California Data Center Intel

Recent California data center news

• Mar 18, 2026

  Jensen Huang After the Keynote: Inside Nvidia’s GTC 2026 Press Briefing

  Nvidia (Jensen Huang) framed the company as building “AI factories,” emphasizing inference, token economics, and ecosystem orchestration.

  • Main announcement: Jensen Huang presented Nvidia’s strategic shift to an inference- and token-centric model, arguing data centers must become “token manufacturing systems”, and confirming Nvidia is helping finance customer data center buildouts with partners including CoreWeave, Nscale, and Nebius. He cited a “trillion-dollar visibility” figure for Blackwell and Vera Rubin through 2027 (figure described as “trillion-dollar visibility” in the briefing).
  • Background and other details: Huang detailed Nvidia’s broader infrastructure role: co-inventing co-packaged optics with TSMC (~100 patents filed across the supply chain), production ramping for co-packaged optics, resumed manufacturing and license activity for H200 systems in China, and noted the complexity of relocating 40% of Taiwan chip capacity to the United States as a longer-term goal.
• Mar 17, 2026

  Suman Kanuganti: Why Telcos, Not Tech Giants, Will Power the AI Grid

  NVIDIA is positioning the “AI Grid” and the piece urges telcos to build identity-bound AI memory using existing telco networks.

  • Main announcement/action: NVIDIA is promoting an “AI Grid” at NVIDIA GTC to move AI workloads onto networks and enable distributed, identity-bound memory; the article asserts telcos are best positioned to deliver this by leveraging identity, authentication, and billing capabilities and placing memory near the tower/device/person.
  • Background and details: The piece is an Expert Opinion by Suman Kanuganti, Co-Founder and CEO of Personal AI (since 2020); it highlights Personal Language Models (PLMs) as memory-first AIs, claims Personal AI is NVIDIA’s top independent software vendor (ISV) for telco and that PLMs deliver 10-30x better cost efficiency versus traditional large language models; it references NVIDIA GTC as the event illuminating this vision.
• Mar 17, 2026 · RMI

  The Interconnection Queue Continues to Be a Barrier to American Economic Competitiveness

  The U.S. Department of Energy directed FERC to initiate a rulemaking for load interconnection, and the authors (RMI) call for expanded, nationwide reforms to generator interconnection to speed studies and reduce costs.

  • Main action & proposed reforms: The article urges scaling existing reforms nationally (building on FERC Order 2023, issued summer 2023) and advancing additional measures that speed study processing and reduce network upgrade costs (e.g., advanced software automation, evaluation of grid-enhancing technologies (GETs), transparency on network upgrade timelines, co-located load/generation planning, and fair competitive fast-tracks).
  • Key facts & context: The piece cites >2.2 terawatts in interconnection queues, an average interconnection timeline of nearly five years in 2024 (vs under two years in 2008), that only 19% of projects requesting interconnection between 2000–2019 had reached commercial operation by end of 2024, and an RMI analysis that incorporating GETs in PJM could enable 6.6 GW of new generation and yield ~$1 billion annually in production-cost savings at a $0.1 billion installation cost.
• Mar 17, 2026 · IT Brief Asia

  Zadara aligns AI cloud platform with NVIDIA security guide

  Zadara has announced alignment of its multi-tenant AI cloud platform with NVIDIA’s Software Reference Guide.

  • Alignment scope and implementation: Zadara says its platform implements NVIDIA design expectations including full-stack workload isolation, network partitioning, per-tenant data volume separation, VM-based GPU tenancy, and tenant-controlled orchestration patterns; Zadara will present this approach at NVIDIA GTC in San Jose (booth 7034).
  • Background, partners and deployment details: The offering targets cloud providers, telecoms operators, and managed service providers, supports NVIDIA Spectrum-X Ethernet and NVIDIA BlueField DPUs, operates more than 500 edge clouds globally, and is being brought to market via partners including Micro Support Group (US northeast) and BCN (which says it deployed Africa’s first Neutral AI Factory).
• Mar 17, 2026

  Berkeley Lab Takes Major Step Toward Doudna with Delivery of Early Access System, Cech

  NERSC has announced delivery and installation of the NERSC-10 Pilot Early Access System (EAS) named Cech at Berkeley Lab to prepare for the full-scale NERSC-10 production system, Doudna.

  • Main announcement: NERSC received the NERSC-10 Pilot Early Access System (EAS) called Cech, which is being installed at Lawrence Berkeley National Laboratory; the EAS is a smaller test system to refine assembly, delivery, installation, and integration processes ahead of deploying the full production system Doudna in late 2026 (also referenced as “end of the year” in comments). Key technical details: 72 NVIDIA Grace CPUs, 144 NVIDIA Blackwell GPUs, 5.76 petaflop/s FP64 and 1.44 exaflop/s NVFP4 capability, networking via NVIDIA Quantum-X800 InfiniBand, Dell IR7000 direct liquid-cooled racks, first U.S. deployment of Rittal in-row V3.5 coolant distribution unit, and Dell’s PowerCool Enclosed Rear Door Heat Exchanger (eRDHx) claiming up to 60% reduced cooling costs versus traditional options.
  • Background & implementation details: The EAS (Cech) is a public–private partnership involving NERSC, Dell Technologies, NVIDIA, and VAST Data; it will be used for software stack development (Omnia, OpenCHAMI), benchmarking, telemetry/monitoring, storage integration with Community File System and HPSS, and ESnet connectivity. The EAS is explicitly a testing/development milestone (not yet available to users) to ensure a smooth deployment of the full Doudna production system; the article is an announcement of this delivery and installation milestone.
• Mar 17, 2026

  Jensen Huang Maps the AI Factory Era at NVIDIA GTC 2026

  Nvidia CEO Jensen Huang announced that AI is entering an infrastructure phase and unveiled hardware, software, and reference architectures to build gigawatt-scale “AI factories” for continuous inference.

  • Main announcement: Nvidia unveiled new infrastructure components including Grace Blackwell NVLink72, Vera Rubin (rack-scale systems with ~3.6 exaflops per rack and 45°C hot-water liquid cooling), the Vera Rubin DSX AI Factory reference architecture and Omniverse DSX digital-twin blueprint, and software layers OpenClaw / Nemo / Nemotron to orchestrate and secure agentic AI systems; Nvidia estimated a $1 trillion AI infrastructure market and cited an industry shift to continuous inference.
  • Details & partners: Nvidia described hybrid architectures integrating Groq accelerators (disaggregated inference via Dynamo orchestration), a production co-packaged optical switch built with TSMC, DSX integrations with partners (Cadence, Dassault Systèmes, Schneider Electric, Siemens, Vertiv, Trane Technologies, Switch) and energy partners (GE Vernova, Siemens Energy, Hitachi Energy, Emerald AI); the keynote also referenced venture funding > $150 billion for AI startups and examples like Nestlé reducing compute costs by 83% on a GPU-accelerated workload.
• Mar 17, 2026

  Chip wafer shortage will run through 2030 as AI demand overwhelms supply: SK Hynix chief

  SK Group Chairman Chey Tae-won warned the global semiconductor wafer shortage could continue until 2030.

  • Main announcement: Chey said the industry faces a wafer deficit of more than 20% and expects at least four to five years of capacity building; he attributed the squeeze to AI-driven HBM demand and said SK Hynix (holding 57% of global HBM and 32% of overall DRAM) is preparing a plan to stabilise DRAM prices (details not disclosed).
  • Background/details: Industry analysts (Greyhound Research, Gartner, IDC) describe the issue as a structural reallocation of memory toward AI workloads; IDC projects 2026 DRAM and NAND supply growth of 16% and 17%, Samsung’s P5 facility is expected online by 2028, and new fab capacity will be largely optimised for AI workloads, limiting near-term relief for conventional enterprise demand.
• Mar 17, 2026

  United States Advanced Thermal Tech Company Frore Systems Raised $143 Million in Series D Funding Round at $1.64 Billion Valuation, Founded in 2018 by Seshu Madhavapeddy & Surya Ganti, Investors Include MVP Ventures, Fidelity Management & Research Company, Top Tier, Mayfield Fund, Clear Ventures, Addition, Qualcomm Ventures, StepStone Group & Alumni Ventures

  Frore Systems has raised $143 million in a Series D funding round at a $1.64 billion valuation (announced 17 March 2026, dateline Hong Kong).

  • Main announcement:Frore Systems has raised $143 million in a Series D round at a $1.64 billion valuation; the company was founded in 2018 by Seshu Madhavapeddy and Surya Ganti and the investors named include MVP Ventures, Fidelity Management & Research Company, Top Tier, Mayfield Fund, Clear Ventures, Addition, Qualcomm Ventures, StepStone Group, and Alumni Ventures.
  • Background and product details: Frore Systems is headquartered in Silicon Valley with manufacturing operations in Taiwan; flagship products include LiquidJet™, LiquidJet™ Nexus, and AirJet®, which are positioned for data center and edge cooling (claims include improved GPU performance, better PUE, and reduced TCO); the article is an announcement/press-style report published on Caproasia on 17 March 2026.
• Mar 16, 2026

  NVIDIA DSX Air Boosts Time to Token With Accelerated Simulation for AI Factories

  NVIDIA announced NVIDIA DSX Air at GTC 2026, a SaaS simulation platform to build full-stack digital twins of AI factories and accelerate deployments.

  • Main announcement:NVIDIA DSX Air is a software-as-a-service platform that provides high-fidelity digital simulations of NVIDIA hardware (GPUs, SuperNICs, DPUs, switches) and integrates with partner solutions via open, API-based connectivity, enabling customers to validate compute, networking, storage, orchestration and security and reduce time to first token from weeks/months to days/hours.
  • Background & implementation details: Partners and customers (examples: CoreWeave, Siam.AI, Hydra Host) are already using DSX Air; GTC demos included a multi-tenant RTX PRO Server simulation with Netris (network orchestration), Rafay (host orchestration), NVIDIA Run:ai (GPU allocation), a VAST AI Operating System video RAG pipeline demo, and security validations with Check Point, TrendAI Vision One, and Keysight running on simulated BlueField DPUs.
• Mar 16, 2026

  Palantir partners with Nvidia to streamline AI data center deployment

  Palantir Technologies and Nvidia have announced the Palantir AI OS Reference Architecture (AIOS-RA).

  • Main announcement: The AIOS-RA is an end-to-end reference architecture and operating system designed to support processes from hardware acquisition to application deployment, running both training and inference on Nvidia Blackwell Ultra systems (each system incorporates eight Blackwell Ultra GPUs and Spectrum-X Ethernet networking). It is built on a hardened Kubernetes substrate and integrates Palantir’s software suite (AIP, Foundry, Apollo, Rubix, AIP Hub) to serve as a blueprint for designing, deploying, and scaling high-performance on-premises, edge, and sovereign cloud AI factories.
  • Additional details / background: Management and security are handled via Palantir Rubix (zero-trust Kubernetes security) and Apollo (autonomous deployment and lifecycle oversight). The collaboration also includes Nvidia’s stack (Nvidia AI Enterprise, CUDA-X Libraries, Nemotron open models, Magnum IO) and targets customers with existing GPU infrastructure, latency-sensitive workflows, data sovereignty requirements, and high geographic distribution.