• Home
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Subscribe to Daily Newsletter
  • NextGenInfra.io
No Result
View All Result
Converge Digest
Sunday, April 19, 2026
  • Home
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Subscribe to Daily Newsletter
  • NextGenInfra.io
No Result
View All Result
Converge Digest
No Result
View All Result

Home » Google Pushes 1 MW AI Rack Power Architecture and Liquid Cooling Blueprint

Google Pushes 1 MW AI Rack Power Architecture and Liquid Cooling Blueprint

May 1, 2025
in Data Centers
A A

At the OCP 2025 EMEA Summit, Google unveiled major infrastructure innovations to power the next wave of AI workloads, including a shift to +/-400 VDC power delivery capable of supporting up to 1 megawatt per IT rack and the open sourcing of its next-generation liquid cooling solution, Project Deschutes. These advances aim to support the explosive growth of AI compute demands, which are expected to surpass 500 kW per rack before the end of the decade. Google is collaborating with Meta and Microsoft under the Mt Diablo project to standardize this new high-voltage power architecture, leveraging the mature EV supply chain for scale and efficiency.

In parallel, Google announced plans to contribute its fifth-generation cooling distribution unit (CDU) to the Open Compute Project. Drawing on nearly a decade of deployment experience with TPU liquid cooling, the Deschutes CDU design enables extremely high availability—99.999% uptime across more than 2,000 TPU pods. The disaggregated cooling approach isolates rack and facility loops and uses cold plates, manifolds, and flexible hoses to manage thermal loads from chips now exceeding 1,000W. Together, these technologies represent the next frontier in AI infrastructure design, driving greater power density, thermal performance, and serviceability across hyperscale deployments.

  • Google introduces +/-400 VDC power architecture to support up to 1 MW per rack, replacing legacy 48 VDC systems
  • AC-to-DC sidecar power racks separate power components from the IT rack, boosting power efficiency by ~3%
  • High-voltage standardization is being developed with Meta and Microsoft via the Mt Diablo project at OCP; draft spec coming May 2025
  • Google to contribute Project Deschutes (5th-gen CDU) to OCP to accelerate liquid cooling adoption
  • Deschutes enables 99.999% uptime and supports compact, cold-plate cooled designs ideal for AI accelerators
  • Google first deployed liquid cooling with TPU v3 in 2018; latest deployments include TPU v5 and Ironwood
  • Liquid-cooled servers reduce size, double chip density, and outperform air cooling in thermal and hydraulic performance
  • Water conducts heat ~4,000x more efficiently and has ~30x better thermal conductivity than air
  • Google calls for industry-wide adoption of +/-400 VDC and Deschutes architecture to prepare for future AI scale

“With the accelerating pace of AI hardware development, we must collectively quicken our pace to prepare data centers for what’s next… The most impactful innovations are still ahead.” — Madhusudan Iyengar & Amber Huffman, Principal Engineers, Google

ShareTweetShare
Previous Post

Meta Expands AI Infrastructure Investment with $72B CapEx Outlook for 2025

Next Post

OCP and UALink Join Forces to Advance AI Interconnect Standards

Jim Carroll

Jim Carroll

Editor and Publisher, Converge! Network Digest, Optical Networks Daily - Covering the full stack of network convergence from Silicon Valley

Related Posts

Cisco, G42, and AMD to Build AI Infrastructure in the UAE
AI Infrastructure

DigitalBridge Teams with KT for AI Data Centers in Korea

November 26, 2025
BerryComm Expands Central Indiana Fiber with Nokia
5G / 6G / Wi-Fi

Telefónica Germany Awards Nokia a 5-Year RAN Modernization Deal

November 26, 2025
AMD’s Compute + Pensando Network Architecture Powers Zyphra’s AI 
AI Infrastructure

AMD’s Compute + Pensando Network Architecture Powers Zyphra’s AI 

November 25, 2025
Bleu, the “Cloud de Confiance” from Capgemini and Orange
Clouds and Carriers

Orange Business Begins Migration of 70% of IT Infrastructure to Bleu Cloud

November 25, 2025
Dell’s server and networking sales rise 16% yoy
Financials

Dell Raises FY26 AI Infrastructure Outlook as AI Server Shipments Surge 150%

November 25, 2025
GlobalFoundries acquires Tagore Technology’s GaN IP
Optical

GlobalFoundries Acquires InfiniLink for Silicon-Photonics Expertise

November 25, 2025
Next Post
OCP and UALink Join Forces to Advance AI Interconnect Standards

OCP and UALink Join Forces to Advance AI Interconnect Standards

Categories

  • 5G / 6G / Wi-Fi
  • AI Infrastructure
  • All
  • Automotive Networking
  • Blueprints
  • Clouds and Carriers
  • Data Centers
  • Enterprise
  • Explainer
  • Feature
  • Financials
  • Last Mile / Middle Mile
  • Legal / Regulatory
  • Optical
  • Quantum
  • Research
  • Security
  • Semiconductors
  • Space
  • Start-ups
  • Subsea
  • Sustainability
  • Video
  • Webinars

Archives

Tags

5G All AT&T Australia AWS Blueprint columns BroadbandWireless Broadcom China Ciena Cisco Data Centers Dell'Oro Ericsson FCC Financial Financials Huawei Infinera Intel Japan Juniper Last Mile Last Mille LTE Mergers and Acquisitions Mobile NFV Nokia Optical Packet Systems PacketVoice People Regulatory Satellite SDN Service Providers Silicon Silicon Valley StandardsWatch Storage TTP UK Verizon Wi-Fi
Converge Digest

A private dossier for networking and telecoms

Follow Us

  • Home
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Subscribe to Daily Newsletter
  • NextGenInfra.io

© 2025 Converge Digest - A private dossier for networking and telecoms.

No Result
View All Result
  • Home
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Subscribe to Daily Newsletter
  • NextGenInfra.io

© 2025 Converge Digest - A private dossier for networking and telecoms.

This website uses cookies. By continuing to use this website you are giving consent to cookies being used. Visit our Privacy and Cookie Policy.
Go to mobile version