Huawei Atlas · Compute Server
Atlas 800 Inference Server (Model 3000)
The Atlas 800 inference server (Model 3000) supports up to eight Atlas 300I/V Pro cards for strong real-time inference and video analytics in data-center scenarios.
Key Highlights
- Up to 8 Atlas 300I/V Pro cards
- 1024ch 1080P 30FPS analysis
- dual Kunpeng 920 CPUs
- 72W per accelerator card for efficient inference.
Technical Specifications
| CPU | Kunpeng 920 x2 |
|---|---|
| System Memory | 32x DDR4 DIMM slots up to 3200MT/s |
| Accelerator Cards | up to 8x Atlas 300I/V Pro |
| AI Compute | up to 1120 TOPS INT8 |
| Video Analytics | 1024ch 1080P 30FPS |
| Interconnect | up to 9x PCIe 4.0 |
| Form Factor | 2U AI server |
| PSU | 1+1 redundant 900W/2000W AC |
Typical Use Cases
Data-center AI inference, large-scale real-time video analytics and content moderation.
翻译稿；img1: main（官网Atlas 800系列服务器机箱图）; 资料链接为官网《Atlas 800推理服务器技术白皮书(型号3000)》落地页。
Related Products
Huawei Atlas
Atlas 200I DK A2 Developer Kit
Easy-to-use AI development kit for fast innovation View details →
Huawei Atlas
Atlas 200I A2 Acceleration Module
Edge AI accelerator module embedded in smart devices, robots and drones View details →
Huawei Atlas
Atlas 300I A2 Inference Card
High-performance AI inference card integrated into servers View details →
Huawei Atlas