Nom du GPU | Aldebaran | GM200 |
---|---|---|
GPU variant | Aldebaran XT | GM200-895-A1 |
Architecture | CDNA 2.0 | Maxwell 2.0 |
Fonderie | TSMC | TSMC |
Taille du processus | 6 nm | 28 nm |
Transistors | 58,200 million | 8,000 million |
Taille des matrices | unknown | 601 mm² |
Date de publication | Nov 8th, 2021 | Nov 10th, 2015 |
---|---|---|
Génération | Radeon Instinct | Tesla |
Production | unknown | End-of-life |
Bus interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Base clock | 1000 MHz | 948 MHz |
---|---|---|
Boost clock | 1700 MHz | 1112 MHz |
Mémoire clock | 1600 MHz 3.2 Gbps effective | 1502 MHz 6 Gbps effective |
Taille de la mémoire | 128 GB | 12 GB |
---|---|---|
Type de mémoire | HBM2E | GDDR5 |
Bus mémoire | 8192 bit | 384 bit |
Bande passante | 3,277 GB/s | 288.4 GB/s |
Shading units | 14080 | 3072 |
---|---|---|
TMUs | 880 | 192 |
ROPs | 0 | 96 |
Compute units | 220 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SMM) |
Cache L2 | 16 MB | 3 MB |
gpu.details.mcm | 2 | — |
Comptage SMM | — | 24 |
Taux de pixel | 0 MPixel/s | 106.8 GPixel/s |
---|---|---|
Taux de texture | 1,496 GTexel/s | 213.5 GTexel/s |
FP16 (half) performance | 383.0 TFLOPS (8:1) | — |
FP32 (float) performance | 47.87 TFLOPS | 6.832 TFLOPS |
FP64 (double) performance | 47.87 TFLOPS (1:1) | 213.5 GFLOPS (1:32) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 267 mm 10.5 inches | 267 mm 10.5 inches |
Taille | 111 mm 4.4 inches | — |
TDP | 500 W | 250 W |
Suggestions pour le PSU | 900 W | 600 W |
Sorties | No outputs | No outputs |
Connecteurs de puissance | 2x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | — | PG600 SKU 202 |
DirectX | — | 12 (12_1) |
---|---|---|
OpenGL | — | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | — | 1.1 |
Shader model | — | 6.4 |
CUDA | — | 5.2 |