{{ '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.
Not enough stock.
Your item was not added to your cart.
Not enough stock.
Please adjust your quantity.
{{'products.quick_cart.out_of_number_hint'| translate}}
{{'product.preorder_limit.hint'| translate}}
Limit {{ product.max_order_quantity }} per order.
Only {{ quantityOfStock }} item(s) left.
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 |