Horovod CANDLE NT3 Benchmark 1.0


Download CSV E-AMOM Scatter-Plot

Application-level Performance

100s 300w

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 64 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=64.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=10.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 64 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=64.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; nt3-hybrid-opt.py; learning rate=0.001, batch size=20.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 64 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=64.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=30.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 64 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=64.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=40.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=50.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 64 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=64.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=60.0, epochs=384.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 1 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=1.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 2 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=2.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 4 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=4.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 6 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=6.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 8 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=8.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 12 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=12.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 16 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=16.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 24 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=24.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 32 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=32.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 48 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=48.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 96 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=96.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 128 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=128.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 192 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=192.0

Horovod CANDLE NT3 Benchmark 1.0; ANL Theta; 384 Processors; 1.301 GHz; nt3-hybrid-opt.py; learning rate=0.001, batch size=70.0, epochs=384.0