# cabinets

Published articles for cabinets.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Three Physical Walls Are Hitting AI Datacenter Infrastructure at the Same Time - And the Compound Effects Are Worse

DevFeed: [Three Physical Walls Are Hitting AI Datacenter Infrastructure at the Same Time - And the Compound Effects Are Worse](<https://devfeed.tech/articles/three-physical-walls-are-hitting-ai-datacenter-infrastructure-at-the-same-time-and-the-compound-effects-are-worse-96458.md>)

Original publisher: [Read original article](<https://www.nextplatform.com/compute/2026/10/23/partner-content-three-physical-walls-are-hitting-ai-datacenter-infrastructure-at-the-same-time-and-the-compound-effects-are-worse/5304697>)

Author: xFusion

Published: 2026-10-23T00:00:00Z

Content type: article

Language: en

Sources: [The Next Platform: In-depth coverage of high end computing](<https://devfeed.tech/sources/the-next-platform-in-depth-coverage-of-high-end-computing.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [data centers](<https://devfeed.tech/topics/data-centers.md>), [systems](<https://devfeed.tech/topics/systems.md>), [clean energy data centers](<https://devfeed.tech/topics/clean-energy-data-centers.md>)

Tags: [2023](<https://devfeed.tech/tags/2023.md>), [2025](<https://devfeed.tech/tags/2025.md>), [2026](<https://devfeed.tech/tags/2026.md>), [800v](<https://devfeed.tech/tags/800v.md>), [accelerators](<https://devfeed.tech/tags/accelerators.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-training](<https://devfeed.tech/tags/ai-training.md>), [air](<https://devfeed.tech/tags/air.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [board](<https://devfeed.tech/tags/board.md>), [bus](<https://devfeed.tech/tags/bus.md>), [business-value](<https://devfeed.tech/tags/business-value.md>), [cabinets](<https://devfeed.tech/tags/cabinets.md>), [compute](<https://devfeed.tech/tags/compute.md>), [datacenter](<https://devfeed.tech/tags/datacenter.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>)

### AI overview

The article argues that AI data centers must co-optimize cooling, power delivery and interconnect as rack and chip demands rise together; solving only one or two constraints can strand capacity. It describes xFusion's proposed responses: diamond-copper cold plates rated for over 4,000 W per chip and a 155 kW liquid-cooled rack designed to avoid secondary-piping modifications; a shift from 54V distribution toward 800V DC to reduce current and resistive losses, alongside wide-bandgap power devices and vertical power delivery; and 224G copper backplanes, with 448G research underway as signaling limits push networks toward optics. Two-phase jet cooling remains in development, with sealing, pressure management and serviceable fluid connections as challenges. The figures and roadmap are presented as xFusion's claims and plans, not independent validation.

### Source excerpt

Here is a trajectory that should focus the mind of anyone designing AI infrastructure right now. In ...