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

Engineering at Meta
Engineering at Meta
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
腾讯CDC
宝玉的分享
宝玉的分享
量子位
Recent Announcements
Recent Announcements
Martin Fowler
Martin Fowler
J
Java Code Geeks
V
Visual Studio Blog
阮一峰的网络日志
阮一峰的网络日志
Blog — PlanetScale
Blog — PlanetScale
大猫的无限游戏
大猫的无限游戏
博客园 - 叶小钗
S
SegmentFault 最新的问题
B
Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
博客园 - 【当耐特】
小众软件
小众软件
The Cloudflare Blog
Y
Y Combinator Blog
I
InfoQ
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
GbyAI
GbyAI
IT之家
IT之家

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 Startup SPAN teams with Nvidia to put data center nodes in your backyard 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
Supply constraints, optical advances dominate Arista's Q1
2026-05-07 · via AI demand is so high, AWS customers are trying to buy out its entire capacity | Network World

It’s mostly good news for Arista Networks, which reported total first-quarter revenue of $2.71 billion, up 35.1% compared to the year-ago quarter. The vendor raised its forecasted growth to 27.7%, aiming now for full-year 2026 revenue of $11.5 billion. Arista also raised its AI revenue target to $3.5 billion, effectively saying it would double its AI sales this year.

Less positive is the supply-chain pressure, as networking components – memory, chips, and wafers – face ongoing shortages and rising costs.

“Our demand is actually the best I’ve ever seen in my Arista tenure. The supply, however, is a slightly different and opposite tail. We are experiencing industry-wide shortages across the board, be it wafers, silicon chips, CPUs, optics and, of course, memory that I referred to last quarter, coupled with elevated costs to procure these,” Jayshree Ullal, CEO and chairperson of Arista, told financial analysts during the vendor’s earnings call. 

“Clearly, our demand is outstripping our supply this year. While we hope the supply chain will ease in the next year or two, the Arista operations team has been diligently engaging with our vendors in strengthening supply agreements and engaging in multiyear purchase commitments,” Ullal said. “I think the supply chain problem is not a one- or two-quarter phenomena. We now think it’s a one- or two-year phenomena… At first, we thought it was memory. Now it’s all the wafer fabrication facilities. Every chip is challenged… So we think a lot of this will continue into next year and keep us constrained for the next couple of years.”

It’s all about the optics

At the recent Optical Fiber Conference, Arista unveiled extended pluggable optics (XPO), a form factor designed specifically for optics at high speed, and assembled 45 optics module suppliers as part of a multi-source agreement to build and support XPO. Arista has seen major industry interest in it since the conference, and it’s now endorsed by more than 100 vendors, according to Ullal.

“Salient features include record-breaking throughput, delivering 12.8 terabits per pluggable module, unprecedented rack density achieving 204.8 terabits per OCP rack unit, integrated cold plate capable of cooling up to 400 watts power per module, and the universality and flexibility across a range of pluggable optics, copper, as well as linear halftime or retimed interfaces,” Ullal said. 

“While the industry has been talking a lot about co-packaged optics, these are still science experiments, and they’re very proprietary with individual vendors doing their own thing. We’ll embrace open CPO a few years from now, but we think XPO has a 10-year run, especially at 1.6T and 3.2T where you need liquid cooling and you need that kind of capacity. So, all the scale-up racks we’re talking about wouldn’t be possible without XPO or CPC or any one of those technologies,” Ullal said.

“Just as the last decade was greatly influenced by OSFP, the next decade will be greatly influenced by XPO,” Ullal said. “And remember, 99% of the optical market today that we connect to is all pluggable optics. So this is a very crucial invention and innovation, not just for Arista, but the industry at large.”

Enterprise AI: Calm before the storm?

When it comes to AI for enterprise network customers, Arista’s Ullal says it’s just getting started.

The company is seeing a shift from AI training toward more AI inference, “which means you don’t always need the GPU,” Ullal said. “You’re going to have high-end CPUs, and you’re going to have a smaller set of parameters and tokens to manage, and you’re going to have specific agentic AI use cases and applications. We’re seeing very, very early trials and stages. Nothing super big yet.” 

Some customers are deploying clusters that are “more inference-based, more agentic AI, edge, inference-based as well,” Ullas said. “I think we’ll see more of that. This is the calm before the storm, if you will. As AI gets more distributed, I think it doesn’t need GPUs alone. It’s going to need more high-performance compute… I think it’s gonna take a couple of years to fully happen.”

SUBSCRIBE TO OUR NEWSLETTER

From our editors straight to your inbox

Get started by entering your email address below.