Home GPU Comparison NVIDIA H100 SXM5 vs NVIDIA Tesla P100 SXM2

NVIDIA H100 SXM5 vs NVIDIA Tesla P100 SXM2

AI GPU We compared a GPU: 80GB VRAM H100 SXM5 and a Professional market GPU: 16GB VRAM Tesla P100 SXM2 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA H100 SXM5 's Advantages
Released 5 years and 11 months late
Boost Clock has increased by 34% (1980MHz vs 1480MHz)
More VRAM (80GB vs 16GB)
Larger VRAM bandwidth (3350GB/s vs 732.2GB/s)
13312 additional rendering cores
NVIDIA Tesla P100 SXM2 's Advantages
Lower TDP (300W vs 700W)

Score

Benchmark

FP32 (float)
H100 SXM5 +530%
66.91 TFLOPS
Tesla P100 SXM2
10.61 TFLOPS
VS

Graphics Card

Mar 2022
Release Date
Apr 2016
Tesla Hopper
Generation
Tesla
AI GPU
Type
Professional
PCIe 5.0 x16
Bus Interface
PCIe 3.0 x16

Clock Speeds

1590 MHz
Base Clock
1328 MHz
1980 MHz
Boost Clock
1480 MHz
1313 MHz
Memory Clock
715 MHz

Memory

80GB
Memory Size
16GB
HBM3
Memory Type
HBM2
5120bit
Memory Bus
4096bit
3350GB/s
Bandwidth
732.2GB/s

Render Config

132
SM Count
56
-
Compute Units
-
16896
Shading Units
3584
528
TMUs
224
24
ROPs
96
528
Tensor Cores
-
-
RT Cores
-
256 KB (per SM)
L1 Cache
24 KB (per SM)
50 MB
L2 Cache
4 MB

Theoretical Performance

47.52 GPixel/s
Pixel Rate
142.1 GPixel/s
1045 GTexel/s
Texture Rate
331.5 GTexel/s
267.6 TFLOPS
FP16 (half)
21.22 TFLOPS
66.91 TFLOPS
FP32 (float)
10.61 TFLOPS
33.45 TFLOPS
FP64 (double)
5.304 TFLOPS

Graphics Processor

GH100
GPU Name
GP100
-
GPU Variant
GP100-890-A1
Hopper
Architecture
Pascal
TSMC
Foundry
TSMC
4 nm
Process Size
16 nm
80 billion
Transistors
15.3 billion
814 mm²
Die Size
610 mm²

Board Design

700W
TDP
300W
1100 W
Suggested PSU
700 W
No outputs
Outputs
No outputs
8-pin EPS
Power Connectors
None

Graphics Features

N/A
DirectX
12 (12_1)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
9.0
CUDA
6.0
N/A
Shader Model
6.4

Related GPU Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy