperchip = 64 node_compute_hw_ncores = 128 node_compute_hw_nchips = 2 node_compute_hw_model = ThinkSystem SR645 node_compute_hw_memory = 1 TB (16 x 64 GB 2Rx4 PC4-3200A-R) node_compute_hw_disk = 1 x 7.68 TB 2.5" SSD node_compute_hw_cpu_name = AMD EPYC 7773X node_compute_hw_cpu_mhz = 2200 node_compute_hw_cpu_char = Max Boost Clock up to 3.5 GHz node_compute_hw_adapter_fs_slot_type = PCI-Express 4.0 x16 node_compute_hw_adapter_fs_ports_used = 1 node_compute_hw_adapter_fs_model = Mellanox ConnectX-6 HDR node_compute_hw_adapter_fs_interconnect = Nvidia Mellanox ConnectX-6 HDR node_compute_hw_adapter_fs_firmware = 20.28.1002 node_compute_hw_adapter_fs_driver = 5.2-1.0.4 node_compute_hw_adapter_fs_data_rate = 200 Gb/s node_compute_hw_adapter_fs_count = 1 node_compute_count = 1 interconnect_fs_syslbl = Nvidia Mellanox ConnectX-6 HDR interconnect_fs_purpose = MPI Traffic, NFS Access interconnect_fs_order = 0 interconnect_fs_label = Nvidia Mellanox ConnectX-6 HDR interconnect_fs_hw_vendor = Nvidia interconnect_fs_hw_topo = Mesh interconnect_fs_hw_switch_fs_ports = 40 interconnect_fs_hw_switch_fs_model = QM8700 Series interconnect_fs_hw_switch_fs_firmware = 3.9.0606 interconnect_fs_hw_switch_fs_data_rate = 200 Gb/s interconnect_fs_hw_switch_fs_count = 1 interconnect_fs_hw_model = Nvidia Mellanox ConnectX-6 HDR CC_VERSION_OPTION = -V -c CXX_VERSION_OPTION = -V -c FC_VERSION_OPTION = -V -c %if %{model} eq 'mpi' submit = mpirun -host localhost -genv coll_hcoll_enable 1 -genv HCOLL_MAIN_IB=mlx5_0:1 -genv UCX_TLS=sm -genv pml ucx -np $ranks $command %else %{model} eq 'omp' submit = mpirun -host localhost -genv coll_hcoll_enable 1 -genv HCOLL_MAIN_IB=mlx5_0:1 -genv UCX_TLS=sm -genv pml ucx --map-by numa -np $ranks $command %endif %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=default: OPTIMIZE = -Ofast -no-prec-div -march=core-avx2 -ipo COPTIMIZE = -ansi-alias CXXOPTIMIZE = -ansi-alias #PORTABILITY = -DSPEC_LP64 %if %{model} eq 'omp' pmodel=OMP OPTIMIZE += -qopenmp %endif default=peak=default: basepeak=1 613.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 # 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_tcache000 = 768 MB I+D on chip per chip node_compute_hw_tcache001 = 96 MB shared / 8 cores node_compute_sw_os000 = Red Hat Enterprise Linux Server release 8.3, node_compute_sw_os001 = Kernel 4.18.0-193.el8.x86_64 node_fileserver_hw_tcache000 = 768 MB I+D on chip per chip node_fileserver_hw_tcache001 = 96 MB shared / 8 cores notes_000 =Environment variables set by runhpc before the start of the run: notes_005 =UCX_MEMTYPE_CACHE = "n" notes_010 =UCX_TLS = "self,shm,cuda_copy" notes_015 = notes_020 =Numa Information: notes_025 =available: 16 nodes (0-15) notes_030 =node 0 cpus: 0 1 2 3 4 5 6 7 notes_035 =node 1 cpus: 8 9 10 11 12 13 14 15 notes_040 =node 2 cpus: 16 17 18 19 20 21 22 23 notes_045 =node 3 cpus: 24 25 26 27 28 29 30 31 notes_050 =node 4 cpus: 32 33 34 35 36 37 38 39 notes_055 =node 5 cpus: 40 41 42 43 44 45 46 47 notes_060 =node 6 cpus: 48 49 50 51 52 53 54 55 notes_065 =node 7 cpus: 56 57 58 59 60 61 62 63 notes_070 =node 8 cpus: 64 65 66 67 68 69 70 71 notes_075 =node 9 cpus: 72 73 74 75 76 77 78 79 notes_080 =node 10 cpus: 80 81 82 83 84 85 86 87 notes_085 =node 11 cpus: 88 89 90 91 92 93 94 95 notes_090 =node 12 cpus: 96 97 98 99 100 101 102 103 notes_095 =node 13 cpus: 104 105 106 107 108 109 110 111 notes_100 =node 14 cpus: 112 113 114 115 116 117 118 119 notes_105 =node 15 cpus: 120 121 122 123 124 125 126 127 notes_110 = notes_submit_000 = The config file option 'submit' was used. notes_submit_005 =submit = mpirun -host localhost -genv coll_hcoll_enable 1 notes_submit_010 =-genv HCOLL_MAIN_IB=mlx5_0:1 -genv UCX_TLS=sm notes_submit_015 =-genv pml ucx --map-by numa -np 16