NVIDIA GRID A100A vs NVIDIA Tesla S870
We compared two Professional market GPUs: 48GB VRAM GRID A100A and 1536MB VRAM Tesla S870 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GRID A100A 's Advantages
Released 13 years late
Boost Clock1005MHz
More VRAM (48GB vs 1536GB)
Larger VRAM bandwidth (1866GB/s vs 76.80GB/s)
6784 additional rendering cores
Lower TDP (400W vs 800W)
Score
Benchmark
FP32 (float)
GRID A100A
+3914%
13.89 TFLOPS
Tesla S870
0.346 TFLOPS
Graphics Card
May 2020
Release Date
May 2007
GRID
Generation
Tesla
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 1.0 x16
Clock Speeds
900 MHz
Base Clock
-
1005 MHz
Boost Clock
-
1215 MHz
Memory Clock
800 MHz
Memory
48GB
Memory Size
1536MB
HBM2e
Memory Type
GDDR3
6144bit
Memory Bus
384bit
1866GB/s
Bandwidth
76.80GB/s
Render Config
-
-
-
108
SM Count
16
6912
Shading Units
128
432
TMUs
32
192
ROPs
24
432
Tensor Cores
-
-
-
-
192 KB (per SM)
L1 Cache
-
48 MB
L2 Cache
96 KB
-
-
-
Theoretical Performance
193.0 GPixel/s
Pixel Rate
14.40 GPixel/s
434.2 GTexel/s
Texture Rate
38.40 GTexel/s
55.57 TFLOPS
FP16 (half)
-
13.89 TFLOPS
FP32 (float)
345.6 GFLOPS
6.947 TFLOPS
FP64 (double)
-
Board Design
400W
TDP
800W
800 W
Suggested PSU
1200 W
No outputs
Outputs
No outputs
None
Power Connectors
-
Graphics Processor
GA100
GPU Name
G80
-
-
-
Ampere
Architecture
Tesla
TSMC
Foundry
TSMC
7 nm
Process Size
90 nm
54.2 billion
Transistors
0.681 billion
826 mm²
Die Size
484 mm²
Graphics Features
N/A
DirectX
11.1 (10_0)
N/A
OpenGL
3.3
3.0
OpenCL
1.1 (1.0)
N/A
Vulkan
N/A
8.0
CUDA
1.0
N/A
Shader Model
4.0
