Systems news | The Register

REG AD

ON-PREM > SYSTEMS

Latest news and analysis of datacenter server systems

Memory godboxes could offer relief from the RAMpocalypse

Amid the AI-fueled memory crunch, will Compute Express Link finally have its moment to shine?

Datacenter to become Arm’s biggest business ‘soon’

Someone other than Meta is buying $1bn of its new AGI chips

Musk has never built a wafer fab, but he wants to burn $119B on one anyway

Initial phases of SpaceX's Terafab project in rural Texas are expected to cost about 1.25 Twitters

Qualcomm teases ‘dedicated CPU for agentic experiences’ and ‘agentic smartphones’

Enters the custom AI silicon business with secret silicon for an un-named hyperscaler

REG AD

Fujitsu confirms mainframe biz to die in 2035, in time for quantum AI supercomputers to take over

In talks with Japan, the UK, and Australia on defense tech that can ‘contribute to global stability’

ZTE-China-National-Clinical-Research-Center-for-Interventional-Medicine

ZTE partners with China's National Clinical Research Center for Interventional Medicine to build a new paradigm of smart interventional medicine

ZTE and NCRCIM leverage "CardioMind" AI and full-stack computing to drive global healthcare equity

Core Scientific accelerates crypto-to-AI pivot, converts Bitcoin mine to gigawatt-scale token farm

They were doing it in Texas...

AWS SVP Peter De Santis presents Graviton 3

Meta Arms itself to the teeth by signing for 'tens of millions' of AWS Graviton cores

After flubbing the Metaverse, Zuck embraces the Neoverse

Loading operation for a Boeing 747-400 Freight - for selfsame model - but for Cargolux Airlines at Milan Malpensa Airport back in 2016

AI now gobbling up power and management chips for servers

Bad news for multiple general server components as vendors switch to more lucrative gear

Musk bets Tesla's AI future on Intel node that isn't finished yet

EV maker leaning on still-in-development 14A process for Terafab, says it needs to build own silicon

SK Hynix’s aspirations for ’Merica-made HBM inch closer to reality

New site set to begin manufacturing and testing HBM memory just in time for Nvidia's Rubin-Ultra GPUs in 2028

Datacenter boom keeps dirty coal plants alive in the US

Happy Earth Day!

REG AD

AMD's Ryzen 9 9950X3D2 Dual Edition tested: Gratuitous overkill with a price to match

An $899 CPU? In this economy?

World's blandest man steps down from CEO job to spend more time in tastefully appointed home

Tim Cook is handing the reins to John Ternus at Apple

Intel eases reliance on TSMC with 'Merica-made Core Series 3 processors

Stripped-down Ultra for laptops and low-power edge boxes

A beginner's guide to GPU virtualization: passthrough, vGPU, and MIG

What every IT generalist needs to know before deploying GPU workloads, and why the platform matters more than the hardware.

Orbital datacenter startup CEO admits launch economics don't fly, presses ahead regardless

Needs SpaceX et al to drop prices and give competitors a ride into space to make it work

AI-powered mainframe exits are a bubble set to pop

Analysts reckon 70 percent of projects will fail, and 75 percent of vendors in the field will go away

Cloud-smart strategy helps Interactive meet GenAI demands

Hybrid cloud strategies emerge as the foundation for secure, AI-ready enterprises

Oracle taps Bloom for 2.8 GW of fuel cells to keep datacenter binge going

With grid hookups slow and turbines scarce, on-site power is starting to look less optional

Britain gives Rolls-Royce the nod to sketch out its mini reactor future

Contract kicks off design work, but SMRs unlikely to generate power before the mid-2030s

Rapidus has broken ground its 2nm wafer plant in Hokkaido

When the IBM PC and shoulder pads were big, Japan led the chip industry. It's trying to get back there now

Local hero Rapidus is on track to begin production of 2nm semis next year, as TSMC expands its Japanese foothold

Andy Jassy during AWS reinvent 2020 keynote

AWS ponders selling its home-grown chips by the rack-load, has almost sold out AI capacity

Annual CEO letter reveals two customers want all Graviton servers, huge drone rollout, a million robots, and more megalomania

NVidia aims to gives games power to handhelds

Nvidia's Rubin GPU is likely to be late thanks to memory shortage and technical challenges

China-bound Hopper accelerators are also likely to ship in smaller volumes than previously forecast, industry watchers say

Supermicro launches probe after staff charged with China export violations

Board-led inquiry follows indictment of two employees and a contractor over alleged diversion of Nvidia GPU servers

Elon Musk and Intel CEO Lip Bu Tan shake hands outside what appears to be the chipmaker's Arizona fab

Intel gets trapped in Elon’s reality distortion field as it joins in megafab delusions

Space is just the next stop on the AI hype train, right after AGI

Anthropic reveals $30bn run rate and plans to use 3.5GW of new Google AI chips

Broadcom's building the silicon and is chuffed about that, but also notes Anthropic remains a risk

How Nvidia learned to embrace the light in its quest for scale

The GPU king's move to optical scale-up was inevitable

IBM wants Arm software on its mainframes to better support AI

Tie-up aims to widen Big Blue’s access to power-efficient compute

AI server farms heat up the neighborhood for miles around, paper finds

Researchers say localized warming can extend well past site edges, raising concerns about community impact

REG AD

Intel DCG chief Kevork Kechichian speaks at ITT 2025

Arm says agentic AI needs a new kind of CPU. Intel's DC chief isn't buying it

Cores it's got what agents crave

Memory-makers' shares are down. Some RAM prices have eased. Blaming Google is not a good idea

Chocolate Factory boffins have found a way to reduce AI’s memory use, but don’t assume that means less demand for DRAM

US PC shipments to fall 13% as memory and storage crunch hits budget systems

Omdia says education, consumer, commercial, and public sector demand will weaken through 2026

REG AD

ZTE showcases end-to-end intelligent computing at CloudFest 2026 in Germany, empowering the digital future

AI boom fuels ZTE’s cloud pitch: more power, less cost, smoother upgrades

South Korean AI chip startup Rebellions eyes new shores for rack-scale invasion

Funding round comes ahead of planned IPO

Enterprise infrastructure is entering an economic reset

Here’s how to cope with the memory pricing squeeze

Three more charged over alleged Nvidia GPU smuggling scheme to China

Prosecutors say trio used Thai front companies to reroute high-end AI servers

Alibaba delivers RISC-V server chip optimized to run China’s top AI models

Claims its set performance records but looks to be years behind western fare

AI-pilled Arm CEO teases mystery products that will turn it into a money machine

Breaking free of its IP licensing shackles

Arm rolls its own 136-core AGI CPU to chase AI hype train

Turns out artificial general intelligence was a CPU this whole time

SoftBank to build massive AI datacenter on former US nuclear weapons site

10GW server farm, 10GW of new generation, and $4.2bn grid upgrade. And someone else is paying for the uranium cleanup

Explainer: AI-ready servers

GPUs and data management is important in an AI push, but don't ignore your legacy kit

Elon Musk and his greatest hits: Tesla, that guy in a robot suit

Elon Musk wants to build 50 times more chips than the world currently produces, using 'new physics'

Like his promise to get a million robocabs on the road, this doesn't add up

Australia to datacenter operators: BYO energy, pay your way, build green, or stay home

PLUS: Singtel’s triple outage; 17,000 counterfeit hard drives seized; Tech wages shift across Asia; And more!

Supermicro co-founder arrested, charged over $2.5B Nvidia GPU sales to China

Indictment claims dummy servers and bogus docs used to slip past US export controls

Jeff Bezos' rocket company Blue Origin applies to launch 51,000 datacenter satellites

‘Project Sunrise’ needs a network that doesn’t exist, a rocket that’s hardly flown, and FCC approval

Alibaba has made 470,000 AI chips, admits they’re inferior and may always be

Sees optimizing its entire cloud around homebrew silicon as the way to compete

REG AD

Your next car might need 300 GB of RAM, and so will autonomous robots

Micron plans to cash in, after already growing revenue $10 billion in a single quarter

Sierpiński fractals in a crystal : three is the magic number. Pic credit: Shutterstock

It's not a binary choice. Independent boffin builds a ternary CPU on an FPGA

Three is the magic number as first off-the-shelf general-purpose ternary hardware since c 1965 lands

Nvidia's on-again off-again H200 sales in China are now on again

Beijing appears to have eased its policy of pushing local GPUs

Meta reveals four Broadcom-built custom AI chips, claims some outperform commercial silicon

Deploying them by the gigawatt but still can’t flag obvious AI slop

ZTE and Orange Morocco launch Livebox 7, world's first prplOS 4.0 ONT for smart home

Livebox 7 brings Morocco’s first prplOS 4.0 ONT, enabling all-scenario smart home experiences

In this render, Ayar shows how it imagnies optical interconnects might be integrated into a GPU platform for scale out compute.

Ayar Labs taps Wiwynn to cram 1,024 GPUs into a photonic rack system

Reference design to stitch more than a thousand accelerators into a single enormous server.

ZTE and Whale Cloud jointly showcase digital transformation solutions at MWC Barcelona 2026

ZTE and Whale Cloud showcase AI-ready NaaS and open-standard platforms to help operators accelerate digital transformation and service innovation

New York skyline at dusk with Empire State Building lit up

AI datacenters may gulp a New York City's worth of water on hot days

Study warns peak cooling demand could strain US water systems by 2030

Mystery outage behind US airline JetBlue asking FAA to ground its flights

Aircraft on the ground briefly halted until systems were up again

HPE tweaks T&Cs so the price it quotes may not be the price you pay

With memory and storage contributing over half the price of a server, Big Green needs to protect its margins

Beijing warns of more chip supply worries after Nexperia China claims it was cut off from SAP

PLUS: Indonesia joins kids social media ban; China frets about AI job impacts; India’s PC market fails to launch, again; And more

Unpacking the deceptively simple science of tokenomics

Inference at scale is much more complex than more GPUs, more tokens, more profits

OpenAI CEO Sam Altman speaking in a Senate hearing

Oracle and OpenAI's Texas Stargate datacenter expansion reportedly on the skids

Meta supposedly considering untapped capacity in deal brokered by Nvidia

Washington reportedly moves to tighten leash on AI chip exports

Draft rules could force Nvidia and AMD to seek government approval before selling abroad

China’s rubber-stamp parliament rubber stamps tech independence plan

Call to do better with chips and put AI everywhere is more than rhetoric because China’s scientists are sprinting ahead

Broadcom says AI companies can’t make their own silicon any time soon

Offers booming customer accelerator biz as evidence, while VMware props up its software business

Intel numbers boss swears big Foundry wins are coming soon

Meanwhile Chipzilla's 18A process tech could see external deployment after all

One vendor doesn't mind high RAM prices: VMware

Memory tiering and pooled memory are having a moment because they offer the chance to use less RAM

Nvidia-backed photonics startup Ayar Labs fills its wallet to mass-produce CPO chiplets

Company aims to stitch tens of thousands of GPUs together for more efficient training and inference

Singapore eyes barge-based hydrogen power for datacenters

Saves real estate by putting the power on the water

Illustration of an AI brain in a light bulb

Extreme Co-design, Building the AI Factory: ZTE Unveils Full-Stack AI Infrastructure

From vertical scaling to digital twins, ZTE redefines AI infrastructure with end-to-end compute-network and software-hardware co-design

PCs and phones to get more boring and expensive in 2026 thanks to memory drought

'This is perhaps the biggest challenge the industry has faced since its inception'

Oak Ridge spawns institute to curb AI datacenter power surge

Lab aims to link power, cooling, and workload management to ease strain on the US grid