{{ 'fb_in_app_browser_popup.desc' | translate }} {{ 'fb_in_app_browser_popup.copy_link' | translate }}
{{ 'in_app_browser_popup.desc' | translate }}
NVIDIA A30 is a high-performance GPU designed for AI inference, HPC, and cloud computing. It's ideal for data centers, offering exceptional energy efficiency for demanding computational tasks.
全店,Free Shipping
商品存貨不足,未能加入購物車
您所填寫的商品數量超過庫存
{{'products.quick_cart.out_of_number_hint'| translate}}
{{'product.preorder_limit.hint'| translate}}
每筆訂單限購 {{ product.max_order_quantity }} 件
現庫存只剩下 {{ quantityOfStock }} 件
NVIDIA A30 is a high-performance GPU designed for AI inference, HPC, and cloud computing. It's ideal for data centers, offering exceptional energy efficiency for demanding computational tasks.
| Feature | NVIDIA A30 | NVIDIA V100 | NVIDIA T4 | NVIDIA A100 | AMD Radeon VII | Google TPU v3 | Intel Habana Gaudi |
|---|---|---|---|---|---|---|---|
| CUDA Cores | 10608 | 5120 | 2560 | 6912 | 3840 | N/A | N/A |
| Memory | 24GB HBM2 | 32GB HBM2 | 16GB GDDR6 | 40GB HBM2 | 16GB HBM2 | High Bandwidth | N/A |
| TFLOPS | 19.5 | 15.7 | 8.1 | 54.1 | 13.8 | 45 | 90 |
| TDP | 250W | 300W | 70W | 400W | 295W | 300W | 300W |
| Process Node | 7nm | 12nm | 12nm | 7nm | 7nm | N/A | N/A |
| Accessory Model | Description |
|---|---|
| NVIDIA NVLink Bridge | High-speed interconnect for multi-GPU setups |
| NVIDIA GPU Rackmount | Efficient mounting solution for data center GPUs |
| NVIDIA Data Center Manager | Software for GPU monitoring and management |
Nvidia A30 Specifications | |
| Product Type | GPU |
| Brand | NVIDIA |
| Model Name | A30 |
| CUDA Cores | 24,576 |
| GPU Memory | 24 GB GDDR6 |
| Memory Bandwidth | 933 GB/s |
| Tensor Cores | 320 |
| Peak FP16 Performance | 10.3 TFLOPS |
| Peak INT8 Performance | 23.1 TOPS |
| Interconnect | PCIe Gen4 |
| Power Consumption | 165W |
| Form Factor | PCIe Full Height/Full Length |
| Cooling Solution | Passive |
| Multi-GPU Support | Yes, NVLink |
| Maximum Displays | 4 |
| Compute Capability | 8.0 |
| ECC Memory | Yes |
| FP64 Peak Performance | 1.03 TFLOPS |
| Unified Memory Architecture | Yes |
| AI Performance | 5.0 TFLOPS |