Home GPU Comparison NVIDIA H100 PCIe vs NVIDIA Tesla K40t

NVIDIA H100 PCIe vs NVIDIA Tesla K40t

AI GPU We compared a GPU: 80GB VRAM H100 PCIe and a Professional market GPU: 12GB VRAM Tesla K40t to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA H100 PCIe 's Advantages
Released 8 years and 4 months late
Boost Clock has increased by 100% (1755MHz vs 876MHz)
More VRAM (80GB vs 12GB)
Larger VRAM bandwidth (2039GB/s vs 288.4GB/s)
11712 additional rendering cores
NVIDIA Tesla K40t 's Advantages
Lower TDP (245W vs 350W)

Score

Benchmark

FP32 (float)
H100 PCIe +915%
51.22 TFLOPS
Tesla K40t
5.046 TFLOPS
VS

Graphics Card

Mar 2022
Release Date
Nov 2013
Tesla Hopper
Generation
Tesla
AI GPU
Type
Professional
PCIe 5.0 x16
Bus Interface
PCIe 3.0 x16

Clock Speeds

1095 MHz
Base Clock
745 MHz
1755 MHz
Boost Clock
876 MHz
1593 MHz
Memory Clock
1502 MHz

Memory

80GB
Memory Size
12GB
HBM2e
Memory Type
GDDR5
5120bit
Memory Bus
384bit
2039GB/s
Bandwidth
288.4GB/s

Render Config

114
SM Count
-
-
Compute Units
-
14592
Shading Units
2880
456
TMUs
240
24
ROPs
48
456
Tensor Cores
-
-
RT Cores
-
256 KB (per SM)
L1 Cache
16 KB (per SMX)
50 MB
L2 Cache
1536 KB

Theoretical Performance

42.12 GPixel/s
Pixel Rate
52.56 GPixel/s
800.3 GTexel/s
Texture Rate
210.2 GTexel/s
204.9 TFLOPS
FP16 (half)
-
51.22 TFLOPS
FP32 (float)
5.046 TFLOPS
25.61 TFLOPS
FP64 (double)
1.682 TFLOPS

Graphics Processor

GH100
GPU Name
GK110B
-
GPU Variant
-
Hopper
Architecture
Kepler
TSMC
Foundry
TSMC
4 nm
Process Size
28 nm
80 billion
Transistors
7.08 billion
814 mm²
Die Size
561 mm²

Board Design

350W
TDP
245W
750 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
1x 16-pin
Power Connectors
-

Graphics Features

N/A
DirectX
12 (11_1)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.1
9.0
CUDA
3.5
N/A
Shader Model
5.1

Related GPU Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy