NVIDIA Tesla V100 FHHL vs NVIDIA Tesla K20X
We compared two Professional market GPUs: 16GB VRAM Tesla V100 FHHL and 6GB VRAM Tesla K20X to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA Tesla V100 FHHL 's Advantages
Released 5 years and 4 months late
Boost Clock1290MHz
More VRAM (16GB vs 6GB)
Larger VRAM bandwidth (829.4GB/s vs 249.6GB/s)
2432 additional rendering cores
NVIDIA Tesla K20X 's Advantages
Lower TDP (235W vs 250W)
Score
Benchmark
FP32 (float)
Tesla V100 FHHL
+235%
13.21 TFLOPS
Tesla K20X
3.935 TFLOPS
Graphics Card
Mar 2018
Release Date
Nov 2012
Tesla
Generation
Tesla
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
937 MHz
Base Clock
-
1290 MHz
Boost Clock
-
810 MHz
Memory Clock
1300 MHz
Memory
16GB
Memory Size
6GB
HBM2
Memory Type
GDDR5
4096bit
Memory Bus
384bit
829.4GB/s
Bandwidth
249.6GB/s
Render Config
-
-
-
80
SM Count
-
5120
Shading Units
2688
320
TMUs
224
128
ROPs
48
640
Tensor Cores
-
-
-
-
128 KB (per SM)
L1 Cache
16 KB (per SMX)
6 MB
L2 Cache
1536 KB
-
-
-
Theoretical Performance
165.1 GPixel/s
Pixel Rate
40.99 GPixel/s
412.8 GTexel/s
Texture Rate
164.0 GTexel/s
26.42 TFLOPS
FP16 (half)
-
13.21 TFLOPS
FP32 (float)
3.935 TFLOPS
6.605 TFLOPS
FP64 (double)
1312 GFLOPS
Board Design
250W
TDP
235W
600 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
1x 8-pin
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
GV100
GPU Name
GK110
-
-
-
Volta
Architecture
Kepler
TSMC
Foundry
TSMC
12 nm
Process Size
28 nm
21.1 billion
Transistors
7.08 billion
815 mm²
Die Size
561 mm²
Graphics Features
12 (12_1)
DirectX
12 (11_0)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.1
7.0
CUDA
3.5
6.6
Shader Model
5.1
