Huawei Rolls Out AI ‘Supernode’ Computing Cluster

(MENAFN) Chinese tech heavyweight Huawei has launched what it calls the world’s most powerful "supernode" AI computing cluster, marking a major milestone in China’s drive to develop artificial intelligence infrastructure free from reliance on U.S. technology, media reported Thursday.

The new system, built entirely with domestic chipmaking processes, bypasses Nvidia’s cutting-edge GPUs, which have been central to global AI development but are now subject to U.S. export restrictions targeting China.

“Huawei is seeking to build a ‘supernode + cluster’ computing solution using chip manufacturing processes available in China to meet the growing compute needs,” said Xu Zhijun, Huawei's Deputy Chairman, during an industry conference held in Shanghai.

The company also unveiled a roadmap for its upcoming Ascend series AI chips, revealing plans to release multiple models over the next three years. That includes the Ascend 950PR, set for rollout in Q1 2026, a launch window aligned with expected next-gen offerings from Nvidia and AMD.

The announcement closely followed fresh guidance from Beijing discouraging the purchase of Nvidia’s China-specific chips, part of a broader strategy to comply with tightening U.S. export rules.

Speaking a day earlier, Nvidia CEO Jensen Huang expressed frustration with the Chinese government's decision. "I’m patient about it, but they have larger agendas to work out between China and the United States."

Responding to the tensions, Chinese Foreign Ministry spokesperson Lin Jian reiterated China’s opposition to what it considers discriminatory trade policies. Quoted by the state-run Global Times, he stated: "China is willing to maintain dialogue and cooperation with all parties to maintain the stability of the global production and supply chain."

Huawei, under sweeping U.S. sanctions since 2019, has become a central figure in China's bid to decouple from American technology in critical sectors like AI. According to media, a “supernode” cluster connects AI processors via high-speed interlinks, enabling them to operate as a unified, high-performance computing system.

MENAFN18092025000045017169ID1110079290