Google’s Virgo fabric to connect A5X instances, scaling AI up to 960,000 distributed GPUs

Shares in the company rose 20% after results were published

Taiwanese firm is a key manufacturer in the Vera Rubin value chain, and said to be ahead of schedule with its efforts

Everpure rival cries profiteering while customers foot the bill

Managed Lustre gets inference upgrade as customers move from training to deployment

DeepMind achieves 'zero global downtime' for distributed AI training

CEO defends delicate pricing moment for Everpure and promises it's not profiteering

Financial terms of the deal have not been disclosed

Stars align for a celestial-named fabric that promises 40% lower unloaded latency and four-times bandwidth per accelerator

Prototype converts between encoding modalities with under 4% fidelity loss

The Aptiv-owned firm joins the CHERI Alliance

Report finds average GPU utilization at 5% across enterprise clusters

Legislation extends the National Quantum Initiative to 2034, adding new centers, testbeds, and supply chain measures

The National Oceanic and Atmospheric Administration (NOAA) to expand N-Wave

Follows announcement that Broadcom will develop Google TPUs until 2031

And it does not expect to be a near-term player in the AI and data center party

Is the Claude developer bolstering already diverse compute stack with MI450?

Wanted: Engineer to wrangle real scale and large spend as terabit backbone sprawl gets out of hand

Its technology is geared toward ensuring data center architectures can evolve with advancing AI models

Will it offer GPShoe-as-a-service?

Deal comes a month after Meta unveiled next four generations of its AI training and inferencing chips

Partners see perk reinstated as AI memory surge shows no signs of abating

New DRAM models does away with the capacitor

CAS approach helps reduce ingestion time and search latency, and enable scaling