NVIDIA A100 PCIe vs NVIDIA Tesla T10 Processor
Processeur graphique
| Nom du GPU | GA100 | GT200B |
|---|---|---|
| Architecture | Ampere | Tesla 2.0 |
| Fonderie | TSMC | TSMC |
| Taille du processus | 7 nm | 55 nm |
| Transistors | 54,200 million | 1,400 million |
| Taille des matrices | 826 mm² | 470 mm² |
Carte graphique
| Date de publication | Jun 22nd, 2020 | Apr 9th, 2009 |
|---|---|---|
| Génération | Tesla | Tesla |
| Production | Active | End-of-life |
| Bus interface | PCIe 4.0 x16 | PCIe 2.0 x16 |
Vitesse clock
| GPU clock | 1410 MHz | 610 MHz |
|---|---|---|
| Mémoire clock | 1215 MHz 2.4 Gbps effective | 800 MHz 1600 Mbps effective |
| Base clock | 765 MHz | — |
| Boost clock | 1410 MHz | — |
| Shader clock | — | 1296 MHz |
Mémoire
| Taille de la mémoire | 40 GB | 4 GB |
|---|---|---|
| Type de mémoire | HBM2E | GDDR3 |
| Bus mémoire | 5120 bit | 512 bit |
| Bande passante | 1,555 GB/s | 102.4 GB/s |
Render Config
| Shading units | 6912 | 240 |
|---|---|---|
| TMUs | 432 | 80 |
| ROPs | 160 | 32 |
| Comptage SM | 108 | 30 |
| Tensor cores | 432 | — |
| Cache L1 | 192 KB (per SM) | — |
| Cache L2 | 40 MB | 256 KB |
Performance théorique
| Taux de pixel | 225.6 GPixel/s | 19.52 GPixel/s |
|---|---|---|
| Taux de texture | 609.1 GTexel/s | 48.80 GTexel/s |
| FP16 (half) performance | 77.97 TFLOPS (4:1) | — |
| FP32 (float) performance | 19.49 TFLOPS | 622.1 GFLOPS |
| FP64 (double) performance | 9.746 TFLOPS (1:2) | 77.76 GFLOPS (1:8) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 267 mm 10.5 inches | 267 mm 10.5 inches |
| TDP | 250 W | 188 W |
| Suggestions pour le PSU | 600 W | 450 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | 8-pin EPS | — |
| Board number | P1001 SKU 200 | — |
Caractéristiques graphiques
| DirectX | — | 11.1 (10_0) |
|---|---|---|
| OpenGL | — | 3.3 |
| OpenCL | 3.0 | 1.1 |
| Vulkan | — | — |
| CUDA | 8.0 | 1.3 |
| Shader model | — | 4.0 |
