2U 2CPU (Kunpeng 920) Server support up to 8 AI Chips(Asend 910)
• Up to 2.24 PFLOPS FP16 in a 4U space • 1.36x the computing density of industry peers
• 8 100G RoCE v2 ports, slashing cross-server chip interconnect latency by 10–70%
• 8 x 100G RoCE v2 high-speed interfaces • Cross-server chip interconnection latency slashed by 10% to 70%