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 6 6 64 learning rate=0.001, batch size=70.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5970544260505132 0.018066315683069 0.0020780305761656 0.0 0.0301676088694115 0.0010190266618865 0.0 0.0 0.1115098592874497 0.0 0.0 0.0 0.0 0.0808921300417216 0.0070989975448907 0.0 0.1880381244375898 0.0060007394498322 0.0 0.0 0.0 0.0 0.0 0.0122235132407902 0.1424541564341809 0.0 0.2219586563028473 0.0060007394498322 0.0 0.428395052504339 0.0 0.0 0.0046671188118121 0.0 0.0 0.0 0.0173224357916783 0.0194004663678439 0.1596945251811643 0.0640748497436773 0.0 0.0 0.0 4.59391540975e-09 150.361745 99.934228 13.728859 6 64 162012930717.090246265390542219
Horovod CANDLE NT3 Benchmark ANL Theta 6 6 64 learning rate=0.001, batch size=60.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5894669515652927 0.0180523744574044 0.0020245188944142 0.0 0.0301455254889804 0.0010231374701297 0.0 0.0 0.1102133170036059 0.0 0.0 0.0 0.0 0.0799376441306365 0.0070102310771658 0.0 0.1790970891444123 0.0059635758471646 0.0 0.0 0.0 0.0 0.0 0.012095540503677 0.1422132265560092 0.0 0.2194547505607615 0.0059635758471646 0.0 0.4295433314515658 0.0 0.0 0.0046585336675978 0.0 0.0 0.0 0.0175589738806029 0.019583492775017 0.1607759135057272 0.064339378011799 0.0 0.0 0.0 4.57887960699e-09 141.796149 82.837317 13.15405 6 64 164397741939.06731765690442037
Horovod CANDLE NT3 Benchmark ANL Theta 6 6 64 learning rate=0.001, batch size=50.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.6153537741738946 0.0183319243073015 0.0020334801626851 0.0 0.0312205946509827 0.001038702764154 0.0 0.0 0.1149420306148897 0.0 0.0 0.0 0.0 0.0833944158726554 0.0070909231917315 0.0 0.1804740401381009 0.0060942117419103 0.0 0.0 0.0 0.0 0.0 0.012005695212455 0.1474519652055027 0.0 0.2239697078571928 0.0060942117419103 0.0 0.4456735683485247 0.0 0.0 0.0047895207027182 0.0 0.0 0.0 0.017301659955686 0.0193351401183711 0.1595648989727199 0.0640708033813286 0.0 0.0 0.0 4.86210830158e-09 146.50098 93.526453 9.797518 6 64 157413480022.91242701521032868
Horovod CANDLE NT3 Benchmark ANL Theta 6 6 64 learning rate=0.001, batch size=30.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5959039371172494 0.0182954943663157 0.0020862944289816 0.0 0.0299330990405305 0.0010094469295726 0.0 0.0 0.1113430685571371 0.0 0.0 0.0 0.0 0.0807523318474071 0.0070472335132779 0.0 0.17917276484292 0.0060613559895257 0.0 0.0 0.0 0.0 0.0 0.0119932656868432 0.141369814822611 0.0 0.2231863322616338 0.0060613559895257 0.0 0.4205063597040642 0.0 0.0 0.0046994423347487 0.0 0.0 0.0 0.0173703402966193 0.0194566347256009 0.159469829585479 0.0639888681439786 0.0 0.0 0.0 4.9385103921e-09 138.575871 83.349502 13.102612 6 64 162832381457.853326281755178166
Horovod CANDLE NT3 Benchmark ANL Theta 6 6 64 learning rate=0.001, batch size=40.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5944833823186911 0.018074959248866 0.002039962555992 0.0 0.0299267024373984 0.0010149337304571 0.0 0.0 0.111086939490836 0.0 0.0 0.0 0.0 0.080644656300795 0.0070016111634351 0.0 0.1739299524538605 0.0059100567768053 0.0 0.0 0.0 0.0 0.0 0.0121868905159074 0.1413734609957523 0.0 0.2218256627219659 0.0059100567768053 0.0 0.425609877824606 0.0 0.0 0.004630392718208 0.0 0.0 0.0 0.0173705548723204 0.0194105174283124 0.1601131830190223 0.0642721448837528 0.0 0.0 0.0 5.48358427829e-09 141.459777 87.724581 11.86257 6 64 163155954498.979029495878221541
Horovod CANDLE NT3 Benchmark ANL Theta 6 6 64 learning rate=0.001, batch size=10.0, epochs=6.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.577119744415703 0.0179768379433439 0.0017712233744284 0.0 0.0300101810755978 0.0010152728096941 0.0 0.0 0.1077851574632672 0.0 0.0 0.0 0.0 0.0782141483603406 0.0070672962322579 0.0 0.1727482700685937 0.0058958438194016 0.0 0.0 0.0 0.0 0.0 0.0126863718159895 0.1418971035959472 0.0 0.2197259544408801 0.0058958438194016 0.0 0.426085870937954 0.0 0.0 0.0046563358371419 0.0 0.0 0.0 0.0176898908744843 0.0194611142489127 0.1582457586656569 0.0634597737527752 0.0 0.0 0.0 5.80521598668e-09 138.8186 82.827798 12.667859 6 64 168535189775.003846162322165254