Xinchuan Server | Enterprise Server Hardware Supplier
The Huawei Atlas 800T A2 is a 4U AI training server built on four Kunpeng 920 CPUs and Ascend 910 AI processors. Beijing Xinchuan Technology supplies and configures these systems for deep learning model development and training workloads, covering computer vision, natural language processing and large-model pre-training. Units are quoted by configuration, because the processor count, memory population and local drive layout are fixed per project rather than per catalogue.
The chassis routes eight 200GE QSFP ports directly from the compute plane and runs RoCE, so parameter synchronisation between nodes travels over the fabric without an internal switching layer. Local storage can be ordered as 7 × 2.5″ SATA plus 3 × 2.5″ NVMe, 8 × 2.5″ SATA plus 2 × 2.5″ NVMe, or 4 × 2.5″ SATA plus 6 × 2.5″ NVMe. Four 2.6 kW power modules run in 2+2 redundancy, and eight hot-swap fan modules provide N+1 cooling. The 4U enclosure measures 175 × 447 × 790 mm and is rated for 5–35 °C operation.
- Form factor: 4U AI server
- CPU: 4 × Kunpeng 920
- AI processors: Ascend 910
- Local storage: 7 × 2.5″ SATA + 3 × 2.5″ NVMe, 8 × 2.5″ SATA + 2 × 2.5″ NVMe, or 4 × 2.5″ SATA + 6 × 2.5″ NVMe
- RAID: 0/1/10/5/50/6/60
- Network: 8 × 200GE QSFP direct, RoCE
- PCIe: up to 3 × PCIe 4.0 slots
- Power: 4 × hot-swap 2.6 kW modules, 2+2 redundant; 200–240 V AC or 240 V DC
- Dimensions: 175 × 447 × 790 mm
Each system is supplied by Beijing Xinchuan Technology with configuration support and shipping to enterprise customers worldwide. Processor count, memory and drive population are confirmed on the quotation, because the same chassis is built in several configurations. Contact us for current availability and a configured quote.
| Parameters | Atlas 800T A2 |
| Hardware specifications | |
| Form factor | 4U AI server |
| CPU | 4 × Kunpeng 920 |
| AI processors | Ascend 910 |
| Local storage | 7 × 2.5″ SATA + 3 × 2.5″ NVMe 8 × 2.5″ SATA + 2 × 2.5″ NVMe 4 × 2.5″ SATA + 6 × 2.5″ NVMe |
| RAID support | RAID 0/1/10/5/50/6/60 |
| Network | 8 × 200GE QSFP direct, RoCE |
| PCIe slots | Up to 3 × PCIe 4.0 |
| Power modules | 4 × hot-swap 2.6 kW, 2+2 redundancy |
| Input power | 200–240 V AC / 240 V DC |
| Cooling | Air cooled, 8 hot-swap fan modules, N+1 |
| Operating temperature | 5 °C to 35 °C |
| Dimensions (H × W × D) | 175 × 447 × 790 mm |
Related models
Frequently asked questions
What workloads is the Atlas 800T A2 built for?
Deep learning model development and training, including computer vision, natural language processing and large-model pre-training. The 4U chassis and direct 200GE fabric are sized for cluster training rather than single-node inference.
Can the processor count be changed after ordering?
No. The Kunpeng 920 CPUs and Ascend processors are fixed at build time, so the processor count is part of the configuration you confirm on the quotation.
How is storage configured?
Three factory layouts are available: 7 × 2.5″ SATA plus 3 × 2.5″ NVMe, 8 × 2.5″ SATA plus 2 × 2.5″ NVMe, or 4 × 2.5″ SATA plus 6 × 2.5″ NVMe. The layout is selected when the order is placed.
Is the system tested before dispatch?
Every unit is powered on, firmware-checked and burn-in tested before dispatch. Test logs or photos can be shared on request.
How are warranty and lead time handled?
Warranty, firmware support period and lead time are confirmed on the quotation, because they follow the configuration and the delivery destination you choose.
