AI Data Centers,
Built in Months
Factory-built IT, power and cooling modules — integrated, FAT-tested, then commissioned on your site. A Tier III Ready reference design engineered for NVIDIA GB300 NVL72.
Factory-built PMDC module — cooling plant, power and IT racks under one shell
Factory-built PMDC module
IT capacity
per-rack density
design PUE, 100% load*
power & cooling redundancy
Built in the factory, proven before it ships
Conventional data center construction takes 24–36 months of sequential site work. PMDC moves most of it into the factory — and runs it in parallel with your civil works.
Speed
Module fabrication and integration testing proceed while the site is prepared, compressing the overall schedule by up to half.
Flexibility
Deploy on idle land, in warehouses, or on constrained plots. Capacity scales in 2,588 kW units without disrupting operations.
Predictability
Standardized factory production with full FAT before shipment — the quality, availability and budget you approved is what arrives.
More than a container — the whole system, modular
PMDC (Prefabricated Modular Data Center) is a construction method where IT data halls, power and cooling systems are pre-fabricated as modules in the factory, integration-tested (FAT), then assembled and commissioned on site.
How is this different from a "container box"?
Beyond power and cooling, fire suppression, security and monitoring are engineered into the modules as one system. You receive a verified data center — not a kit of parts.
Three module types, one data center
The compute space
- Racks & hot-aisle containment (auto doors)
- Rack PDUs · dual A/B busway
- Front-flow fan wall cooling units
- Fire detection & suppression (optional)
- Integrated monitoring & access control (optional)
The powertrain
- MV/LV switchgear · RMU
- UPS & battery systems
- Automatic transfer switches (ATS)
- Transformers · distribution boards
- 3 MW × 4 powertrains, 4-to-make-3 (N+1)
The heat-rejection plant
- Free-cooling air-cooled chillers (N+1)
- Dry coolers (N+2)
- CDUs · variable-speed pump skids
- TES tank — uninterrupted cooling
- Piping · glycol (PG25) systems
The compute space
- Racks and hot-aisle containment with auto doors
- Rack PDUs on dual A/B busway
- Front-flow fan wall cooling units
- Fire detection and suppression
- Integrated monitoring and access control
The powertrain
- MV/LV switchgear and RMU
- UPS strings and battery systems
- Automatic transfer switches (ATS)
- Transformers and distribution boards
- 4-to-make-3 redundancy across powertrains
The heat-rejection plant
- Free-cooling air-cooled chillers (N+1)
- Dry coolers (N+2)
- CDUs with variable-speed pump skids
- TES tank for uninterrupted cooling
- Piping and PG25 glycol systems
Engineered for GB300-class density
142 kW per rack — over 15× the density of a typical data center. The PMDC reference design engineers power, cooling and airflow for NVIDIA GB300 NVL72 systems from the ground up.
- Supports up to 48 NVIDIA GB300 NVL72 systems
- 72-rack layout: 48 GPU · 12 switch · 12 management
- Scalable Units of 2,588 kW — grow with demand
- Full hot-aisle containment · rack-on-slab · overhead trunking
GB300 NVL72 systems (max)
TDP per GPU rack (123.5 liquid + 18.5 air)
IT load per scalable unit
power circuits per GPU rack (4 busways × 2)
87% Direct-to-Chip Liquid Cooling
A dual-loop chilled water system: free-cooling chillers serve the air-side load, while dry coolers feed the DLC loop through CDUs. TCS supply is held at 35 °C for GB300-class racks, and a thermal storage tank guarantees uninterrupted cooling through any power transfer.
- Heat capture — 87% liquid : 13% air
- TCS 35/45 °C (PG25) · FWS 30/42 °C · air side 20/30 °C
- Free cooling minimizes compressor hours — lower energy
- TES tank + dedicated 1,000 kW UPS — continuous cooling
Designed for Concurrent Maintenance
A distributed-redundant power architecture lets any single component be shut down for maintenance without touching IT operations. A fault in one powertrain never interrupts the data center.
- UPS 4-to-make-3 — three trains carry the full load (N+1)
- Dual A/B busways — 8 circuits per GPU rack (3P+E 63A)
- Critical cooling loads (CDU, fan walls, pumps) on dedicated UPS, 2N feeds
- Diesel generator backup with on-site fuel autonomy
Factory and site, moving at the same time
The key is step 3 — while modules are built and verified in the factory, civil works proceed in parallel on site. The waiting time of conventional builds is engineered out.
The 7.8 MW reference design
| Overall | |
|---|---|
| IT capacity | 7,764 kW = 3 × Scalable Units (2,588 kW) — adjustable per requirement |
| Design basis | Uptime Institute Tier III Ready · concurrent maintainability |
| Annual PUE | 1.15 – 1.25 (100% load, design simulation) |
| Total footprint | ≈ 2,030 m² (power 445 / cooling 869 / IT hall 714) |
| Ambient design | ASHRAE N=20-year extreme: 37.8 °C DB / 28.7 °C WB |
| IT & Racks | |
| Rack configuration | 72 racks = 48 GPU (142 kW) + 12 switch (56 kW) + 12 management (23 kW) |
| Supported platform | NVIDIA GB300 NVL72, up to 48 systems · room-based air + DLC in parallel |
| Rack liquid interface | PG25 coolant · 1.3 lpm/kW · FD83 couplings · 20 psid @ 45 °C |
| Airflow design | Full hot-aisle containment (auto doors) · rack-on-slab · dual A/B busway · overhead trunking |
| Cooling | |
| Heat capture ratio | 87% liquid : 13% air |
| Water temperatures | TCS 35/45 °C (PG25) · FWS 30/42 °C · air-side CHW 20/30 °C |
| Heat rejection | Free-cooling air-cooled chillers (N+1) + dry coolers (N+2) + TES tank |
| Room units | Front-flow modular fan walls (N+1) · in-row CDUs (N+1) |
| Power | |
| Redundancy | Distributed redundant · UPS 4-to-make-3 (N+1) · power modules 3 MW × 4 · cooling on 2N feeds |
| Rack feeds | GPU racks: 4 busways × 2 circuits = 8 feeds (3P+E 63A) · management racks on dual-source PDUs |
| Backup | Diesel generator support · 12-hour on-site fuel autonomy · dedicated 1,000 kW UPS for critical cooling |
| Grid interface | Customized per site (reference: 22.9 kV / 380 V / 60 Hz) |
All figures reflect the 7.8 MW reference design and are adjusted to site conditions and requirements. PUE values are design simulations.
Clear from day one — what we deliver, what you prepare
✓ Delivered by HeTone
- System & module enclosure design and fabrication
- Power systems & electrical distribution
- Cooling systems & chilled-water piping/distribution
- IT infrastructure — racks, rack PDUs, busway
- Factory integration testing (FAT) & site commissioning (SAT)
- Optional: fire, monitoring, access control, ventilation systems
Prepared by the client (typical)
- Civil & building works — foundations, slabs, trenches
- Freight to site, offloading and final positioning
- External electrical works — utility feeds, genset tie-ins
- IT data cabling and fiber-optic installation
- Emergency gensets & fuel tanks (procurement support available)
The exact scope is finalized in each project proposal. The split above reflects the typical baseline of the reference design.
Vendor-neutral integration, Asian supply-chain speed
One-stop AI infrastructure
From GPU servers and racks to power, cooling and construction — liquid cooling technology and global computing infrastructure under one team.
Proven global architecture
Built on proven module architecture from global infrastructure partners, applying Uptime Institute Tier III Ready design criteria.
Taiwan & SEA supply chain
Manufacturing bases across Taiwan and Southeast Asia shorten lead times and diversify supply-chain risk.
Turnkey, single accountability
Design → factory build → FAT → shipping → installation → SAT, all under one contract and one responsible team.
Frequently asked questions
A Prefabricated Modular Data Center: the IT hall, power and cooling systems are built as factory modules, integration-tested (FAT), then assembled and commissioned on your site. Unlike a simple containerized server box, the entire facility system — including fire suppression, security and monitoring — is engineered as one modular product.
Because factory fabrication runs in parallel with your civil works, overall schedules are typically up to half of a conventional build. The exact timeline depends on capacity and site conditions — we provide a stage-by-stage schedule at consultation.
CAPEX consists of the three module systems (IT / power / cooling), plus transport, installation and commissioning. It varies mainly with rack density, redundancy level and site conditions. Request a consultation for a budgetary estimate based on your requirements.
Tier III Ready means the module design follows Uptime Institute Tier III criteria — concurrent maintainability with N+1 redundancy. Site-level certification is a separate process performed on the completed facility; our design basis is intended to support that path, and we can assist with the certification process.
Often yes — via staged retrofit paths (e.g. rear-door heat exchangers as a transition, then DLC), or by adding PMDC modules alongside an existing facility to expand without major building works. A site assessment determines the best route.
The reference design targets NVIDIA GB300 NVL72 (142 kW/rack, 87% liquid capture). The cooling architecture’s temperature envelope (TCS 35/45 °C) is designed to accommodate future high-density platforms as well.
Planning an AI data center?
Tell us your project scale and timeline — our engineers will follow up with materials based on the 7.8 MW reference design.