Unveiling the OceanStor M900 Huawei also launched OceanStor M900 – a UnifiedBus-powered context memory storage cluster that supports direct one-hop access and provides a petabytescale KV cache for the L3.5 layer. It is designed for agentic inference with multi-tier KV caching for agent-heavy and longer-context workloads. It uses hybrid media and an optimised retention algorithm, which Huawei says extends SSD read / write lifespan by 16-fold.
The Shenzhen-based vendor adds that its Ascend chip series is the most vital component in its 11-chip UnifiedBus-powered portfolio for SuperPoDs and SuperClusters. The Ascend 960DT will be available in Q1 2027, three quarters ahead of the company’ s original roadmap, while the Ascend 960PR will be ready in Q3 2027, one quarter ahead of schedule.
“ We’ re evolving our Ascend chip series on a one-generation-a-year cycle,” said David.
“ In 2028 and 2029, we will roll out the Ascend 970 and 980 chips, respectively. Thanks to the Tau Scaling Law, not only will their compute specifications continue to double, but you can also expect to see huge improvements across the board in terms of memory.”
“ At Huawei, we are committed to building powerful AI infrastructure and going open source and open system to build out computing ecosystems,” David said.“ We will continue to support native training for mainstream foundation models on Ascend and actively embrace a wide range of models and applications. Together, let’ s lead the charge into the intelligent world.” WATCH NOW
Tau Scaling Law announced to the world
huawei. com 11