NVIDIA Tesla K40m vs NVIDIA Tesla P100 PCIe 16 GB
Graphics Processor
| GPU model | GK110B | GP100 |
|---|---|---|
| Architecture | Kepler | Pascal |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 16 nm |
| Number of transistors | 7,080 million | 15,300 million |
| Die size | 561 mm² | 610 mm² |
| GPU variant | — | GP100-893-A1 |
Graphics Card
| Release date | Nov 22nd, 2013 | Jun 20th, 2016 |
|---|---|---|
| Generation | Tesla | Tesla |
| Production | End-of-life | End-of-life |
| Launch price | 7,699 USD | 5,699 USD |
| Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Clocks
| Base clock | 745 MHz | 1190 MHz |
|---|---|---|
| Boost clock | 876 MHz | 1329 MHz |
| Memory clock | 1502 MHz 6 Gbps effective | 715 MHz 1430 Mbps effective |
Memory Configuration
| Memory size | 12 GB | 16 GB |
|---|---|---|
| Memory type | GDDR5 | HBM2 |
| Memory bus width | 384 bit | 4096 bit |
| Bandwidth | 288.4 GB/s | 732.2 GB/s |
Render Configuration
| Shading units | 2880 | 3584 |
|---|---|---|
| TMUs | 240 | 224 |
| ROPs | 48 | 96 |
| SMX count | 15 | — |
| Cache L1 | 16 KB (per SMX) | 24 KB (per SM) |
| Cache L2 | 1536 KB | 4 MB |
| SM count | — | 56 |
Performance
| Pixel rate | 52.56 GPixel/s | 127.6 GPixel/s |
|---|---|---|
| Texture rate | 210.2 GTexel/s | 297.7 GTexel/s |
| FP32 (float) performance | 5.046 TFLOPS | 9.526 TFLOPS |
| FP64 (double) performance | 1.682 TFLOPS (1:3) | 4.763 TFLOPS (1:2) |
| FP16 (half) performance | — | 19.05 TFLOPS (2:1) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
| TDP | 245 W | 250 W |
| Suggested PSU | 550 W | 600 W |
| Outputs | No outputs | No outputs |
| Power connectors | — | 1x 8-pin |
API Support & Features
| DirectX | 12 (11_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.1 | 1.2 |
| CUDA | 3.5 | 6.0 |
| Shader model | 5.1 | 6.4 |
