NVIDIA Tesla M2090 vs NVIDIA GRID A100B
We compared two Professional market GPUs: 6GB VRAM Tesla M2090 and 48GB VRAM GRID A100B to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA Tesla M2090 's Advantages
Lower TDP (250W vs 400W)
NVIDIA GRID A100B 's Advantages
Released 8 years and 10 months late
Boost Clock1005MHz
More VRAM (48GB vs 6GB)
Larger VRAM bandwidth (1866GB/s vs 177.4GB/s)
6400 additional rendering cores
Score
Benchmark
FP32 (float)
Tesla M2090
1.332 TFLOPS
GRID A100B
+942%
13.89 TFLOPS
Graphics Card
Jul 2011
Release Date
May 2020
Tesla
Generation
GRID
Professional
Type
Professional
PCIe 2.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
-
Base Clock
900 MHz
-
Boost Clock
1005 MHz
924 MHz
Memory Clock
1215 MHz
Memory
6GB
Memory Size
48GB
GDDR5
Memory Type
HBM2e
384bit
Memory Bus
6144bit
177.4GB/s
Bandwidth
1866GB/s
Render Config
-
-
-
16
SM Count
108
512
Shading Units
6912
64
TMUs
432
48
ROPs
192
-
Tensor Cores
432
-
-
-
64 KB (per SM)
L1 Cache
192 KB (per SM)
768 KB
L2 Cache
48 MB
-
-
-
Theoretical Performance
20.83 GPixel/s
Pixel Rate
193.0 GPixel/s
41.66 GTexel/s
Texture Rate
434.2 GTexel/s
-
FP16 (half)
55.57 TFLOPS
1332 GFLOPS
FP32 (float)
13.89 TFLOPS
666.1 GFLOPS
FP64 (double)
6.947 TFLOPS
Board Design
250W
TDP
400W
600 W
Suggested PSU
800 W
No outputs
Outputs
No outputs
1x 6-pin + 1x 8-pin
Power Connectors
None
Graphics Processor
GF110
GPU Name
GA100
-
-
-
Fermi 2.0
Architecture
Ampere
TSMC
Foundry
TSMC
40 nm
Process Size
7 nm
3 billion
Transistors
54.2 billion
520 mm²
Die Size
826 mm²
Graphics Features
12 (11_0)
DirectX
N/A
4.6
OpenGL
N/A
1.1
OpenCL
3.0
N/A
Vulkan
N/A
2.0
CUDA
8.0
5.1
Shader Model
N/A
