| OMP1 | OMP2 | OMP4 | OMP8 | OMP16 | OMP24 | |
|---|---|---|---|---|---|---|
| experiment_name | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| external_libraries | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| number_processes | 4 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| number_nodes | 1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| number_processes_per_node | 4 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| mpi_command | mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| run_command | <executable> | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| batch_script | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| batch_command | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| pinning_command | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| custom_categories | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| dataset | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| dataset_handler | link | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| run_directory | . | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| frequencies | { 1 = 0 ; }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| filter | { type = number ; value = 10 ; }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| profile_start | { unit = none ; value = 0 ; }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| additional_hwc | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| excluded_areas | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| decan_multi_variant | true | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| decan_all_variants | true | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| maximal_path_number | 4 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| included_areas | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| is_sudo_available | false | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| keep_executable_location | false | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| lprof_params | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| lprof_post_process_params | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| vprof_params | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| decan_params | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 | |
| cqa_params | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| multiruns_params | { { script_variables = { }, __number_processes = 4 ; throughput_core = -1 ; profile_start = { unit = none ; value = 0 ; }, dataset = ; number_processes_per_node = 4 ; mpi_command = mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> ; throughput_max_core = 0 ; __dataset = /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//dataset-run1 ; __number_nodes = 1 ; __number_processes_per_node = 4 ; name = OMP2 ; number_nodes = 1 ; environment_variables = { { name = OMP_NUM_THREADS ; value = 2 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, run_directory = . ; number_processes = 4 ; pinning_command = ; run_command = <executable> ; }, { script_variables = { }, __number_processes = 4 ; throughput_core = -1 ; profile_start = { unit = none ; value = 0 ; }, dataset = ; number_processes_per_node = 4 ; mpi_command = mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> ; throughput_max_core = 0 ; __dataset = /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//dataset-run1 ; __number_nodes = 1 ; __number_processes_per_node = 4 ; name = OMP4 ; number_nodes = 1 ; environment_variables = { { name = OMP_NUM_THREADS ; value = 4 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, run_directory = . ; number_processes = 4 ; pinning_command = ; run_command = <executable> ; }, { script_variables = { }, __number_processes = 4 ; throughput_core = -1 ; profile_start = { unit = none ; value = 0 ; }, dataset = ; number_processes_per_node = 4 ; mpi_command = mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> ; throughput_max_core = 0 ; __dataset = /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//dataset-run1 ; __number_nodes = 1 ; __number_processes_per_node = 4 ; name = OMP8 ; number_nodes = 1 ; environment_variables = { { name = OMP_NUM_THREADS ; value = 8 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, run_directory = . ; number_processes = 4 ; pinning_command = ; run_command = <executable> ; }, { script_variables = { }, __number_processes = 4 ; throughput_core = -1 ; profile_start = { unit = none ; value = 0 ; }, dataset = ; number_processes_per_node = 4 ; mpi_command = mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> ; throughput_max_core = 0 ; __dataset = /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//dataset-run1 ; __number_nodes = 1 ; __number_processes_per_node = 4 ; name = OMP16 ; number_nodes = 1 ; environment_variables = { { name = OMP_NUM_THREADS ; value = 16 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, run_directory = . ; number_processes = 4 ; pinning_command = ; run_command = <executable> ; }, { script_variables = { }, __number_processes = 4 ; throughput_core = -1 ; profile_start = { unit = none ; value = 0 ; }, dataset = ; number_processes_per_node = 4 ; mpi_command = mpirun -n <number_processes> --bind-to core --map-by ppr:<number_processes>:node:PE=<OMP_NUM_THREADS> ; throughput_max_core = 0 ; __dataset = /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//dataset-run1 ; __number_nodes = 1 ; __number_processes_per_node = 4 ; name = OMP24 ; number_nodes = 1 ; environment_variables = { { name = OMP_NUM_THREADS ; value = 24 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, run_directory = . ; number_processes = 4 ; pinning_command = ; run_command = <executable> ; }, }, | |||||
| scalability_reference | main | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| environment_variables | { { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, { name = OMP_NUM_THREADS ; value = 1 ; }, }, | { { name = OMP_NUM_THREADS ; value = 2 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, | { { name = OMP_NUM_THREADS ; value = 4 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, | { { name = OMP_NUM_THREADS ; value = 8 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, | { { name = OMP_NUM_THREADS ; value = 16 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, | { { name = OMP_NUM_THREADS ; value = 24 ; }, { name = OMP_PROC_BIND ; value = close ; }, { name = OMP_PLACES ; value = cores ; }, }, |
| script_variables | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| thread_filter_threshold | 1% | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| object_coverage_threshold | 0.01 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| localbinary | /home/eoseret/hpl-2.3/OV1_hpl_N100K_NB384_G4_OMP1to24//binaries/xhpl | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| bucket_threshold | 1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| is_all_external_libraries_in_cc | false | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| decan_threshold | 500 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| ranges_count | 20 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| #__scalability_reference | true | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| _scalability_bins | { }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| delay | 0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| base_run_index | 0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| repetitions | 31 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| _is_loaded | config_G4.json | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| qplot_path | nil | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| outliers_count | 0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| throughput_core | -1 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| job_submission_threshold | 0s | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| throughput_max_core | 0 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| base_run_name | OMP1 | OMP2 | OMP4 | OMP8 | OMP16 | OMP24 |
| optimizer_loop_count | 10 | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| __maximal_path_number | true | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| __filter | true | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| binary | ./bin/Linux_AArch64/xhpl | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| comments | HPL benchmark compiled with ARM ACfL/Armpl 24.10. Matrix order: 100K, block size 384. Run on AWS Graviton 4 with 1 NUMA node and 96 cores. Using 4 MPI ranks to limit multithreading overhead | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| basebinary | xhpl | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| _is_custom_categories | false | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |
| filter_decan | { type = all ; }, | same as r0 | same as r0 | same as r0 | same as r0 | same as r0 |