04 / Services
Data Center Liquid Cooling Services_
Data center liquid cooling services for GPU racks: CDU installation, rack-level manifold routing, quick-disconnect fittings, leak detection, and thermal commissioning. GB200 NVL72 and GB300 NVL72 racks require liquid cooling — there is no air-cooled option. We integrate it, across Texas and the United States.
Required for all NVIDIA GB200 and GB300 NVL72 deployments
Why Liquid Cooling Is Required_
The GB200 NVL72 rack draws approximately 120kW. The GB300 NVL72 draws more. At these power densities, air cooling physically cannot remove heat fast enough. The thermal design power of 72 Blackwell GPUs in a single rack exceeds what fans and heat sinks can dissipate.
Direct liquid cooling circulates coolant through cold plates mounted on GPU and CPU packages, removing heat at the source. It is not an upgrade or an optimization — it is the only viable thermal solution for current-generation high-density GPU platforms.
Every GB200 and GB300 NVL72 deployment includes liquid cooling as a core infrastructure requirement, not an add-on. Our scope covers the full cooling integration from CDU to cold plate.
Scope of Work_
CDU Installation
Coolant Distribution Unit placement, connection to the facility chilled water loop, and commissioning. We install, connect, and validate CDU operation including heat exchanger performance, pump function, filtration, and control system configuration.
Rack-Level Manifold Routing
Supply and return manifolds routed from CDU to each GPU rack. Manifolds distribute coolant evenly across all server trays. Proper manifold design ensures balanced flow — every GPU receives adequate cooling.
Quick-Disconnect Fittings
Dry-break QD fittings installed at each server tray allow individual servers to be removed for maintenance without draining the entire cooling loop. These push-pull couplings seat and latch by hand — there is no torque screw. We install them, confirm full engagement, and verify seal integrity.
Leak Detection
Leak detection sensors installed under each rack, at CDU connections, at manifold junctions, and at quick-disconnect points. Sensors wired to monitoring systems that generate alerts before coolant reaches electrical components.
Pressure Testing & Flushing
The entire cooling loop is pressure tested before coolant is introduced. This identifies leaks, weak fittings, or manufacturing defects. After pressure testing, the loop is flushed to remove particulates.
Coolant Fill & Flow Balancing
Coolant introduced to the loop with controlled fill procedures to avoid air pockets. Flow rates balanced across all racks and server trays to ensure uniform cooling.
Thermal Commissioning
GPU racks brought under load and thermal performance validated. Coolant inlet and outlet temperatures, flow rates, and GPU junction temperatures monitored to confirm the cooling system meets platform requirements.
CDU Architecture_
A CDU (Coolant Distribution Unit) contains a heat exchanger, circulation pumps, filtration, a reservoir, and a control system. Facility chilled water (typically 10–20°C) flows through the primary side of the heat exchanger. Rack coolant flows through the secondary side.
The CDU maintains coolant at the target temperature and flow rate regardless of GPU load fluctuations.
One CDU typically serves one to two GPU racks depending on the platform and thermal load. GB200 NVL72 deployments at 120kW per rack often require dedicated CDUs per rack.
The CDU must be sized to handle peak thermal load with margin — undersized cooling is a deployment failure that cannot be fixed without hardware changes.
How We Work_
Facility Water Review
Confirm chilled-water temperature, flow, pressure, and water quality at the CDU tie-in with the facility mechanical team.
CDU & Manifold Install
Place CDUs, route supply and return manifolds to each rack, and land quick-disconnect fittings at every server tray.
Pressure Test & Flush
Pressure-test the dry loop, correct any leaks, then flush to remove particulates before coolant is introduced.
Fill & Flow Balance
Controlled coolant fill to avoid air pockets, then balance flow across racks and trays so every GPU is cooled evenly.
Thermal Commissioning
Bring racks under load and verify inlet/outlet temperatures, flow, and GPU junction temperatures against the platform spec.
Coordination With Facility Mechanical_
Liquid-cooled GPU deployments require coordination between the IT infrastructure team and the facility mechanical team. The building must provide chilled water at the correct temperature, flow rate, and pressure to feed the CDUs. Piping, valves, and isolation points must be in place before GPU rack installation begins.
We coordinate with facility mechanical teams and general contractors to ensure the building-side water infrastructure is ready for GPU cooling integration. This includes reviewing mechanical drawings, confirming flow capacity, and validating water quality before connecting CDUs.
Standards We Work To_
Thermal design and facility water conditions follow the OEM platform spec together with ASHRAE TC 9.9 guidance for data center liquid cooling classes. Loop assembly, leak detection, and commissioning follow the CDU and rack OEM procedures.
Pressures, coolant temperatures, and torque values for the fittings that take them are made to the relevant OEM spec — we defer to those documents rather than publish fixed numbers that vary by CDU and platform. The dry-break quick-disconnects at the trays are push-pull couplings with no torque screw.
Platform coverage: GB200 NVL72 and GB300 NVL72 — the current liquid-cooled NVL72 platforms — plus rear-door and direct-to-chip retrofits on earlier air-cooled halls.
Where we work: based in Texas, providing data center liquid cooling services across the United States.
Track Record_
Active GB200 NVL72 and GB300 NVL72 liquid-cooled deployments. CDU installation, manifold routing, leak detection, and thermal commissioning across hyperscale AI training facilities.
Our technicians have hands-on liquid cooling deployment experience at facilities operated by Meta, Oracle, and Computacenter.
Liquid Cooling FAQ_
Why do GB200 and GB300 NVL72 racks require liquid cooling?
An NVL72 rack concentrates 72 Blackwell GPUs and 36 Grace CPUs in a single enclosure at power densities where air cooling cannot remove heat fast enough. NVIDIA specifies direct liquid cooling for these platforms — there is no air-cooled option. Coolant circulates through cold plates on the GPU and CPU packages, removing heat at the source.
What does data center liquid cooling integration include?
Our scope covers CDU installation and connection to the facility chilled-water loop, rack-level supply and return manifold routing, quick-disconnect fittings at each server tray, leak detection, pressure testing and flushing, controlled coolant fill and flow balancing, and thermal commissioning under load.
Do quick-disconnect fittings need to be torqued?
No. The dry-break quick-disconnect couplings used at server trays are push-pull connectors — they seat and latch by hand and have no torque screw. We verify correct engagement and seal integrity rather than applying a torque value. Torqued fittings appear elsewhere in the loop, such as certain manifold and CDU connections, and those are made to the OEM spec.
What standards do you work to for liquid cooling?
Thermal targets follow the OEM platform spec together with ASHRAE TC 9.9 guidance for data center liquid cooling classes and facility water conditions. Loop assembly, leak detection, and commissioning follow the CDU and rack OEM procedures; we defer to those documents for pressures, temperatures, and torque values rather than publishing fixed figures.
Where do you provide liquid cooling services?
Leviathan Systems is based in Texas and provides data center liquid cooling integration across the United States for GB200 NVL72 and GB300 NVL72 deployments.
Ready to Deploy Your GPU Infrastructure?_
Tell us about your project. Book a call and we’ll discuss scope, timeline, and the best approach for your deployment.
Book a Call