Kingdom of Saudi Arabia · Licensed Operator

The Kingdom’s
AI Compute Backbone.

Tier III-certified immersion and direct-to-chip cooling. 187 megawatts. Engineered from silicon up for the GPUs that will train Saudi Arabia’s future.

Reserve CapacityTake the 360° Tour
Deployed + committed
187MW
High-density racks
7,000+
Per rack
65–100kW
Guaranteed uptime
99.982%

Certified. Audited. Aligned.

Uptime Institute Tier III · DesignTier III · Constructed FacilityTier III · OperationsLEED AlignedISO 27001SAMA CompliantNCA Cybersecurity FrameworkVision 2030 Aligned

The thermal ceiling

Air-cooled data centers can’t host the AI era.

NVIDIA H100s draw 700W. B200s draw 1,000W+. A single AI training rack now exceeds 100 kW — eight times what a traditional data center was built for. The hyperscale playbook is being rewritten in real time, and the Kingdom needs infrastructure built for what comes next, not what came before.

  • Standard hot-aisle containment caps thermally near 20 kW per rack.
  • Dense GPU clusters require dielectric or cold-plate fluid management.
  • Retrofitting an air hall to liquid is six to eighteen months of disruption.
  • Building wrong costs more than building twice.
Rack power densitykW per rack
Traditional DC8 kW
Modern DC20 kW
Desert Dragon100 kW

Our immersion platform supports up to 100 kW/rack today, with thermal headroom for the next two GPU generations.

Why Desert Dragon

Three reasons hyperscalers and sovereigns choose us.

Every operator says they're different. Here's what is actually different about us — and why it matters when you're committing serious capacity.

Sovereign by design

$2.5B committed. Saudi-owned. Vision 2030 aligned. Engineered for KSA’s regulatory, security, and energy environment from day one.

How we’re built

Built for the GPU era

Tier III-certified immersion and direct-to-chip cooling supporting 100 kW racks. The first AI-native data center platform in the Kingdom.

See the platform

Operationally certified

Uptime Institute Tier III across Design, Construction, and Operations. Run by a bench with 150+ combined years in global hyperscale.

Meet the bench

The platform

Three cooling architectures. One platform.

Choose the thermal envelope that matches your workload — or blend all three in a single deployment.

Immersion

Single-phase dielectric immersion.

Compute hardware fully submerged in non-conductive thermal fluid. Eliminates fans, hot spots, and the thermal throttling that caps GPU performance. Engineered for the racks the air-cooled world can't touch.

Rack density
Up to 100 kW
Annualized PUE
1.05–1.08
Water consumption
Zero
Explore the immersion architecture
Immersion cooling architecture illustration

Workloads at home in our halls

Built for the workloads that matter.

If your workload demands density, determinism, or sovereignty — Desert Dragon is the only purpose-built platform for it in the Kingdom.

LLM Training Clusters

Up to 100 kW immersion racks for H100/H200/B200 multi-thousand-GPU pods.

GPU Inference at Scale

Direct-to-chip cooled racks with sub-millisecond intra-rack networking.

Sovereign Cloud Regions

Air-gapped, Saudi-cleared regions for sovereign hyperscaler deployments.

High-Frequency Trading

Sub-microsecond cross-connects to regional exchanges with 2N power.

Genomics & Drug Discovery

Dense compute for protein folding, simulation, and parallel HPC workloads.

Government Mission Systems

NCA-aligned secure enclaves with classified-tier access controls.

Capacity calculator

Size your deployment.

A reference estimate for your team's planning. We'll follow up with a precision quote tailored to site, workload, and timing.

Your deployment

Adjust the inputs. Numbers update live.

50
1500
Cooling preference
Redundancy
Total power
3.44 MW
3,445 kW facility load
Annualized PUE
1.06
Among the lowest in operation
Annual energy
25,651 MWh
At 85% utilization
Estimated annual cost
$1,026,059
At KSA reference rate $0.04/kWh

Our recommendation

Based on your inputs, we recommend a deployment of immersion-cooled pods with 2N power architecture. At this rack density, immersion cooling is required to maintain GPU thermal envelopes and protect hardware investment.

Carbon comparison · annual

At a Desert Dragon (KSA grid mix)16,673 t CO₂e
If deployed at a US average facility10,261 t CO₂e
If deployed at an EU average facility6,413 t CO₂e

Reference grid factors. KSA factor improves as renewable integration scales through 2029.

Get a precision quoteOpen detailed calculator

Reference estimate for planning. We respond to qualified RFPs with site-specific pricing within one business day.

Proof of operation

Trusted by the operators building Saudi Arabia's digital future.

Live customers spanning telecommunications, sovereign clients, and the largest enterprise IT estates in the Kingdom.

MEEZA
HOSHAN
GO
TUWAIQ
LIFE·SHIELD
XDS
How a regional telco deployed 12 MW of AI-ready capacity in 90 days.
Telecommunications

How a regional telco deployed 12 MW of AI-ready capacity in 90 days.

12 MW
Megawatts delivered
90 days
From contract to live
65 kW
Rack density supported
Read case study
A government client's path to domestic GPU compute capability.
Sovereign client

A government client's path to domestic GPU compute capability.

256 H100
Compute capability
100% KSA
Data residency
< 6 weeks
Time to first job
Read case study
Migrating 5,000 servers to immersion cooling — without a minute of downtime.
Enterprise migration

Migrating 5,000 servers to immersion cooling — without a minute of downtime.

5,000+
Servers migrated
0 min
Customer downtime
−42%
Power footprint reduced
Read case study

The operating bench

The team that’s already done it.

A leadership group with combined experience across Hydro66, Northern Data, News Corp, and 30+ years in global hyperscale infrastructure — now building it for the Kingdom.

Meet the full bench

Sustainability

Pragmatic sustainability. Measurable progress.

Closed-loop cooling. Free-air optimization where viable. Hot-aisle containment. AI-driven workload balancing reducing power load by 30%. Carbon footprint monitoring rolling out across all sites by 2029, with continuous improvement aligned to Vision 2030 sustainability targets.

1.15
Target PUE across portfolio
Near-zero
Water consumption (immersion halls)
−30%
Power reduction from AI workload balancing
LEED
Aligned design across all five sites

Insights

From the operating floor.

Field reports, market analysis, and engineering briefings from the team building the Kingdom’s AI infrastructure.

Browse all insights
engineering 12 min

Why immersion cooling is no longer optional for AI training clusters

NVIDIA H100 GPUs draw 700W. B200 modules draw over 1,000W. A single AI training rack now exceeds 100kW — eight times what a traditional data center was built to dissipate. We walk through the thermodynamics, the operating economics, and the design choices that made dielectric immersion the only viable path for serious AI compute.

Khanh Ho Si· Chief Technology Officer
engineering 14 min

From design to operations: how we earned three Uptime certifications in 18 months

A field report from the engineering team. The pre-design discipline that pays off at certification. The four mistakes most operators make on Tier III Design submission. The audit prep cadence that turned what is normally a six-month back-and-forth into a clean first-pass approval.

Ahmad Mansour· Executive Director, Engineering
company 9 min

Inside our 187 MW buildout: a site visit

A walk-through of the active facility — from the substation step-down to the immersion gallery floor. Photographs, build sequencing, and the why behind some of the less obvious design choices visitors notice on a tour but don't always get explained.

Editorial team

Capacity is being committed daily.
Reserve yours.

Tour the facility. Talk to engineering. Get a precision quote. We respond to qualified inquiries within one business day.

Reserve CapacitySchedule a Site Tour

Olaya District, Riyadh · Open for tours by appointment