James Carroll

SK hynix and Sandisk have jointly unveiled the industry’s first standard specification for High Bandwidth Flash (HBF), marking a significant milestone in the evolution of AI memory architectures. Announced at FMS 2026 in Santa Clara, the specification establishes an open framework for a new storage tier positioned between high-bandwidth memory (HBM) and traditional SSDs, addressing one of the […]

CoreWeave plans three AI data center facilities in Indonesia totaling 360 MW of contracted IT power, marking its first expansion into the Asia-Pacific region.

FuriosaAI, I/ONX HPC and Velox plan a 15 MW Stockholm AI data center using Furiosa RNGD accelerators alongside GPUs for energy-efficient inference.

MediaTek outlined an advanced packaging and connectivity roadmap spanning CoWoS, EMIB-T, 3.5D integration, 448G SerDes, co-packaged copper and CPO as it prepares to scale two generations of custom AI accelerator ASICs.

Sunil Sadani and Dmitry Shokarev discuss how AI is transforming network infrastructure requirements and the routing solutions designed to meet these demands. The conversation explores the evolution from AI training to AI consumption, and the critical role of data center interconnect in supporting distributed GPU clusters. The presenters examine the PTX router portfolio and its […]

Dale Brown, Global Head of Growth AI Solutions at HPE, explores critical decisions enterprises face when deploying AI workloads and reveals how organizations are rethinking their AI infrastructure strategies. Brown discusses the evolving landscape of AI deployment, from initial public cloud adoption to the growing trend of workload repatriation, and examines why companies are increasingly […]

Lightstorm secured a ₹2,500 crore, fully underwritten 10-year debt facility from IndusInd Bank to accelerate an AI-focused terrestrial and subsea fiber corridor linking India with Singapore and Malaysia.

Dell’Oro Group forecasts the data center IT semiconductor and components market will reach $1.8 trillion by 2030 as AI investment expands across accelerators, memory, storage, CPUs and NICs, while component power consumption exceeds 200 GW.

Agentic AI and inference are expanding AI networking demand into front-end data center networks, with Dell’Oro forecasting AI-related opportunities will drive more than half of front-end switch sales growth and accelerate 800G and 1.6T adoption.

Global Switch has started construction of its 56 MW Madrid 2 data center, expanding its Madrid campus to 70 MW to support AI, cloud and high-performance computing workloads.