ABCDEFGHIJKLMNOPQRSTUVWXYZAA
1
Performance Comparisons (TF 2.0 on DGX Station)
2
TaskGPUsAMP/XLAVal Acc (Avg)Time-to-resultPeak ThroughputE2E Speed-upThroughput Speed-up
3
CoLA1 x V100 16GB0.824451591.01.0
4
CoLA1 x V100 16GBAMP0.820284981.61.7
5
CoLA1 x V100 16GBAMP+XLA0.8232341571.92.7
6
CoLA4 x V100 16GB0.8122072182.23.7
7
CoLA4 x V100 16GBAMP0.8281843652.56.2
8
CoLA4 x V100 16GBAMP+XLA0.8282515051.88.6
9
MRPC1 x V100 16GB0.835204581.01.0
10
MRPC1 x V100 16GBAMP0.823134971.51.7
11
MRPC1 x V100 16GBAMP+XLA0.8271411581.42.7
12
MRPC4 x V100 16GB0.8271502181.43.8
13
MRPC4 x V100 16GBAMP0.8171433651.46.3
14
MRPC4 x V100 16GBAMP+XLA0.8182225290.99.1
15
SST21 x V100 16GB0.9223436591.01.0
16
SST21 x V100 16GBAMP0.9182088971.61.6
17
SST21 x V100 16GBAMP+XLA0.92213521582.52.7
18
SST24 x V100 16GB0.92910252163.43.7
19
SST24 x V100 16GBAMP0.9266723645.16.2
20
SST24 x V100 16GBAMP+XLA0.9256015085.78.6
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
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100