Huawei Introduces OceanStor M900 Storage for AI Inference

Sep 17, 2026
Huawei introduced OceanStor M900 Context Memory Storage for AI inference in hyperscale data centers, with capacity of up to 64 PB per cluster.
Huawei Introduces OceanStor M900 Storage for AI Inference
Image: Huawei via PR Newswire

Huawei introduced OceanStor M900 Context Memory Storage at HUAWEI CONNECT 2026, the company said in a press release. The system gives AI computing clusters a shared memory space with capacity of up to 64 PB per cluster.

OceanStor M900 uses Huawei's UnifiedBus network to manage KV cache across on chip memory, DRAM, and SSDs. Its architecture provides direct NPU to SSD access, with claimed latency of 60 microseconds and aggregate bandwidth of 40 TB per second.

Huawei says the system can double inference cluster token throughput and reduce time to first token by half in typical AI programming workloads. Its storage technology distributes data based on predicted KV cache lifecycles and supports up to 24 drive writes per day.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

Guide
Private AI chip companies, by latest reported valuation

Private AI chip companies, by latest reported valuation

Private companies building AI compute silicon (the chips that run AI workloads), shown with each company's most recent reported or disclosed valuation. The figures are measured on different dates and bases, so this is a snapshot, not a live ranking or a like-for-like comparison.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.