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 24 24 64 learning rate=0.001, batch size=60.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5905696369405977 0.0183196573409144 0.0020066548681068 0.0 0.0302834668789588 0.0010248021656386 0.0 0.0 0.1102837611807162 0.0 0.0 0.0 0.0 0.0799836693098817 0.0070488489708714 0.0 0.1779992093509856 0.0060377436739552 0.0 0.0 0.0 0.0 0.0 0.0123066176750544 0.142925646002961 0.0 0.2241739944407496 0.0060377436739552 0.0 0.4309611505142707 0.0 0.0 0.0047405366354621 0.0 0.0 0.0 0.0174927442101862 0.019499399078293 0.1594671219518183 0.0639111825600766 0.0 0.0 0.0 4.79257833379e-09 134.162008 80.878653 14.675349 24 64 164176799668.033788663846823546
Horovod CANDLE NT3 Benchmark ANL Theta 24 24 64 learning rate=0.001, batch size=50.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5933102305700378 0.0180615708043187 0.0020852682000698 0.0 0.0303738016053014 0.0010208848360736 0.0 0.0 0.1107614763579251 0.0 0.0 0.0 0.0 0.0803019286732581 0.0070216144961227 0.0 0.1812415175444181 0.0060133032313259 0.0 0.0 0.0 0.0 0.0 0.0120958172720236 0.1433484670499131 0.0 0.2233748389119663 0.0060133032313259 0.0 0.4258481611188983 0.0 0.0 0.0046975182457557 0.0 0.0 0.0 0.0173395085204018 0.0194247767204716 0.1596069042646069 0.0638499001962934 0.0 0.0 0.0 4.93826486965e-09 136.67205 87.024224 12.020497 24 64 162997305783.852994710741092377
Horovod CANDLE NT3 Benchmark ANL Theta 24 24 64 learning rate=0.001, batch size=40.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5842742256472907 0.0182996962238048 0.0020730358412446 0.0 0.0302057007923636 0.0010208396428486 0.0 0.0 0.1091984233152396 0.0 0.0 0.0 0.0 0.0791799690386209 0.007100684210565 0.0 0.1795500560903828 0.0060200343148798 0.0 0.0 0.0 0.0 0.0 0.0119407805960876 0.1427321088079718 0.0 0.2202436584471433 0.0060200343148798 0.0 0.4316928178846588 0.0 0.0 0.0046830867955294 0.0 0.0 0.0 0.0172593399283905 0.0193323757696351 0.1589381536512135 0.063818306586668 0.0 0.0 0.0 4.84478578769e-09 135.44052 81.218711 10.099133 24 64 166427890588.83208688989366457
Horovod CANDLE NT3 Benchmark ANL Theta 24 24 64 learning rate=0.001, batch size=70.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.583771671409029 0.0179276754018809 0.0020607417916687 0.0 0.0301771657375992 0.0010310682243878 0.0 0.0 0.1090269877962271 0.0 0.0 0.0 0.0 0.0790890118033824 0.0069819968805544 0.0 0.179949561157999 0.0058893770971119 0.0 0.0 0.0 0.0 0.0 0.0121058454288734 0.142494504381162 0.0 0.2230951388452581 0.0058893770971119 0.0 0.4301304689070348 0.0 0.0 0.0046970524047557 0.0 0.0 0.0 0.0171369432522241 0.0191976850438927 0.1599789973490535 0.0643080236122764 0.0 0.0 0.0 4.90286462362e-09 137.420664 92.038745 12.108856 24 64 165702422841.966464354890019181
Horovod CANDLE NT3 Benchmark ANL Theta 24 24 64 learning rate=0.001, batch size=30.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5849038640526565 0.0180607447984097 0.002065088994094 0.0 0.0302158332427239 0.0010220044392165 0.0 0.0 0.1093098552755647 0.0 0.0 0.0 0.0 0.0792704490673442 0.0070769921515533 0.0 0.1771674725933533 0.0059449873183444 0.0 0.0 0.0 0.0 0.0 0.0119605332202767 0.1426513525163674 0.0 0.2223583283734444 0.0059449873183444 0.0 0.4289691735251547 0.0 0.0 0.0047310002150733 0.0 0.0 0.0 0.0170773770405158 0.0191424660346098 0.1614431339797705 0.0647009823712008 0.0 0.0 0.0 5.05950347562e-09 140.709178 87.622765 13.358164 24 64 165738368011.940581152105412811
Horovod CANDLE NT3 Benchmark ANL Theta 24 24 64 learning rate=0.001, batch size=10.0, epochs=24.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5907113665931677 0.0179740955647888 0.0020393984443926 0.0 0.03045971027018 0.0010201448012209 0.0 0.0 0.1102765185050558 0.0 0.0 0.0 0.0 0.0800002396059138 0.0070800743528792 0.0 0.1760419788108951 0.0058989017627038 0.0 0.0 0.0 0.0 0.0 0.0121729480638372 0.1437602540455177 0.0 0.2222677989681615 0.0058989017627038 0.0 0.4298171479069975 0.0 0.0 0.0046523924531452 0.0 0.0 0.0 0.0176509038328068 0.0196903022771994 0.1587528802678953 0.0636947206546379 0.0 0.0 0.0 6.71461841986e-09 133.842915 81.291082 13.007243 24 64 164359716217.948593461624108655