Atlas 800I A2-G

The Atlas 800I A2-G inference server adopts 8-module efficient inference mode and provides strong AI inference capability. It has advantages in computing power, memory bandwidth and interconnection capability, and can be widely used in generative large model inference, such as intelligent customer service, copy generation, knowledge precipitation and other content generation scenarios.

Technical Specification

Mode	4U AI server
NPU	8 * Ascending AI module
CPU	4 * Kunpeng 920
Memory	32 DDR4 memory slots with a maximum capacity of 3200 MT/s A single memory module supports 16/32/64 GB
Local storage	8 * 2.5 SATA+2 * 2.5 NVMe
Local storage	4 * 2.5 SATA+6 * 2.5 NVMe
RAID	RAID 0, 1/10/5/5/50/6/60 is supported
Network	8 * 200GE QSFP interface direct outgoing, RoCE protocol
PCIe Ports	Supports a maximum of three PCIe 4.0 expansion slots
Power Supply	Four hot-swapable 2.6 kW power modules in 2+2 redundancy
Power Supply	200 ~ 240 V AC
Power Supply	240 V DC
Heat Dissipation Mode	air cooling
Fan Module	Supports eight hot-swappable fan modules in N+1 redundancy
Working Temperature	5℃～35℃（41℉～95℉）
Structure Size	175 mm (H) × 447 mm (W) × 790 mm (D)

* The descriptions and information displayed in the product promotional materials are for reference only. The actual delivered product shall prevail. The final interpretation right belongs to GLORY.

Resources

Documents

Atlas 800I A2-G

Download