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 32 32 64 learning rate=0.001, batch size=70.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5915456245453429 0.0181991053350652 0.0021098986962289 0.0 0.0303195386126334 0.0010125920824603 0.0 0.0 0.1105720735916597 0.0 0.0 0.0 0.0 0.0801930960041473 0.0069916771146888 0.0 0.1780824328434437 0.0060007800229333 0.0 0.0 0.0 0.0 0.0 0.012249626943342 0.143099579302906 0.0 0.2255412096864838 0.0060007800229333 0.0 0.4314312738585189 0.0 0.0 0.004636451270625 0.0 0.0 0.0 0.0169322292796207 0.0190421279758496 0.1579426935324858 0.0634309286677008 0.0 0.0 0.0 4.65373142023e-09 134.342243 86.565841 12.02412 32 64 164684584647.156656395773258662
Horovod CANDLE NT3 Benchmark ANL Theta 32 32 64 learning rate=0.001, batch size=60.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.589260439861016 0.017970948018464 0.0020350221923251 0.0 0.0307127600885935 0.0010195269688555 0.0 0.0 0.1100551175421051 0.0 0.0 0.0 0.0 0.0798333875485896 0.0070311799512325 0.0 0.1777726571684621 0.0058706528941053 0.0 0.0 0.0 0.0 0.0 0.0120897635805487 0.1451450003429502 0.0 0.2231477411063066 0.0058706528941053 0.0 0.4288679774325745 0.0 0.0 0.0047927386714248 0.0 0.0 0.0 0.0175844608188044 0.0196194830111296 0.1593571856230822 0.0639176631482258 0.0 0.0 0.0 4.70285678919e-09 144.247902 89.906391 13.665591 32 64 164553810521.89951685148775467
Horovod CANDLE NT3 Benchmark ANL Theta 32 32 64 learning rate=0.001, batch size=50.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5918506774986815 0.0177891947150531 0.0020669347997115 0.0 0.0300868364410104 0.0010155994215832 0.0 0.0 0.1105300575170683 0.0 0.0 0.0 0.0 0.0801996816286535 0.007070226726729 0.0 0.1764256669501891 0.0058401880053059 0.0 0.0 0.0 0.0 0.0 0.0120494684791785 0.1420857063300847 0.0 0.2220799638279443 0.0058401880053059 0.0 0.4358182371620772 0.0 0.0 0.0047580645011046 0.0 0.0 0.0 0.0171405340561962 0.0192074688559077 0.1594288582535308 0.0640201404732082 0.0 0.0 0.0 4.80783233811e-09 149.461002 92.985415 9.329106 32 64 163899662380.857973491609229015
Horovod CANDLE NT3 Benchmark ANL Theta 32 32 64 learning rate=0.001, batch size=40.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5925683280595216 0.0180943537581342 0.0020420635696528 0.0 0.030487636424052 0.0010200389064262 0.0 0.0 0.1106685334045081 0.0 0.0 0.0 0.0 0.0802674103051942 0.0070361516812861 0.0 0.1807417604508595 0.0059731541119499 0.0 0.0 0.0 0.0 0.0 0.0120694432152028 0.1440706208378984 0.0 0.2200172725314616 0.0059731541119499 0.0 0.4292425111140519 0.0 0.0 0.0046225474592547 0.0 0.0 0.0 0.0172805108898466 0.0193225744594993 0.1585045975382122 0.0636239889706511 0.0 0.0 0.0 4.89250883077e-09 136.812617 90.482823 9.978139 32 64 163581272039.123212073972920496
Horovod CANDLE NT3 Benchmark ANL Theta 32 32 64 learning rate=0.001, batch size=30.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5922223626963528 0.0181816478640797 0.0020639437029233 0.0 0.0302944066997383 0.0010117485188188 0.0 0.0 0.1106372795604314 0.0 0.0 0.0 0.0 0.0802539193792707 0.0069539534979366 0.0 0.1744592465560536 0.0059682435450362 0.0 0.0 0.0 0.0 0.0 0.0121642336403148 0.1433331406446414 0.0 0.2237476130733807 0.0059682435450362 0.0 0.4317223761322315 0.0 0.0 0.0046765019542518 0.0 0.0 0.0 0.0173392402613385 0.0194031839642618 0.1610490437345628 0.0646898407917667 0.0 0.0 0.0 5.01021572516e-09 146.664229 91.669104 12.015235 32 64 163673702887.077224112554084513
Horovod CANDLE NT3 Benchmark ANL Theta 32 32 64 learning rate=0.001, batch size=10.0, epochs=32.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.591045431487568 0.0180499782570371 0.0020540402207359 0.0 0.0304078906050813 0.001019655260747 0.0 0.0 0.1103420707330948 0.0 0.0 0.0 0.0 0.0800319590779838 0.0071308652500253 0.0 0.1707174847671214 0.0059276159218867 0.0 0.0 0.0 0.0 0.0 0.0122343016764494 0.1436723099200096 0.0 0.2243329874908706 0.0059276159218867 0.0 0.4322112508034198 0.0 0.0 0.0047787347259205 0.0 0.0 0.0 0.0175064270951072 0.019560467315843 0.1596996332528668 0.0641384554121246 0.0 0.0 0.0 6.56055486918e-09 133.612245 83.03757 11.014842 32 64 164260006583.054949039410298875