Huawei Unveils Peerium Computing Architecture to Scale Data Center Processing
New architecture aims to link up to 1 million processors using UnifiedBus interconnect technology.
Quick Look
- Huawei has introduced its Peerium Computing Architecture, designed to connect up to 1 million processors into a single system.
- The technology utilizes UnifiedBus interconnects to bypass limitations imposed by US semiconductor manufacturing restrictions.
AI-generated summary
Why It Matters
Huawei is facing US restrictions that limit its access to advanced semiconductor manufacturing technologies. The company is pivoting to system-level computing power to maintain competitiveness.
Huawei Technologies on Thursday unveiled a new computing architecture that it says could make as many as 1 million processors work as a single computer, putting its UnifiedBus interconnect technology at the centre of an ambitious effort to rethink how large data centres are built.
The Peerium Computing Architecture combines nested parallelism, unified memory addressing and peer-to-peer interconnect, with UnifiedBus serving as the high-speed fabric connecting processors, memory, storage and networking equipment.
Huawei said an Atlas 950 SuperCluster containing 256,000 computing cards was already being deployed, while its next-generation Atlas 960 system using near-packaged optics, or NPO, was under testing.
Known in Chinese as Lingqu, UnifiedBus has become central to Huawei’s efforts to extract more computing power at the system level, a strategy that has taken on greater importance as US restrictions constrain the company’s access to advanced semiconductor manufacturing technologies.
What to Watch
AI outlook — possibilities, not facts
Deployment of Atlas 960 system following current testing phase.
Likely · Within months
Open Questions
- What is the power efficiency of the 1 million processor cluster?
- When will the Atlas 960 system be commercially available?
