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 16 16 64 learning rate=0.001, batch size=70.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5902453071597196 0.0183618776263757 0.0020440129739131 0.0 0.0302006780438515 0.0010176232792456 0.0 0.0 0.110209523654373 0.0 0.0 0.0 0.0 0.079908639549218 0.0070164081491071 0.0 0.1742999836879979 0.0060561309321939 0.0 0.0 0.0 0.0 0.0 0.0121589114223572 0.142638347123886 0.0 0.2204992857525742 0.0060561309321939 0.0 0.4333336572110789 0.0 0.0 0.0046377618291061 0.0 0.0 0.0 0.0172488174311998 0.0192928304051129 0.1604731371042588 0.0644000760223946 0.0 0.0 0.0 4.60645291532e-09 141.854305 92.727815 14.276159 16 64 163713645805.844871713429777031
Horovod CANDLE NT3 Benchmark ANL Theta 16 16 64 learning rate=0.001, batch size=60.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.599008865515171 0.0182486551196868 0.0020510600137418 0.0 0.0301562316718229 0.0010133509141791 0.0 0.0 0.1118883089615874 0.0 0.0 0.0 0.0 0.081170902186719 0.0070033979488303 0.0 0.1703282566094464 0.0060337006009503 0.0 0.0 0.0 0.0 0.0 0.0120235817839664 0.1425919209581549 0.0 0.2237304996019348 0.0060337006009503 0.0 0.4300705548087597 0.0 0.0 0.0046888681351057 0.0 0.0 0.0 0.0173766712031944 0.0194277312169362 0.1609822690803477 0.0645540774537041 0.0 0.0 0.0 4.71224512834e-09 143.537705 91.457049 15.015082 16 64 161659985050.1316291207496042
Horovod CANDLE NT3 Benchmark ANL Theta 16 16 64 learning rate=0.001, batch size=40.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5981282104755422 0.0182070456558854 0.0020243299448889 0.0 0.0303682318924875 0.0010266812701709 0.0 0.0 0.1117123147809814 0.0 0.0 0.0 0.0 0.0810718034657957 0.007140552446513 0.0 0.1825299809258488 0.0059847133176629 0.0 0.0 0.0 0.0 0.0 0.0120783374670875 0.1433990127784874 0.0 0.2231384631564113 0.0059847133176629 0.0 0.4281904706660653 0.0 0.0 0.004783376423954 0.0 0.0 0.0 0.0174476111453071 0.019471941090196 0.1600010811430122 0.0642426631659923 0.0 0.0 0.0 4.82044274873e-09 139.303011 91.356182 12.522742 16 64 161751878747.118174157102079717
Horovod CANDLE NT3 Benchmark ANL Theta 16 16 64 learning rate=0.001, batch size=50.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5936584796517225 0.0184282477999906 0.0021084601853717 0.0 0.030543795251353 0.0010186646477035 0.0 0.0 0.1108590910715803 0.0 0.0 0.0 0.0 0.0804215634869583 0.0071748021602618 0.0 0.1787417755823633 0.0060664064223079 0.0 0.0 0.0 0.0 0.0 0.0120403289696458 0.1440836309479853 0.0 0.2275112232164613 0.0060664064223079 0.0 0.421874054213967 0.0 0.0 0.0047190157637236 0.0 0.0 0.0 0.0170791805846844 0.0191876407700562 0.1593785127925163 0.0639704075846351 0.0 0.0 0.0 4.78229231507e-09 141.471191 89.413572 11.503201 16 64 163209255850.05260225095552693
Horovod CANDLE NT3 Benchmark ANL Theta 16 16 64 learning rate=0.001, batch size=30.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5962119400507195 0.0181589383124256 0.0020433955932721 0.0 0.0306680879785701 0.0010296498329541 0.0 0.0 0.1113580662669094 0.0 0.0 0.0 0.0 0.0808240746218883 0.0070354956450498 0.0 0.1737192763747356 0.006008455759876 0.0 0.0 0.0 0.0 0.0 0.0121021054779025 0.1450617201979529 0.0 0.2239261989470354 0.006008455759876 0.0 0.4285110030532234 0.0 0.0 0.0047444510641917 0.0 0.0 0.0 0.0174728552254369 0.019516250818709 0.1587565230443437 0.0636860371184063 0.0 0.0 0.0 5.00883552582e-09 140.989558 89.775184 14.050983 16 64 162481810753.162016671012000605
Horovod CANDLE NT3 Benchmark ANL Theta 16 16 64 learning rate=0.001, batch size=10.0, epochs=16.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5912371005287071 0.0185004833452092 0.0020680852686407 0.0 0.0303284505156681 0.0010319701343939 0.0 0.0 0.1104244564685805 0.0 0.0 0.0 0.0 0.080130288325442 0.0070514975043409 0.0 0.1792607486342704 0.0060984505501577 0.0 0.0 0.0 0.0 0.0 0.0123223515542964 0.1433813517982944 0.0 0.2233414661950228 0.0060984505501577 0.0 0.4290080278650433 0.0 0.0 0.0048217708142125 0.0 0.0 0.0 0.0173189294674885 0.0193870147361292 0.1592853942981643 0.0638852049824321 0.0 0.0 0.0 5.87910875121e-09 136.54117 86.610047 13.226308 16 64 164143085939.920207465578273062