Industry Statistics
AI Infrastructure Statistics: Power, Cooling, Latency, and Networking
A living reference page of useful AI infrastructure ranges and bounds. Use it for orientation, then read primary sources before making design or investment decisions.
· 12 min read
By Seabase Editorial
← Articles index · Home · Infrastructure · Nori · About
Topics: Infrastructure · Cooling · AI Inference · Energy · Geography
People looking for subsea data centers, AI inference infrastructure, cooling, and coastal placement often need a short set of comparable numbers. This page collects commonly cited industry ranges, physics bounds, and deployment comparisons in one place.
It is designed to be cited and updated. It is not a substitute for vendor datasheets, utility studies, or site-specific engineering.
Treat every table as a planning range until you validate it against primary sources for your site and generation.
How to use this page
- Use ranges for orientation, not as contract specifications.
- Prefer annualized or generation-specific sources when making procurement decisions.
- Separate physics bounds (speed of light) from commercial averages (PUE, rack power).
- Follow links into deeper essays and glossary definitions for context.
Canonical short URL: /statistics. This article is the same living reference.
Rack power evolution (typical planning ranges)
AI has pushed rack power far beyond classic enterprise densities. Exact values depend on accelerator generation, networking, liquid cooling coverage, and facility limits. The ranges below are planning envelopes commonly discussed in industry materials, not a single standard.
| Era / class (illustrative) | Typical rack power (planning range) | Notes |
|---|---|---|
| Legacy enterprise / light density | 5–15 kW | Air-cooled, limited GPU density |
| Early AI / mixed GPU racks | 20–40 kW | Often still air-assisted |
| Dense AI racks | 40–80 kW | Liquid cooling often required |
| High-density AI racks | 80–120+ kW | Facility and CDU limits dominate |
Related: Direct Liquid Cooling, Coolant Distribution Unit, and Can Underwater Data Centers Reduce Cooling Costs?.
Power Usage Effectiveness by Facility Type
PUE is total facility energy divided by IT energy. Lower overhead is better, but PUE is not a complete sustainability or cost score. Climate, utilization, electrical design, and cooling architecture all matter. See PUE.
| Facility context | Typical PUE range | Capital and operating context |
|---|---|---|
| Older enterprise data centers | 1.5–2.0+ | Performance varies widely by facility age, utilization, climate, and cooling architecture. |
| Modern hyperscale campuses | 1.1–1.4 | Achieving these levels typically requires substantial investment in optimized buildings, cooling plants, electrical systems, controls, and site infrastructure. |
| Leading high-efficiency campuses | Approximately 1.1 or lower | These results are generally associated with large, purpose-built facilities, favorable climates, high utilization, and significant capital investment. |
| Seabase design target | Approximately 1.05 | Seabase is designing for comparable or better infrastructure efficiency by taking advantage of modern direct liquid cooled AI systems, seawater as a stable heat sink, and a modular architecture intended to reduce the amount of cooling and facility infrastructure required. |
PUE alone does not describe total infrastructure economics. Two facilities can report similar PUE while requiring very different levels of capital investment, cooling equipment, land, construction, and ongoing operations.
Seabase is modeling toward approximately 1.05 PUE while targeting lower infrastructure capital requirements and lower cooling-related operating costs than conventional high-efficiency campuses. These remain design objectives subject to engineering validation, site conditions, utilization, equipment configuration, and consistent metering boundaries.
Seabase is not pursuing low PUE by recreating a hyperscale cooling campus offshore. It is designing a simpler thermal architecture intended to reach approximately 1.05 PUE with less supporting infrastructure, lower cooling overhead, and a smaller physical footprint.
Related: Environmental Accountability.
Network latency and distance (physics-informed)
Light in fiber is slower than in vacuum. A practical planning rule of thumb is that one-way fiber delay is on the order of about 5 microseconds per kilometer of fiber path (path length, not straight-line map distance). Round-trip delay is roughly twice the one-way path delay, before equipment and queuing.
| Approximate fiber path | Rough RTT floor (order of magnitude) | What it means |
|---|---|---|
| Same metro / nearby campus (~10–50 km path) | low single-digit to ~0.5 ms path floor | Application and peering often dominate |
| Regional coastal corridor (~200–500 km path) | ~2–5 ms path floor | Still strong for many interactive inference cases |
| Cross-country continental path (~4,000 km) | ~40 ms path floor | Before peering and congestion |
| Intercontinental paths | tens to 100+ ms | Route and undersea cable path matter |
These are path floors, not measured user latency. Queuing, Wi-Fi, mobile cores, and multi-step agent calls add more. Deep dive: The Latency Tax and Why AI Inference Changes Infrastructure Geography.
Why multi-step AI makes geography matter
If one network round trip costs 60 ms and an agent pipeline needs four sequential model or tool calls, network delay alone can impose about 240 ms before counting inference time. That compounding effect is why metro and coastal placement appear repeatedly in inference planning.
Cooling methods by deployment type
| Deployment type | Primary heat rejection | Typical tradeoff |
|---|---|---|
| Air-cooled land hall | CRAH / CRAH + chillers / evaporative | Familiar ops; limited at high rack density |
| Direct liquid cooling (land) | Cold plates + CDU + facility loop | Enables denser racks; adds fluid systems |
| Immersion (land) | Dielectric fluid baths | High density; service model changes |
| Floating / barge marine | Seawater exchangers + surface access | Access easier; motion and weather exposure |
| Subsea sealed modules | Module-to-seawater heat rejection | Strong sink coupling; retrieval/service design critical |
Related: What Is a Subsea Data Center?, Floating vs. Subsea Data Centers, Underwater vs Land Data Centers.
Cable landing stations and coastal networking
Most intercontinental internet traffic travels on subsea fiber. Cable landing stations (CLS) are where those systems come ashore and hand off to terrestrial networks.
- Landing density is uneven: some coasts have many diverse landings, others have few.
- Proximity to a landing helps but does not guarantee low latency. Terrestrial routing and peering still decide outcomes.
- Route diversity (multiple landings and paths) matters as much as raw capacity for resilience.
- Coastal metros often combine population demand, carrier hotels, and landing access in one geography.
Related: The Durable Geography of AI Demand and Metro AI Infrastructure.
AI power context (facility planning)
AI clusters are persistent electrical loads. Nameplate renewable capacity is not the same as firm delivered capacity for a continuous compute load. Batteries help with ride-through and smoothing. They are not a primary energy source.
- Plan for firm power at the intended continuous load, not only peak marketing numbers.
- Include networking, storage, pumps, controls, and conversion losses in site power budgets.
- Evaluate fuel logistics and interconnect timelines at the same priority as IT bill of materials.
Related: An Electron Is an Electron and power-flexible AI infrastructure.
Subsea thermal orientation
Ocean-adjacent and subsea designs use seawater as a stable heat sink for direct liquid cooled AI hardware. Modern DLC systems operate at higher coolant temperatures, increasing the temperature differential relative to surrounding seawater. That larger ΔT supports compact heat exchange and favors modular architectures that can follow successive hardware generations. Local thermal effects still depend on site conditions, flow design, equipment configuration, and monitoring practice.
See environmental research and Environmental Accountability.
Methodology and update policy
- Physics-based latency bounds use fiber-path approximations and are labeled as floors.
- Rack power and PUE bands are industry planning ranges synthesized for orientation. They will be revised as public datasheets and annual reports change.
- Cooling and landing sections are comparative frameworks, not ranked vendor scorecards.
- When this page changes materially, the modified date above is updated.
If you cite this page, include the access date. For industry narrative context, also see State of the Subsea Data Center Industry.
Related reading cluster
- What Is a Subsea Data Center?
- State of the Subsea Data Center Industry
- The Latency Tax
- Glossary
- Articles library
Frequently asked questions
What is a typical AI rack power density?
Planning ranges vary widely by generation. Dense AI racks often fall in the tens of kilowatts, and high-density designs can exceed 80–100 kW per rack. Always validate against the specific accelerator and cooling design.
What PUE should I expect?
Modern hyperscale campuses often report roughly 1.1–1.4. Leading high-efficiency campuses sometimes report approximately 1.1 or lower. Seabase is designing toward approximately 1.05 PUE as a modeled target, subject to engineering validation and operating conditions.
Next step
If you want help interpreting these ranges for a coastal deployment, contact Seabase.
Related research
- What Is a Subsea Data Center? (Complete Guide)
A complete guide to subsea and underwater data centers: history, benefits, challenges, power, networking, cooling, maintenance, environmental concerns, floating vs subsea, and AI inference.
- The Latency Tax: Why AI Infrastructure Geography Still Matters
Why network geography imposes a measurable cost on AI inference, how LEO satellite networks and coastal compute compare as solutions, and why raw accelerator performance is not the same as delivered throughput.
- Can Underwater Data Centers Reduce Cooling Costs?
How underwater data centers use seawater for heat rejection, when cooling costs fall, what costs remain, and how to compare marine cooling with land-based plants and liquid cooling.
- An Electron Is an Electron: Why AI Compute Should Follow Demand, Not a Single Energy Source
AI data centers need firm power, reliable fuel logistics, terrestrial fiber, and proximity to demand. Seabase explains why ocean compute should remain power-source flexible.
- State of Subsea AI Infrastructure 2026
A living quarterly brief on subsea and underwater data centers: deployments, GPU and rack-power trends, cooling, AI inference demand, announcements, regulation, and cable landing growth.