Monday, 30 June 2025
29.3 C
Singapore
32.2 C
Thailand
20.6 C
Indonesia
28.5 C
Philippines

Western Digital and Ingrasys collaborate on advanced fabric-attached storage for AI workloads

Western Digital and Ingrasys team up to develop a TOR switch with embedded storage, targeting AI-ready disaggregated data centre solutions.

Western Digital and Ingrasys, a subsidiary of Foxconn Technology Group, have announced a long-term partnership to develop a next-generation Top-of-Rack (TOR) switch with embedded storage. The collaboration aims to address the growing demands of artificial intelligence (AI) workloads by introducing high-performance, fabric-attached disaggregated storage solutions.

The new TOR Ethernet Bunch of Flash (EBOF) switch is designed to deliver distributed storage at the network edge, reducing latency by avoiding the need for centralised storage arrays. It removes the dependency on separate storage networks and brings data closer to compute resources. Ingrasys will handle manufacturing, while Western Digital will lead the architecture and market strategy using its RapidFlex NVMe-oF (Non-Volatile Memory Express over Fabrics) bridge technology.

Enabling flexible disaggregated infrastructure

This partnership marks a key milestone in the AI infrastructure space, with both companies aiming to promote a scalable and efficient storage architecture for data centres. Ingrasys brings extensive experience in building GPU servers, while Western Digital contributes its expertise in fabric-attached storage.

The TOR EBOF switch is set for release in 2027 and will feature Western Digital’s next-generation RapidFlex Fabric bridge device. The solution supports 100G Ethernet connectivity and NVMe/PCIe Gen6 interfaces for E3.S/L solid-state drives (SSDs), enabling seamless integration of flash storage into the networking layer. The system will be powered by the NVIDIA Spectrum-4 switch ASIC, providing high-performance switching with flexible support for 400 and 800GbE cabling to meet future data centre demands.

Western Digital’s RapidFlex NVMe-oF bridge technology stands out for its hardware-accelerated design, which removes firmware from the performance path. This approach allows I/O read and write payloads to pass directly through the adapter, reducing latency while maintaining direct Ethernet connectivity. As a result, the system allows data centres to scale storage independently from compute resources, improving flexibility and efficiency.

A shared vision for next-generation AI data centres

Executives from all three companies involved have expressed optimism about the strategic impact of this collaboration. Kurt Chan, vice president and general manager of Western Digital’s Platforms Business, stated, “Together with Ingrasys, we continue to accelerate the shift toward disaggregated infrastructure by co-developing cutting-edge, fabric-attached solutions designed for the data demands of AI and modern workloads. This collaboration brings together two leaders in storage infrastructure modernisation to deliver flexible, scalable architectures that unlock new levels of efficiency and performance for our customers.”

Benjamin Ting, president of Ingrasys, echoed the sentiment, adding, “Our collaboration with Western Digital reflects a shared commitment to long-term innovation and customer-centric design. By combining our expertise in scalable system integration with Western Digital’s leadership in storage technologies, we’re building a foundation for future-ready, fabric-attached solutions that will meet the evolving demands of AI and disaggregated infrastructure.”

Gilad Shainer, senior vice president of networking at NVIDIA, also commented on the partnership’s significance: “The collaboration between Western Digital and Ingrasys brings together the high-performance storage and scalable system transformation needed to fully unlock the potential of accelerated computing. As AI and data-intensive workloads push infrastructure limits, this joint effort is set to deliver the performance, low latency, and disaggregated scalability that next-generation data centres require.”

This development underscores the importance of close industry collaboration in addressing the infrastructure needs of AI and advanced computing environments.

Hot this week

AWS and DISG launch AI Springboard to help 300 Singapore enterprises adopt AI

AWS and DISG launch AI Springboard to help 300 Singapore enterprises scale AI adoption with funding, training, and technical support.

Salesforce launches Agentforce 3 to improve control and visibility over AI agents

Salesforce launches Agentforce 3 with new tools for monitoring, interoperability and performance to help enterprises scale AI agents confidently.

Nintendo’s GameCube controller returns for Switch 2 with new features

Nintendo’s GameCube controller for the Switch 2 is back in stock with wireless upgrades, GameChat support, and USB-C charging.

HPE launches modular AI factory solutions in partnership with NVIDIA

HPE expands its AI factory solutions with NVIDIA to simplify enterprise AI adoption through integrated infrastructure, software, and services.

Garmin launches quatix 8 smartwatch for mariners in Singapore

Garmin introduces quatix 8 in Singapore, a rugged smartwatch for mariners with marine tools, AMOLED display, and long battery life.

Cheapest SIM-only plans in Singapore 2025: Flexible, contract-free mobile data

Compare the cheapest SIM-only plans in Singapore for 2025, with up to 1TB data, 5G access, roaming, and no-contract options from S$8/month.

Android 16 to alert you if your phone connects to a fake cell tower

Android 16 will warn you if your phone connects to a fake tower, helping protect your calls, texts, and location from silent spying.

YouTube Create is finally coming to iOS devices

YouTube Create is coming to iOS, offering free mobile video editing tools as Google aims to catch up with CapCut and InShot.

OpenAI turns to Google’s AI chips in the shift from Microsoft and Nvidia

OpenAI begins renting Google's AI chips to run ChatGPT, shifting away from Microsoft and Nvidia to lower computing costs.

Related Articles

Popular Categories