Dell PowerEdge C4140 Deep Learning Performance Comparison - Scale-up vs. Scale - Page 47

Other Explored Aspects

Get Dell PowerEdge C4140 PDF manuals and user guides View all Dell PowerEdge C4140 manuals
Add to My Manuals
Save this manual to your list of manuals

Page 47 highlights

Deep Learning Performance: Scale-up vs Scale-out Figure 40. Multi-node training PowerEdge C4140-V100-SXM2- Configuration-K with IntelXeon4116 cpu, Multi-node training PowerEdge C4140-V100-SXM2 Configuration-M with IntelXeon6148 cpu, versus single-node training non Dell 8xV100-16GB-SXM2 In the Figure 40 we can see how the system C4140-V100-SXM2 Configuration-M outperforms in terms of training time in different batch sizes compared the other systems. 7.4 Other Explored Aspects This section shows the results of aspects explored during this project such as the hyper parameter tuning, learning rate effect on single-node and multi-node mode, and critical kernels executed in the TensorFlow benchmarks. These aspects could be subject of deeper study for future projects. Architectures & Technologies Dell EMC | Infrastructure Solutions Group 46

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
Deep Learning Performance: Scale-up vs Scale-out
Architectures & Technologies
Dell
EMC
| Infrastructure Solutions Group
46
Figure 40. Multi-node training PowerEdge C4140-V100-SXM2- Configuration-K with IntelXeon4116 cpu,
Multi-node training PowerEdge C4140-V100-SXM2 Configuration-M with IntelXeon6148 cpu, versus
single-node training non Dell 8xV100-16GB-SXM2
In the Figure
40
we can see how the system C4140-V100-SXM2 Configuration-M outperforms in terms of
training time in different batch sizes compared the other systems.
7.4
Other Explored Aspects
This section shows the results of aspects explored during this project such as the hyper
parameter tuning, learning rate effect on single-node and multi-node mode, and critical kernels
executed in the TensorFlow benchmarks. These aspects could be subject of deeper study for
future projects.