Hajra Naz
Huawei, the Chinese telecom giant, has revealed major new plans in artificial intelligence (AI) chips and computing power. The company wants to challenge U.S. tech giant Nvidia, which currently dominates the global AI chip market.
At the annual Huawei Connect conference, rotating chairman Eric Xu announced that Huawei will speed up its computing roadmap. He said the company will release new AI computing clusters every year. Each release will aim to double the computing power of the previous one.
Atlas 950 SuperCluster: A Major Leap
Huawei’s first step will be the launch of the Atlas 950 SuperCluster. It is expected to arrive as early as next year. This system will use more than 500,000 Ascend AI chips working together at high speed.
For comparison, Huawei’s earlier version, the Atlas 950 Supernode, supported 8,192 chips. The new cluster is a huge jump forward in scale and power.
Read More: China Blocks Nvidia AI Chip Sales Amid US Tech Tensions
Future Roadmap: Atlas 960 and 970
Huawei also shared its long-term plans. The Atlas 960 SuperCluster will launch in 2027. It will use 15,488 Ascend chips per node, pushing performance even higher.
By 2028, the company plans to release the Atlas 970 SuperCluster, a system expected to include over 1 million interconnected chips. These supernodes will use Huawei’s strength in networking to deliver massive computing power.
Why Huawei Is Moving Fast
China is currently facing restrictions that prevent its companies from buying Nvidia’s advanced AI chips. In response, Huawei is building its own large-scale systems. Experts say the company is also using China’s strong power supply network to make these ambitious projects possible.
Wang Shen, a data center expert at Omdia, noted that Huawei’s networking expertise gives it an edge in creating large clusters. This approach may help China reduce its dependence on foreign chip makers.
A Shift in Global Tech Competition
The move comes as China investigates Nvidia for alleged violations of antimonopoly law. Tensions between the U.S. and China over advanced technology have made AI and chip production a key battleground.
Huawei’s push into AI computing shows that it aims to be a global leader in the next wave of artificial intelligence. If successful, these superclusters could power industries such as healthcare, robotics, autonomous vehicles, and national research projects.
Read More: Amazon Launches Smart AI Assistant for Marketplace Seller Support
FAQs
1. What is the Atlas 950 SuperCluster?
It is Huawei’s upcoming AI computing system. It will use over 500,000 Ascend AI chips, making it one of the most powerful clusters in the world.
2. Why is Huawei developing these AI clusters?
China faces restrictions on buying U.S. chips like Nvidia’s. Huawei wants to create its own powerful systems to stay competitive.
3. When will the Atlas 960 and Atlas 970 launch?
The Atlas 960 is planned for 2027, and the Atlas 970 is expected in 2028.
4. How are Huawei’s chips different from Nvidia’s?
Nvidia dominates global AI chips today. Huawei’s Ascend chips are designed to compete by powering massive supercomputing clusters.
5. What industries could benefit from these clusters?
Healthcare, finance, transportation, robotics, and scientific research could all use the computing power from Huawei’s AI systems.
