惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

F
Fortinet All Blogs
有赞技术团队
有赞技术团队
量子位
N
Netflix TechBlog - Medium
博客园 - 叶小钗
博客园 - 三生石上(FineUI控件)
Google DeepMind News
Google DeepMind News
aimingoo的专栏
aimingoo的专栏
GbyAI
GbyAI
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
Blog — PlanetScale
Blog — PlanetScale
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
月光博客
月光博客
Martin Fowler
Martin Fowler
Y
Y Combinator Blog
宝玉的分享
宝玉的分享
博客园 - 司徒正美
云风的 BLOG
云风的 BLOG
V
Visual Studio Blog
V
V2EX
IT之家
IT之家
L
LangChain Blog
大猫的无限游戏
大猫的无限游戏
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More

AI demand is so high, AWS customers are trying to buy out its entire capacity | Network World

Cisco: Latest news and insights 2026 network outage report and internet health check Selector targets the network visibility gap in multi-cloud infrastructure AI reshapes cybersecurity workforce priorities as IT teams brace for new risks Top network and data center events of 2026 How AI is transforming network incident response (and where it still falls short) Google opens TPUs to enterprises beyond its own cloud via Blackstone JV AI, cybersecurity skills top IT pay premiums Startup Bolt Graphics promises 5x performance over Nvidia’s best GPU Wireless security is a battle of AI vs. AI NetOps teams look to AI to automate Day 2 operations Digital twins reshape network and data center management Network outages, power failures strain data center resiliency Five takeaways from Cisco's blowout quarter and what it means to customers Cisco to cut nearly 4,000 jobs despite strong growth in AI, enterprise networking Hard drive shortage affecting enterprise storage needs Wi-Fi 8 is closer than you think. Here’s what you need to know Cisco open-sources agentic AI security spec HPE revamps private cloud stack for enterprises rethinking VMware Versa takes aim at fragmented enterprise security with CSPM, orchestration update, and AI agent controls Red Hat opens Ansible to AI agents, within limits Red Hat offers endless Linux support — for a fee Red Hat: Sovereignty is more than just compliance Tech job postings hit three-year high as AI demand fuels hiring rebound HPE memory server targets compute-heavy and agentic AI workloads PCI group begins work on new spec to support bandwidth-hungry apps like AI, HPC Q&A: Quantum physicist Sonia Fernández-Vidal on why classical computing isn't going anywhere OpenAI-led consortium seeks to address AI processing bottlenecks AWS hit by US-East-1 outage after data center thermal event Gluware's Titan rises to meet Mythos network vulnerability challenge
Startup SPAN teams with Nvidia to put data center nodes i...
2026-05-14 · via AI demand is so high, AWS customers are trying to buy out its entire capacity | Network World

Even though many communities are in open revolt against massive data centers needed to power artificial intelligence, a startup is betting that homeowners will welcome miniature data centers in their backyards.

The startup, an intelligent power management company called SPAN, has partnered with Nvidia and homebuilder PulteGroup to make use of spare electrical transmission capacity already available in many neighborhoods—something SPAN says its smart panels can detect.

Rather than building massive new data centers with their own ZIP code, SPAN is proposing a network of small units, called XFRA nodes, installed outside of homes or in small commercial locations. These nodes are no bigger than an HVAC or power generator found outside of any home, according to SPAN.

A SPAN representative told CNBC that it can install 8,000 XFRA units about six times faster and at five times lower cost than the construction of a typical centralized 100-megawatt data center of the same size.

“One big reason the XFRA model works is that the average American home only uses about 40 percent of its electrical capacity,” a SPAN spokesperson told Realtor.com. “As big data center developers struggle to find power sources and distribution capacity, XFRA uses capacity that’s already available.”

The hardware going into these devices is not exactly modest. The device packs 16 RTX6000 cards, four AMD Epyc CPUs, and 3TB of DDR5 memory. The cards are liquid cooled, and the design is meant to minimize sound, a major complaint of people living near data centers.

That’s some serious and expensive hardware. The memory alone would cost almost $100,000. The RTX6000 cards are between $9000 and $10,000 each, and the Epyc processors are anywhere from $8500 to $14,000 each.

The SPAN spokesperson told CNBC the exact arrangement will vary from one neighborhood or region to the next, but it’s likely that SPAN will take on paying the host’s electricity and internet bills directly and charge a flat fee every month that’s much lower than what the host would otherwise pay to their electric utility and internet service provider.

Span’s role in this installation typically includes a smart panel, an outdoor XFRA unit, a backup battery, and sometimes solar panels. XFRA is SPAN’s creation, a distributed data-center concept that uses underutilized power capacity in homes and small commercial buildings to run AI compute nodes closer to end users.

Alex Cordovil, senior analyst for infrastructure at the Dell’Oro Group, says the device is worth taking seriously, but the realistic ceiling is narrower.

“The potential is real where homes pair smart panels with solar and battery storage,” he said. The economics only stack up if these nodes consume locally generated surplus that would otherwise flow back to the grid at a low feed-in tariff.”

Cordovil notes that AI accelerators are an expensive ticket for the average homeowner, they perform best in tightly coupled clusters rather than single-rack islands, the hardware is iterating rapidly, servicing a dispersed fleet is costly, and the security model of compute bolted to a residential wall is very different from a Tier III facility.

“A more instructive parallel is how telcos are positioning their existing footprint for AI inference at the edge — they already have power, connectivity, security and a distributed node structure, but still wrestle with running compute across a small number of GPUs per site,” Cordovil said.

He concluded this can have a future as a complement to large campuses with thousands of GPUs, not a replacement.

SUBSCRIBE TO OUR NEWSLETTER

From our editors straight to your inbox

Get started by entering your email address below.