前回は Scale 28 で 32ノード, 128GPU だったが、今回は 32ノード, 64 GPU に変更。しかし、今回の方が性能が高い。。。
◯ Graph500 & Scale 28
median_TEPS: 1.15001e+10
============= Result ==============
SCALE: 28
edgefactor: 16
NBFS: 64
graph_generation: 17.0261831284
num_mpi_processes: 64
construction_time: 25.7171049118
redistribution_time: 2.76876497269
min_time: 0.332682
firstquartile_time: 0.354009
median_time: 0.373469
thirdquartile_time: 0.394995
max_time: 0.4714
mean_time: 0.374731
stddev_time: 0.0268339
min_nedge: 4294927670
firstquartile_nedge: 4294927670
median_nedge: 4294927670
thirdquartile_nedge: 4294927670
max_nedge: 4294927670
mean_nedge: 4294927670
stddev_nedge: 0
min_TEPS: 9.111e+09
firstquartile_TEPS: 1.08734e+10
median_TEPS: 1.15001e+10
thirdquartile_TEPS: 1.21323e+10
max_TEPS: 1.291e+10
harmonic_mean_TEPS: 1.14614e+10
harmonic_stddev_TEPS: 1.03402e+08
min_validate: 4.44144
firstquartile_validate: 4.60263
median_validate: 4.76156
thirdquartile_validate: 5.0697
max_validate: 6.86868
mean_validate: 4.90372
stddev_validate: 0.427448
TSUBAME-KFC - LX 1U-4GPU/104Re-1G Cluster, Intel Xeon E5-2620v2 6C 2.100GHz, Infiniband FDR, NVIDIA K20x
◯ Graph500 & Scale 28
median_TEPS: 1.15001e+10
============= Result ==============
SCALE: 28
edgefactor: 16
NBFS: 64
graph_generation: 17.0261831284
num_mpi_processes: 64
construction_time: 25.7171049118
redistribution_time: 2.76876497269
min_time: 0.332682
firstquartile_time: 0.354009
median_time: 0.373469
thirdquartile_time: 0.394995
max_time: 0.4714
mean_time: 0.374731
stddev_time: 0.0268339
min_nedge: 4294927670
firstquartile_nedge: 4294927670
median_nedge: 4294927670
thirdquartile_nedge: 4294927670
max_nedge: 4294927670
mean_nedge: 4294927670
stddev_nedge: 0
min_TEPS: 9.111e+09
firstquartile_TEPS: 1.08734e+10
median_TEPS: 1.15001e+10
thirdquartile_TEPS: 1.21323e+10
max_TEPS: 1.291e+10
harmonic_mean_TEPS: 1.14614e+10
harmonic_stddev_TEPS: 1.03402e+08
min_validate: 4.44144
firstquartile_validate: 4.60263
median_validate: 4.76156
thirdquartile_validate: 5.0697
max_validate: 6.86868
mean_validate: 4.90372
stddev_validate: 0.427448
TSUBAME-KFC - LX 1U-4GPU/104Re-1G Cluster, Intel Xeon E5-2620v2 6C 2.100GHz, Infiniband FDR, NVIDIA K20x