MuMMI

Create Training Set

Train application_name system_name processors num_mpi num_omppn input frequency executable_name scenario frequency papi_tot_cyc papi_tot_ins papi_l1_tcm papi_l2_tcm papi_fml_ins papi_br_tkn papi_l2_ldm papi_l2_stm papi_fdv_ins papi_br_cn papi_tot_iis papi_ca_shr papi_ca_cln papi_ca_itv papi_br_ntk papi_br_msp papi_fp_ins papi_res_stl papi_l1_ldm papi_l1_stm papi_l2_tcw papi_l2_dca papi_l2_dcr papi_l2_dcw papi_l1_icm papi_br_ins papi_l1_dch papi_l1_dca papi_l1_dcm papi_l1_tca papi_l1_ica papi_fp_ops papi_l2_ica papi_tlb_dm papi_vec_ins papi_l2_dcm papi_l2_icm papi_l2_tch papi_l2_tca papi_ld_ins papi_sr_ins papi_tlb_im papi_hw_int papi_l3_tcm nruntime power_sys power_cpu power_mem num_mpi num_omppn scaling_factor
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=70.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5865728383576915 0.0181238807586422 0.0021063577491397 0.0 0.0299850570709074 0.0010243900374038 0.0 0.0 0.1095455494633736 0.0 0.0 0.0 0.0 0.0794289782621785 0.0070137370280357 0.0 0.1762074056889654 0.0059988387214728 0.0 0.0 0.0 0.0 0.0 0.012131803056078 0.1416401380949018 0.0 0.2233039597162136 0.0059988387214728 0.0 0.4283765617740316 0.0 0.0 0.0047341640383901 0.0 0.0 0.0 0.0171123878532477 0.0192187456023873 0.1586333135249496 0.0635508576823305 0.0 0.0 0.0 4.50374073938e-09 138.947721 87.232574 12.22319 8 64 165561233238.893756705923114295
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=60.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5843289657892179 0.0180368521242723 0.0020276832340809 0.0 0.0302390787905344 0.0010185469367107 0.0 0.0 0.1089805321081254 0.0 0.0 0.0 0.0 0.0790207462291949 0.0070422897213285 0.0 0.1755557633775722 0.0059510488872261 0.0 0.0 0.0 0.0 0.0 0.012001069086742 0.1427674635933407 0.0 0.2216335156403962 0.0059510488872261 0.0 0.4289415684473353 0.0 0.0 0.0047283788580254 0.0 0.0 0.0 0.0173951349909652 0.0194228182250461 0.1599595790891253 0.0640808617782744 0.0 0.0 0.0 4.56158214697e-09 149.838838 96.074637 11.351387 8 64 165874140292.003432424158008476
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=50.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5945493387252277 0.0186867960744913 0.0020642418581543 0.0 0.0301328696907343 0.0010355156149993 0.0 0.0 0.1110230649347236 0.0 0.0 0.0 0.0 0.0805150483288327 0.007021185215304 0.0 0.1819794862469819 0.0064566293956137 0.0 0.0 0.0 0.0 0.0 0.0121211822430051 0.1422906258956313 0.0 0.224538229812467 0.0064566293956137 0.0 0.4253911727737478 0.0 0.0 0.0047153844758772 0.0 0.0 0.0 0.0172062275875747 0.0192704694457289 0.1647594378768366 0.0661002459066736 0.0 0.0 0.0 4.71292748448e-09 141.435179 84.696417 12.48013 8 64 162758538620.84925333470128954
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=40.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5853535770997174 0.0180286602480425 0.0020684561425005 0.0 0.0304384511313821 0.0010086074923475 0.0 0.0 0.109352558416638 0.0 0.0 0.0 0.0 0.0793387714589063 0.0070366627444883 0.0 0.1815859556217758 0.0059306673410105 0.0 0.0 0.0 0.0 0.0 0.0120926094537625 0.1436145095472503 0.0 0.2208983825436993 0.0059306673410105 0.0 0.4299171605891068 0.0 0.0 0.004779175019018 0.0 0.0 0.0 0.0172895811894797 0.0193580373319801 0.1579267831719402 0.0631934571826501 0.0 0.0 0.0 4.69178964106e-09 146.530547 98.811576 13.388424 8 64 165591469447.141079067225711379
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=30.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5883977465771791 0.0185392127445565 0.0020224457040855 0.0 0.0297309048983493 0.0009813073582168 0.0 0.0 0.1098567530416189 0.0 0.0 0.0 0.0 0.0796845270178404 0.0073370099202001 0.0 0.1548449819164836 0.0063719829894863 0.0 0.0 0.0 0.0 0.0 0.0126518238030866 0.1405183188432872 0.0 0.2159087942008302 0.0063719829894863 0.0 0.4285773273654466 0.0 0.0 0.0049016412044487 0.0 0.0 0.0 0.0173422793211202 0.0193647250252057 0.158797005422505 0.0637545646396757 0.0 0.0 0.0 4.87922439587e-09 141.024269 92.473553 12.102676 8 64 164590249360.074058052567916277
Horovod CANDLE NT3 Benchmark ANL Theta 8 8 64 learning rate=0.001, batch size=10.0, epochs=8.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5893684496693916 0.0184796685675343 0.0021317590848209 0.0 0.0305260140071032 0.0010280321057641 0.0 0.0 0.110050243467334 0.0 0.0 0.0 0.0 0.0798231656087956 0.0070100246129276 0.0 0.174363994899656 0.0061838647881095 0.0 0.0 0.0 0.0 0.0 0.0121562749136298 0.1440045352513995 0.0 0.224409274457544 0.0061838647881095 0.0 0.4476338894380935 0.0 0.0 0.0047260236307033 0.0 0.0 0.0 0.016906712326088 0.0190384714109089 0.1609109664765326 0.0645958369598749 0.0 0.0 0.0 5.78847535132e-09 140.90804 84.623226 12.278508 8 64 164309427832.859572471121495635