Tag: chip
-
The Register: Congress wants to know if Nvidia superchips slipped through Singapore to DeepSeek
Source URL: https://www.theregister.com/2025/04/17/congress_demands_nvidia_explain_deepseek/ Source: The Register Title: Congress wants to know if Nvidia superchips slipped through Singapore to DeepSeek Feedly Summary: As Huang jets to Middle Kingdom after H20 ban forces $5.5B hit Nvidia’s troubles with the US government have just begun: The day after the Trump administration’s export restrictions on its AI chips triggered…
-
Slashdot: Nvidia To Make AI Supercomputers in US for First Time
Source URL: https://news.slashdot.org/story/25/04/14/1352243/nvidia-to-make-ai-supercomputers-in-us-for-first-time?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Nvidia To Make AI Supercomputers in US for First Time Feedly Summary: AI Summary and Description: Yes Summary: Nvidia’s initiative to manufacture AI supercomputers domestically marks a significant move towards bolstering the U.S. AI infrastructure. This expansion not only aims to meet the escalating demand for AI hardware but…
-
The Register: Google offers 7th-gen Ironwood TPUs for AI, with AI-inspired comparisons
Source URL: https://www.theregister.com/2025/04/10/googles_7thgen_ironwood_tpus_debut/ Source: The Register Title: Google offers 7th-gen Ironwood TPUs for AI, with AI-inspired comparisons Feedly Summary: Sure, we’re doing FP8 versus a supercomputer’s FP64. What of it? Cloud Next Google’s seventh-generation Tensor Processing Units (TPU), announced Wednesday, will soon be available to cloud customers to rent in pods of 256 or 9,216…
-
Cloud Blog: GKE at 65,000 nodes: Evaluating performance for simulated mixed AI workloads
Source URL: https://cloud.google.com/blog/products/containers-kubernetes/benchmarking-a-65000-node-gke-cluster-with-ai-workloads/ Source: Cloud Blog Title: GKE at 65,000 nodes: Evaluating performance for simulated mixed AI workloads Feedly Summary: At Google Cloud, we’re continuously working on Google Kubernetes Engine (GKE) scalability so it can run increasingly demanding workloads. Recently, we announced that GKE can support a massive 65,000-node cluster, up from 15,000 nodes. This…
-
The Register: Lightmatter says it’s ready to ship chip-to-chip optical highways as early as summer
Source URL: https://www.theregister.com/2025/04/01/lightmatter_photonics_passage/ Source: The Register Title: Lightmatter says it’s ready to ship chip-to-chip optical highways as early as summer Feedly Summary: AI accelerators to see the light, literally Lightmatter this week unveiled a pair of silicon photonic interconnects designed to satiate the growing demand for chip-to-chip bandwidth associated with ever-denser AI deployments.… AI Summary…
-
The Register: Arm reckons it’ll own 50% of the datacenter by year’s end
Source URL: https://www.theregister.com/2025/04/01/arm_datacenter_cpu_market/ Source: The Register Title: Arm reckons it’ll own 50% of the datacenter by year’s end Feedly Summary: Optimistic much? Arm expects to see its architecture account for half of the datacenter CPU market by the end of this year, up from 15 percent in 2024, all thanks to the AI boom.… AI…
-
The Register: Nvidia challenger Cerebras says it’s leaped Mid-East funding hurdle on way to IPO
Source URL: https://www.theregister.com/2025/03/31/cerebras_ipo_roadblock/ Source: The Register Title: Nvidia challenger Cerebras says it’s leaped Mid-East funding hurdle on way to IPO Feedly Summary: Wafer-scale AI chip startup apparently smoothed over American concerns around UAE’s G42 planned stake AI chip startup Cerebras Systems says it has cleared a key hurdle ahead of its planned initial public offering…