Product Compare
NVIDIA Tesla K40s vs NVIDIA H100 PCIe
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Nov 2013 | Mar 2022 |
| Generation | Tesla Kepler | Tesla Hopper |
| Type | Professional | AI GPU |
| Bus Interface | PCIe 3.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 706 MHz | 1095 MHz |
| Boost Clock | 706 MHz | 1755 MHz |
| Memory Clock | 1502 MHz | 1593 MHz |
| Memory | ||
| Memory Size | 12GB | 80GB |
| Memory Type | GDDR5 | HBM2e |
| Memory Bus | 384bit | 5120bit |
| Bandwidth | 288.4GB/s | 2039GB/s |
| Render Config | ||
| Shading Units | 2880 | 14592 |
| TMUs | 240 | 456 |
| ROPs | 48 | 24 |
| L1 Cache | 16 KB (per SMX) | 256 KB (per SM) |
| L2 Cache | 1536 KB | 50 MB |
| SM Count | 114 | |
| Tensor Cores | 456 | |
| Theoretical Performance | ||
| Pixel Rate | 42.36 GPixel/s | 42.12 GPixel/s |
| Texture Rate | 169.4 GTexel/s | 800.3 GTexel/s |
| FP32 (float) | 4.067 TFLOPS | 51.22 TFLOPS |
| FP64 (double) | 1355.5 GFLOPS | 25.61 TFLOPS |
| FP16 (half) | 204.9 TFLOPS | |
| Graphics Processor | ||
| GPU Name | GK110B | GH100 |
| Architecture | Kepler | Hopper |
| Foundry | TSMC | TSMC |
| Process Size | 28 nm | 4 nm |
| Transistors | 7.08 billion | 80 billion |
| Die Size | 561 mm² | 814 mm² |
| Board Design | ||
| TDP | 245W | 350W |
| Suggested PSU | 550 W | 750 W |
| Outputs | No outputs | No outputs |
| Power Connectors | 1x 16-pin | |
| Graphics Features | ||
| DirectX | 12 (11_1) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2.175 | N/A |
| CUDA | 3.5 | 9.0 |
| Shader Model | 6.5 (5.1) | N/A |