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 96 96 64 learning rate=0.001, batch size=70.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5893746224070637 0.0178899090970817 0.0020958567196867 0.0 0.0300908196943404 0.0010187155292407 0.0 0.0 0.1100863717407495 0.0 0.0 0.0 0.0 0.0798539408373474 0.0069932850707592 0.0 0.1786224266161412 0.005933122452732 0.0 0.0 0.0 0.0 0.0 0.0120500877576261 0.1419731856659226 0.0 0.2231827882532304 0.005933122452732 0.0 0.4263425215558452 0.0 0.0 0.0047300816180086 0.0 0.0 0.0 0.0172466989352644 0.0193425556549511 0.156828406138186 0.0629058476579795 0.0 0.0 0.0 4.91817427576e-09 134.949969 82.620754 13.403953 96 64 164545459275.117991005089043299
Horovod CANDLE NT3 Benchmark ANL Theta 96 96 64 learning rate=0.001, batch size=60.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.6000653731833757 0.0179277636199167 0.0020832947624995 0.0 0.0302177495429402 0.0009900177723089 0.0 0.0 0.1120901800484508 0.0 0.0 0.0 0.0 0.0813529633180383 0.0070202269663414 0.0 0.1740640288339367 0.0059065169952091 0.0 0.0 0.0 0.0 0.0 0.0121279714203954 0.1428003202624744 0.0 0.2203994032563909 0.0059065169952091 0.0 0.4275647870557664 0.0 0.0 0.0046580764914193 0.0 0.0 0.0 0.0172940912312759 0.0193773859937754 0.1607384820138966 0.0644096691151989 0.0 0.0 0.0 5.03038054123e-09 142.060848 91.609096 13.859865 96 64 161568962295.896241753521021422
Horovod CANDLE NT3 Benchmark ANL Theta 96 96 64 learning rate=0.001, batch size=40.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5948714207651038 0.0179969390839134 0.0020159294191316 0.0 0.0300185273094794 0.0010296559292915 0.0 0.0 0.1111306895751197 0.0 0.0 0.0 0.0 0.0805861022960251 0.0069442236511101 0.0 0.1738334680188239 0.0059478854921014 0.0 0.0 0.0 0.0 0.0 0.0119830493431656 0.1417350877871271 0.0 0.2240258602159276 0.0059478854921014 0.0 0.4301308362882124 0.0 0.0 0.0046848508780608 0.0 0.0 0.0 0.0175922387648392 0.0196081681839707 0.1597537482700035 0.0640494387925227 0.0 0.0 0.0 5.10964105265e-09 139.94841 89.823635 14.010798 96 64 162959367482.018658859148345465
Horovod CANDLE NT3 Benchmark ANL Theta 96 96 64 learning rate=0.001, batch size=50.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5897015450916218 0.0179112965395152 0.0021193759239344 0.0 0.030120789980383 0.0010183996751516 0.0 0.0 0.1101550463739201 0.0 0.0 0.0 0.0 0.0799021207123486 0.0071189576993739 0.0 0.1717054054076819 0.0058987952061862 0.0 0.0 0.0 0.0 0.0 0.0121375238842405 0.1421730077996323 0.0 0.2222322591658069 0.0058987952061862 0.0 0.4230695360196011 0.0 0.0 0.0046784529480637 0.0 0.0 0.0 0.0169578796382531 0.0190772555621876 0.1581528834637143 0.0633559319430308 0.0 0.0 0.0 5.06904266343e-09 141.036571 90.784173 11.127698 96 64 164382257228.067766489802546058
Horovod CANDLE NT3 Benchmark ANL Theta 96 96 64 learning rate=0.001, batch size=30.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5950600090768826 0.0182302613702387 0.0021077742577725 0.0 0.0304424139944914 0.0010200616139186 0.0 0.0 0.1111415397715463 0.0 0.0 0.0 0.0 0.0805970413409401 0.0070337836282213 0.0 0.1789997657320741 0.0060052863040597 0.0 0.0 0.0 0.0 0.0 0.0121241101949518 0.143845951094657 0.0 0.2195356830632363 0.0060052863040597 0.0 0.4302360373773635 0.0 0.0 0.0046643259435434 0.0 0.0 0.0 0.0170732098198228 0.0191809840775953 0.1615322484438923 0.0648078141639161 0.0 0.0 0.0 5.29316443459e-09 144.17913 94.730435 13.521159 96 64 162839417261.890553958839014818
Horovod CANDLE NT3 Benchmark ANL Theta 96 96 64 learning rate=0.001, batch size=10.0, epochs=96.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5990764222363061 0.0181239152609982 0.0021118694300413 0.0 0.0303303060633457 0.0010218467042417 0.0 0.0 0.1118580547303678 0.0 0.0 0.0 0.0 0.081189546531359 0.0070140041941036 0.0 0.1699921082990375 0.005955053441451 0.0 0.0 0.0 0.0 0.0 0.0120757359596238 0.1433527696971186 0.0 0.2219488792151318 0.005955053441451 0.0 0.4295629693064829 0.0 0.0 0.0047385681532622 0.0 0.0 0.0 0.0167446258534697 0.018856495283511 0.161020196365336 0.0646070022367016 0.0 0.0 0.0 8.99388354241e-09 133.745199 83.582305 12.35048 96 64 161894861784.015649273186195299