Intel Gaudi 2 vs NVIDIA Tesla P4
Intel Gaudi 2
NVIDIA Tesla P4
|
6144 Shaders
96GB HBM2e
550MHz
|
2560 Shaders
8GB GDDR5
1063MHz
|
|
Peak AI Performance
865.08 TFLOPS
FP8 Tensor (FP32 Accumulate)
|
Peak AI Performance
21.77 TOPS
INT8
|
|
FP32
5.5 TFLOPS
|
FP32
5.44 TFLOPS
|
|
FP16
11 TFLOPS
|
FP16
90 GFLOPS
|
|
Form Factor
OAM Module
-
|
Form Factor
PCIe Card
1.0-Slots
|
|
TDP
600W
|
TDP
75W
|
|
-
-
-
-
-
|
-
-
-
-
-
|
|
GB6 OpenCL N/A
0%
|
GB6 OpenCL N/A
0%
|
|
GB6 Metal N/A
0%
|
GB6 Metal N/A
0%
|
|
GB6 Vulkan N/A
0%
|
GB6 Vulkan N/A
0%
|
|
GB5 OpenCL N/A
0%
|
GB5 OpenCL N/A
0%
|
|
GB5 CUDA N/A
0%
|
GB5 CUDA N/A
0%
|
|
GB5 Metal N/A
0%
|
GB5 Metal N/A
0%
|
|
GB5 Vulkan N/A
0%
|
GB5 Vulkan N/A
0%
|
|
OCT 2020.1 N/A
0%
|
OCT 2020.1 N/A
0%
|
|
OCT Metal N/A
0%
|
OCT Metal N/A
0%
|
|
Peak AI
Performance
865.08 TFLOPS
FP8 Tensor (FP32 Accumulate)
|
Peak AI
Performance
21.77 TOPS
INT8
|
|
-
-
-
|
-
-
-
|
|
FP8
-
-
-
865.08 TFLOPS
Tensor (FP32 Accumulate)
-
|
-
-
-
-
-
-
|
|
FP16
11 TFLOPS
-
-
-
-
|
FP16
90 GFLOPS
-
-
-
-
|
|
FP32
5.5 TFLOPS
-
-
|
FP32
5.44 TFLOPS
-
-
|
|
-
-
-
|
FP64
170 GFLOPS
-
|
|
BF16
11 TFLOPS
432.54 TFLOPS
Tensor
-
|
-
-
-
-
|
|
-
-
-
|
-
-
-
|
|
-
-
-
|
-
-
-
|
|
-
-
-
-
|
INT8
21.77 TOPS
-
-
|
|
-
-
|
-
-
|
|
-
-
|
-
-
|
|
Pixel
Fillrate
-
|
Pixel
Fillrate
68.032 GPixel/s
|
|
-
-
|
-
-
|
|
Texture
Fillrate
-
|
Texture
Fillrate
170.08 GTexel/s
|
|
Manufacturer
Intel
|
Manufacturer
NVIDIA
|
|
Chip Designer
Intel
|
Chip Designer
NVIDIA
|
|
Architecture
Gaudi 2
|
Architecture
Pascal
|
|
Family
Gaudi
|
Family
Tesla P
|
|
Codename
Gaudi 2
HL-2080
Variant
HL-2080
|
Codename
NV134
GP104
Variant
GP104-895-A1
|
|
Market Segment
Server
|
Market Segment
Server
|
|
Release Date
5/1/2022
|
Release Date
9/13/2016
|
|
Foundry
TSMC
-
|
Foundry
TSMC
-
|
|
Fabrication Node
N7
-
|
Fabrication Node
16FF
-
|
|
-
-
-
|
Die Size
314 mm²
-
|
|
-
-
|
Transistor Count
7.2 Billion
-
|
|
-
-
-
|
Transistor Density
22.93M/mm²
-
|
|
Form
OAM Module
|
Form
PCIe Card
|
|
Shading Units
6144 Shaders
-
|
Shading Units
2560 Shaders
-
|
|
Texture Mapping Units
-
|
Texture Mapping Units
160 TMUs
|
|
Render Output Units
-
|
Render Output Units
64 ROPs
|
|
Tensor Cores
24 T-Cores
|
-
-
|
|
-
-
|
-
-
|
|
-
-
|
Streaming Multiprocessors
20 SMs
|
|
Compute Units
2 CUs
|
-
-
|
|
-
-
|
-
-
|
|
-
-
|
-
-
|
|
-
-
-
550MHz
|
-
-
810MHz Base
1063MHz
|
|
-
-
|
-
-
|
|
L1
-
-
-
Unknown
|
L1
-
48KB/SM
-
-
|
|
L2
Unknown
|
L2
2MB Shared
|
|
-
-
-
|
-
-
-
|
|
96GB
HBM2e
ECC
|
8GB
GDDR5
-
|
|
Bus Width
6144Bit
|
Bus Width
256Bit
|
|
Clock
1596MHz
Transfer Rate
3.2GT/s
Bandwidth
2451.5GB/s
|
Clock
1500MHz
Transfer Rate
6GT/s
Bandwidth
192GB/s
|
|
-
-
-
-
eSRAM
48MB
6400GB/s
-
-
|
-
-
-
-
-
-
-
-
-
|
|
TDP
600W
|
TDP
75W
|
|
-
-
|
-
-
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
|
|
Max Resolution
Unknown
|
Max Resolution
Unknown
|
|
Max Resolution Refresh Rate
-
|
Max Resolution Refresh Rate
-
|
|
Variable Refresh Rate
-
-
-
|
Variable Refresh Rate
G-Sync
FreeSync
-
|
|
Display Stream Compression (DSC)
Not Supported
|
Display Stream Compression (DSC)
Not Supported
|
|
Multi Monitor Support
Unknown
|
Multi Monitor Support
Unknown
|
|
-
-
|
-
-
|
|
No Encoders
-
|
Model
2x NVENC 4
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
No Decoders
|
Model
NVDEC 3
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
-
VP9
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
-
-
-
-
|
Direct X
12
Direct 3D
12_1
|
|
-
-
OpenCL
3.0
-
-
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
|
|
-
-
-
-
-
-
-
-
-
-
|
Shader Model
6.7
CUDA
6.1
-
-
PureVideo HD
VP8
VDPAU
Feature Set H
|
|
Not a Card
-
-
-
|
-
-
-
-
|
|
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
|
|
-
-
PCIe Version
4.0
PCIe Lanes
16
|
Slots Required
1.0
PCIe Version
3.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
Type
RoCE
|
Multi GPU Support
Supported
-
-
|
|
-
-
-
-
-
-
|
Height
69 mm (2.72 in)
Width
168 mm (6.61 in)
Depth
20 mm (0.79 in)
|
Copy Link