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 12 12 64 learning rate=0.001, batch size=70.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5905599758973704 0.0182352415995998 0.0020665473176776 0.0 0.0302121231873295 0.0010300540371593 0.0 0.0 0.1102832132823528 0.0 0.0 0.0 0.0 0.0800115786302654 0.0070763907693718 0.0 0.1796224632393872 0.0059996603028849 0.0 0.0 0.0 0.0 0.0 0.0121975161388209 0.1427055347503005 0.0 0.2223258963930038 0.0059996603028849 0.0 0.428499066232255 0.0 0.0 0.0047591111679606 0.0 0.0 0.0 0.0169892136591827 0.0190557609768602 0.1579514213796619 0.0633427472912602 0.0 0.0 0.0 4.65994615007e-09 140.004587 88.771953 15.802097 12 64 163704527784.84160875014855856
Horovod CANDLE NT3 Benchmark ANL Theta 12 12 64 learning rate=0.001, batch size=60.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5937973055920216 0.0181804919542728 0.0020780515646049 0.0 0.0306097001459567 0.0010298492397739 0.0 0.0 0.1108768884787699 0.0 0.0 0.0 0.0 0.0804011362171251 0.0070933373746833 0.0 0.1811005102118027 0.0059713096428383 0.0 0.0 0.0 0.0 0.0 0.0121749409654076 0.1446761793854808 0.0 0.2236554906695608 0.0059713096428383 0.0 0.4271194029422438 0.0 0.0 0.0047861367798119 0.0 0.0 0.0 0.017307053877765 0.0193851054423699 0.1585698057438726 0.0637245940876512 0.0 0.0 0.0 4.7167009406e-09 140.156047 91.726918 14.311443 12 64 163026337409.08411241238235735
Horovod CANDLE NT3 Benchmark ANL Theta 12 12 64 learning rate=0.001, batch size=50.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5993306558809934 0.0182162479386154 0.0020776386387958 0.0 0.0303001721494604 0.0010108294186592 0.0 0.0 0.111982837167557 0.0 0.0 0.0 0.0 0.0812314832947612 0.0071252745970445 0.0 0.1792013645035724 0.0059931660193971 0.0 0.0 0.0 0.0 0.0 0.0122441774533844 0.1430579387097882 0.0 0.2237006004696862 0.0059931660193971 0.0 0.4308018362231344 0.0 0.0 0.0047512808497828 0.0 0.0 0.0 0.0170185744428895 0.0190962130816853 0.1578077290003017 0.0632735547203709 0.0 0.0 0.0 4.85662872196e-09 137.652036 87.70229 11.295165 12 64 161833623279.896532500718481173
Horovod CANDLE NT3 Benchmark ANL Theta 12 12 64 learning rate=0.001, batch size=40.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5847809032141257 0.0178985733728161 0.0020838380041405 0.0 0.0302735465869405 0.0010115565634837 0.0 0.0 0.1092286669980563 0.0 0.0 0.0 0.0 0.079215704156627 0.0070327691264416 0.0 0.1782867541657459 0.0058954301016393 0.0 0.0 0.0 0.0 0.0 0.0120861034480551 0.1429936325111031 0.0 0.2203073997322404 0.0058954301016393 0.0 0.4190884398858453 0.0 0.0 0.0046290014631536 0.0 0.0 0.0 0.0172801626917311 0.0193640006958716 0.1587591121884697 0.0635721980889096 0.0 0.0 0.0 4.81866584579e-09 135.272386 86.103319 13.227301 12 64 165557884180.11983387441245809
Horovod CANDLE NT3 Benchmark ANL Theta 12 12 64 learning rate=0.001, batch size=30.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5877901765698914 0.0179383903632154 0.0020140577314275 0.0 0.0306822011650852 0.001018499356914 0.0 0.0 0.1098390067033959 0.0 0.0 0.0 0.0 0.0796770522554947 0.0070537504783518 0.0 0.1813047526258809 0.0058990452433343 0.0 0.0 0.0 0.0 0.0 0.0120764058643971 0.1448379525949758 0.0 0.2248241126304125 0.0058990452433343 0.0 0.4344863480480539 0.0 0.0 0.0046638860311867 0.0 0.0 0.0 0.0175352172563291 0.0195492749877566 0.1597341887796029 0.0641146698068669 0.0 0.0 0.0 4.97501848552e-09 140.215937 89.842457 13.076642 12 64 165165713733.848333401398179254
Horovod CANDLE NT3 Benchmark ANL Theta 12 12 64 learning rate=0.001, batch size=10.0, epochs=12.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5960266478408816 0.0183871813907884 0.0020644056771857 0.0 0.0304440813110471 0.0010168920591978 0.0 0.0 0.1112708677571296 0.0 0.0 0.0 0.0 0.0807239409622486 0.0071062392033495 0.0 0.1756131731042745 0.0060106366971864 0.0 0.0 0.0 0.0 0.0 0.0122338694944504 0.1439281205542697 0.0 0.219742651243754 0.0060106366971864 0.0 0.4300249371212961 0.0 0.0 0.0046878575512012 0.0 0.0 0.0 0.0171466179503858 0.0192110236275715 0.1589251596387447 0.0638543091944444 0.0 0.0 0.0 6.09831143431e-09 136.587437 87.952261 15.728643 12 64 163184368118.877027145797966078