SendTech Times
News
INFRASTRUCTURE CONSTRAINT:

Delta GoCool-150 Shows The Cooling Cost Of A 150kW AI Rack

Newsroom brief

ServeTheHome reported that Delta displayed a GoCool-150 liquid-to-air cooling unit for Nvidia Vera Rubin NVL72 racks, with 150kW heat removal, 225 litres per minute of coolant flow and a 32-fan rear array.

Verified against source materialEdited by SendTech Times Cloud & Infrastructure DeskSource: ServeTheHome
Delta GoCool-150 Shows The Cooling Cost Of A 150kW AI Rack
Image source: ServeTheHome

ServeTheHome reported that Delta rates the GoCool-150 to remove up to 150kW of heat from one liquid-cooled AI system.

The cabinet-sized unit appeared beside ASRock Rack's Nvidia Vera Rubin NVL72 setup at Computex 2026.

The cooling distribution unit is not a central plant for a hall of servers.

Delta positions it for facilities that want direct-liquid-cooled racks but lack a building-scale liquid loop.

The cabinet takes heat from the rack's liquid circuit and rejects it into the room air.

One Rack Adds A Second Machine

The displayed specifications list circulation of 225 litres of coolant per minute and a liquid-side target of 45C.

Five hot-swappable pumps sit at the bottom of the cabinet, allowing a failed pump to be replaced without stopping the CDU or the hardware it supports.

ServeTheHome also recorded a rear array of 32 Delta PFD2048HT fans moving 17,658 cubic feet of air per minute; the displayed specifications rate each 200mm fan at up to 336 watts on 48VDC power and list the complete cabinet at about 1,200kg before coolant is counted.

Those numbers make cooling a separate infrastructure purchase rather than a feature hidden inside the server rack.

The CDU needs power, monitoring, maintenance access and enough surrounding air-handling capacity to absorb the heat it releases.

Retrofit Flexibility Comes With An Air-Side Limit

Many existing data halls were built for lower-density, air-cooled racks.

A rack-level liquid-to-air unit can introduce denser systems without an immediate building-wide liquid loop, but the room's air-handling system must still carry away the rejected heat.

Hot-swappable pumps and fans reduce maintenance interruption, while the management console and control card give the cooler its own operating layer.

Operators must therefore track the unit alongside servers, networking gear and power systems.

Delta has not disclosed production customer deployments or long-duration reliability results for the GoCool-150.

Share this article
inXf

Related articles

More
AI Rack Density Pushes Data Centres Toward Power and Cooling Limits
Cloud & Data Centers

AI Rack Density Pushes Data Centres Toward Power and Cooling Limits

Data Center Knowledge analysis shows AI rack power moving from typical 11 kW rooms to 100 kW-plus deployments, with cooling, power distribution and grid access setting the ceiling.

NVIDIA’s 45-Degree Cooling Push Turns AI Factories Into A Water Test
Cloud & Data Centers

NVIDIA’s 45-Degree Cooling Push Turns AI Factories Into A Water Test

NVIDIA says its Rubin-generation AI infrastructure can run closed-loop liquid cooling at up to 45 degrees Celsius, cutting the need for chillers and water as denser AI factories put facility design under pressure.

NVIDIA Adds DSX Ready Checklist For AI Factory Power And Cooling
Cloud & Data Centers

NVIDIA Adds DSX Ready Checklist For AI Factory Power And Cooling

NVIDIA is starting DSX Ready with qualified battery and cooling hardware, giving AI factory customers a vetted supplier list for rack-scale deployments.

Waterless Cooling Becomes A Siting Test For AI Data Centres
Cloud & Data Centers

Waterless Cooling Becomes A Siting Test For AI Data Centres

Capacity wrote that zero-water and chip-level cooling systems are moving into AI data centre planning as operators try to reduce local water withdrawals while managing new energy trade-offs.

Frore LiquidJet Claim Targets Rubin GPU Cooling Efficiency
Cloud & Data Centers

Frore LiquidJet Claim Targets Rubin GPU Cooling Efficiency

Tom’s Hardware reported that Frore Systems says its LiquidJet coldplate could lower Nvidia Rubin GPU junction temperatures by 6°C to 12°C in an analytical thermal model. Frore Systems' white paper claims a 10°C reduction would raise token generation efficiency by about 15%, while Tom’s Hardware noted that delidded packages add mechanical risk.

Nvidia Japan AI Factory Plan Lists 140MW Capacity
Cloud & Data Centers

Nvidia Japan AI Factory Plan Lists 140MW Capacity

Tech Wire Asia reported that Nvidia’s Japan expansion includes a 140-megawatt AI factory using 13,750 Vera CPUs and 27,500 Rubin GPUs for the METI-backed FRONTia Project. The report did not name the site, capital cost, power supplier, construction timetable or first enterprise workloads.

Equinix HK6 Links Hong Kong AI Capacity To Shenzhen’s Innovation Corridor
Cloud & Data Centers

Equinix HK6 Links Hong Kong AI Capacity To Shenzhen’s Innovation Corridor

Equinix has opened the first phase of HK6 in Hong Kong, adding 1,000 cabinets, direct-to-chip liquid cooling and a private low-latency link to the Hong Kong-Shenzhen Innovation and Technology Park.

KDDI’s Osaka AI Data Center Turns Liquid Cooling Into A Power Test
Cloud & Data Centers

KDDI’s Osaka AI Data Center Turns Liquid Cooling Into A Power Test

KDDI is moving liquid cooling into an Osaka AI data center after a 2023 immersion test cut server cooling energy use by 94 percent and lowered PUE to 1.05.

Keep Reading

More Stories

Latest
Kepler Targets 2027 Production for HBM Replacement MemoryCloud & Data CentersOct 6, 2026Kepler Targets 2027 Production for HBM Replacement MemoryEE Times reports that Kepler Computing is preparing 3D ferroelectric memory for 2027 production, promising higher capacity and bandwidth per watt while limiting reliance on advanced-node lithography.Yokogawa Opens Singapore Hub For Industrial Cyber ResilienceCapital & PolicyOct 6, 2026Yokogawa Opens Singapore Hub For Industrial Cyber ResilienceYokogawa Engineering Asia has launched a Singapore center focused on OT cyber resilience, training, response planning and recovery coordination for Southeast Asia, Oceania and Taiwan.ClickFix Attack Uses Browser Cache To Hide Malware PayloadCybersecurityOct 6, 2026ClickFix Attack Uses Browser Cache To Hide Malware PayloadMicrosoft Threat Intelligence traced a ClickFix cache-smuggling method that preloads malware into browser caches, then uses file size checks and a pasted Run command to launch later credential-theft stages.VOA Tests Six-Month Startup Buildout Before Funding DecisionsFintech & Digital PaymentsOct 6, 2026VOA Tests Six-Month Startup Buildout Before Funding DecisionsTechCabal’s interview with VOA Venture Partners founder Victoria Olayide Adesanya describes a six-month build programme that lets the firm work inside African financial-infrastructure startups before deciding whether to invest.Bitcoin Holds $86,000 As Dollar Index Hits 18-Month HighCrypto/Web3Oct 6, 2026Bitcoin Holds $86,000 As Dollar Index Hits 18-Month HighCoinDesk reported that bitcoin stayed near $86,000 while the U.S. Dollar Index reached about 102.5, with U.S. rate expectations and European political risks strengthening the dollar backdrop.Google Freezes OSS Bug Bounty Reports After AI Submission FloodCybersecurityOct 6, 2026Google Freezes OSS Bug Bounty Reports After AI Submission FloodGoogle has stopped accepting new product vulnerability reports in its OSS VRP after invalid automated submissions swamped reviewers, while older reports and some Cloud VRP routes remain open.Fleuret AI Raises €4M For Continuous AI Pentesting PlatformCybersecurityOct 6, 2026Fleuret AI Raises €4M For Continuous AI Pentesting PlatformTech.eu reported that French startup Fleuret AI raised €4 million in pre-seed funding to develop an agentic-AI platform that turns penetration testing into a continuous security process.GFT Analysis Says AI Documentation Can Cut Maintenance Work 30%Fintech & Digital PaymentsOct 6, 2026GFT Analysis Says AI Documentation Can Cut Maintenance Work 30%A GFT Technologies analysis says AI-linked software documentation can cut maintenance effort and speed developer onboarding when knowledge assets stay synchronized with code changes.Schneider Electric Lines Up $22.6 Billion PTC DealAIOct 5, 2026Schneider Electric Lines Up $22.6 Billion PTC DealSchneider Electric plans to buy PTC in a cash transaction valuing the US engineering software provider’s equity at about $22.6 billion, adding product-lifecycle software to its industrial AI push.Aggarwal Pledges Ola Electric Stake To Fund ₹1,000 Cr Rights IssueCapital & PolicyOct 5, 2026Aggarwal Pledges Ola Electric Stake To Fund ₹1,000 Cr Rights IssueOla Electric founder Bhavish Aggarwal pledged 20 Cr shares to finance his participation in a rights issue that forms part of a larger ₹1,500 Cr fundraising plan.Natrona Schools AI Review Puts Student Privacy Ahead Of Classroom Tool UseAIOct 5, 2026Natrona Schools AI Review Puts Student Privacy Ahead Of Classroom Tool UseNatrona County trustees questioned whether teacher AI tools expose student data, even as existing district rules already ban unauthorized generative AI use by students.AMD Prices 256-Core EPYC 9996 At $14,904 For Server BuyersChips & SemiconductorsOct 5, 2026AMD Prices 256-Core EPYC 9996 At $14,904 For Server BuyersTechRadar reports that AMD’s 6th Gen EPYC 9006 “Venice” lineup includes a 256-core EPYC 9996 with 512 threads, 1GB of L3 cache, a 600W default power rating and a $14,904 list price for 1,000-unit orders.