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 384 384 64 learning rate=0.001, batch size=70.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5767456860818951 0.0183700419065774 0.0020767278749708 0.0 0.0304216885634371 0.0010302305296566 0.0 0.0 0.10769736892067 0.0 0.0 0.0 0.0 0.0780935341571011 0.0070712765751572 0.0 0.1779157947300316 0.0059640515220533 0.0 0.0 0.0 0.0 0.0 0.012127565463727 0.1434831448938046 0.0 0.2215717657066732 0.0059640515220533 0.0 0.4255432228848478 0.0 0.0 0.0046925226223873 0.0 0.0 0.0 0.0170370645694796 0.0191137924444504 0.1593113595598412 0.0637317243062886 0.0 0.0 0.0 5.14253393999e-09 138.822441 88.492192 14.046848 384 64 168162901420.08895113178404993
Horovod CANDLE NT3 Benchmark ANL Theta 384 384 64 learning rate=0.001, batch size=60.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5991310839874605 0.0180501826974004 0.0020711277912226 0.0 0.0304366550485465 0.0010202658342328 0.0 0.0 0.1118341083964543 0.0 0.0 0.0 0.0 0.0811374632999218 0.0071409907795675 0.0 0.1757411193439441 0.0058859937462033 0.0 0.0 0.0 0.0 0.0 0.012187698285996 0.1436988317394115 0.0 0.2206015319943636 0.0058859937462033 0.0 0.4251951590313489 0.0 0.0 0.0047710028357023 0.0 0.0 0.0 0.0173099739119599 0.0193811017031826 0.1594474368348209 0.0638416144252249 0.0 0.0 0.0 5.43704917649e-09 136.184888 92.673726 16.14024 384 64 161839638089.876010592838668636
Horovod CANDLE NT3 Benchmark ANL Theta 384 384 64 learning rate=0.001, batch size=50.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5910871717690319 0.0178937654802443 0.0020975717407637 0.0 0.030682610847263 0.0010272888803993 0.0 0.0 0.110392145622044 0.0 0.0 0.0 0.0 0.0800583262575179 0.0071515790766937 0.0 0.1764468402944107 0.0057967142148636 0.0 0.0 0.0 0.0 0.0 0.0121627583805013 0.1447813328487712 0.0 0.2212100251910452 0.0057967142148636 0.0 0.4281874228340115 0.0 0.0 0.0048193398960884 0.0 0.0 0.0 0.0169993471464469 0.0190969188872106 0.1590961677406637 0.0637155141931279 0.0 0.0 0.0 5.46627931064e-09 144.953553 93.303302 11.016228 384 64 163968105737.915615645285056974
Horovod CANDLE NT3 Benchmark ANL Theta 384 384 64 learning rate=0.001, batch size=40.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5972636467758045 0.0181526332626914 0.002090877363758 0.0 0.0302041189251558 0.0010247902939581 0.0 0.0 0.1115095748482728 0.0 0.0 0.0 0.0 0.0808372586278756 0.0070908773590087 0.0 0.175276675238516 0.0059350913749328 0.0 0.0 0.0 0.0 0.0 0.0122287607917344 0.1425558274923084 0.0 0.2226827270513131 0.0059350913749328 0.0 0.4291277735480372 0.0 0.0 0.0048116955855283 0.0 0.0 0.0 0.0172709688201709 0.019361846183929 0.1605633076400427 0.0643796308956279 0.0 0.0 0.0 5.54169469927e-09 140.827298 89.754875 12.004457 384 64 162137051707.009418571202934647
Horovod CANDLE NT3 Benchmark ANL Theta 384 384 64 learning rate=0.001, batch size=30.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5919246921207357 0.0183342214572128 0.0020433611508855 0.0 0.030109288626777 0.0010358984375537 0.0 0.0 0.1105676513393149 0.0 0.0 0.0 0.0 0.0802071257334332 0.0071855838199463 0.0 0.1747392499546604 0.0059841953232036 0.0 0.0 0.0 0.0 0.0 0.0122365931486786 0.1419964914879005 0.0 0.2253706259640952 0.0059841953232036 0.0 0.4323424323720279 0.0 0.0 0.0047442107476366 0.0 0.0 0.0 0.0172520493453521 0.0192954104962376 0.1586580351008943 0.063456431317947 0.0 0.0 0.0 5.67288425109e-09 134.156926 87.358766 14.97132 384 64 163687640343.019597487135162107
Horovod CANDLE NT3 Benchmark ANL Theta 384 384 64 learning rate=0.001, batch size=10.0, epochs=384.0 1.301 nt3-hybrid-opt.py empirical 1.301 1 0.5868832862079264 0.018090518535481 0.0020827238141784 0.0 0.0300251241235378 0.0010082566627619 0.0 0.0 0.1095581977334403 0.0 0.0 0.0 0.0 0.0794531104534876 0.0070110713883852 0.0 0.1826634608200495 0.0058341476018203 0.0 0.0 0.0 0.0 0.0 0.0122939435250822 0.1418596558592256 0.0 0.2219559228938106 0.0058341476018203 0.0 0.4235183788919247 0.0 0.0 0.0046015191673394 0.0 0.0 0.0 0.017278931362591 0.0193616551767694 0.1608631335369455 0.0644198998886053 0.0 0.0 0.0 1.009097595628e-08 128.908518 76.35093 12.361428 384 64 165109501719.019787100429640506