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 2 2 64 learning rate=0.001, batch size=70.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5933089095689126 0.0182918735109577 0.0020811722133773 0.0 0.0300530240388353 0.0010242182730947 0.0 0.0 0.1108055733082197 0.0 0.0 0.0 0.0 0.0804114451570653 0.0069542953214643 0.0 0.1778893597189587 0.0060703813003564 0.0 0.0 0.0 0.0 0.0 0.0122087137639322 0.1419143467836369 0.0 0.2170894874302511 0.0060703813003564 0.0 0.4308530311272061 0.0 0.0 0.0046903364231472 0.0 0.0 0.0 0.0170656187446911 0.0191467909580684 0.1591829922191528 0.0639139308563158 0.0 0.0 0.0 4.43435838271e-09 144.831724 92.817931 11.311034 2 64 163630662516.853431359178326272
Horovod CANDLE NT3 Benchmark ANL Theta 2 2 64 learning rate=0.001, batch size=60.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5937890330576944 0.0179808649565451 0.0020550851745009 0.0 0.0303238012983609 0.0010394537698039 0.0 0.0 0.1109027307452538 0.0 0.0 0.0 0.0 0.0804457721963184 0.0069962298157857 0.0 0.1717692034266843 0.0059393173976361 0.0 0.0 0.0 0.0 0.0 0.0120067586396306 0.1434235262886118 0.0 0.225492749904214 0.0059393173976361 0.0 0.4279955781099293 0.0 0.0 0.0047277006610068 0.0 0.0 0.0 0.0172758043339088 0.0193308895084097 0.1601519525308599 0.0642652248067337 0.0 0.0 0.0 4.4871900131e-09 147.017747 89.729693 12.301706 2 64 163132145254.149901646829371706
Horovod CANDLE NT3 Benchmark ANL Theta 2 2 64 learning rate=0.001, batch size=50.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5887377452109142 0.0182207190341103 0.0020712445481611 0.0 0.0300072322577823 0.0010119464601488 0.0 0.0 0.1099470096102978 0.0 0.0 0.0 0.0 0.0797418997433747 0.0071336611392575 0.0 0.1706681114288826 0.0059528944751342 0.0 0.0 0.0 0.0 0.0 0.0121988783212738 0.1416780462907861 0.0 0.224341992643827 0.0059528944751342 0.0 0.422631907349759 0.0 0.0 0.004781774405265 0.0 0.0 0.0 0.0173573677377399 0.019428612285901 0.1578644494303785 0.0633939151804322 0.0 0.0 0.0 4.49355665828e-09 147.410014 94.64885 9.966171 2 64 164423118964.924271060525527283
Horovod CANDLE NT3 Benchmark ANL Theta 2 2 64 learning rate=0.001, batch size=40.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5875655824206287 0.017656809137457 0.0020111609142144 0.0 0.0301631520839049 0.0010398630881675 0.0 0.0 0.1097315336382338 0.0 0.0 0.0 0.0 0.0796065774471679 0.0070013518478146 0.0 0.1766366302121883 0.0058150415875536 0.0 0.0 0.0 0.0 0.0 0.0120857612324704 0.1423241046277838 0.0 0.226366469002164 0.0058150415875536 0.0 0.4262478661926403 0.0 0.0 0.004646672744658 0.0 0.0 0.0 0.0175913902627004 0.0196025511769148 0.1619307697732199 0.0650308204582004 0.0 0.0 0.0 4.51206193881e-09 139.679435 88.729839 12.387097 2 64 164898499420.029950721340417006
Horovod CANDLE NT3 Benchmark ANL Theta 2 2 64 learning rate=0.001, batch size=30.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5929518786185441 0.0181984191237751 0.0020817572118503 0.0 0.0302715814904161 0.0010299486785833 0.0 0.0 0.1107244112279021 0.0 0.0 0.0 0.0 0.0802902046304562 0.0070938272908007 0.0 0.1728788893537301 0.0060652481052999 0.0 0.0 0.0 0.0 0.0 0.0120733448652857 0.1430879686702952 0.0 0.2233812337146583 0.0060652481052999 0.0 0.4211365627327683 0.0 0.0 0.0047214581941892 0.0 0.0 0.0 0.0171893229527602 0.0192710801646105 0.1583193537522034 0.063517873182186 0.0 0.0 0.0 4.69893271525e-09 144.919218 90.562215 11.022801 2 64 163321213881.048240659844804744
Horovod CANDLE NT3 Benchmark ANL Theta 2 2 64 learning rate=0.001, batch size=10.0, epochs=2.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5929793812623476 0.0182999906634938 0.0021404107656443 0.0 0.030373430325144 0.0010172128981136 0.0 0.0 0.1107014991000119 0.0 0.0 0.0 0.0 0.0803226465434351 0.0069714843725486 0.0 0.1757301086438128 0.0060185948540264 0.0 0.0 0.0 0.0 0.0 0.0122367540047815 0.1436330409380683 0.0 0.2217710783570254 0.0060185948540264 0.0 0.4305468707505187 0.0 0.0 0.0046462546663758 0.0 0.0 0.0 0.0169731627662809 0.0191135735319252 0.1575463663972967 0.0631714997486158 0.0 0.0 0.0 5.73213600859e-09 140.909382 84.392324 12.125267 2 64 163602743478.984524009116835114