Distributed edge network

Lower latency, higher resilience, more deployment flexibility.

The network concept emphasizes proximity to users, redundancy across many locations and the ability to stage deployment over time rather than waiting for mega-campus approvals. More importantly, it is the project’s experimental route to a larger claim: that dedicated dark fibre plus NVIDIA Spectrum-XGS scale-across networking can make many small AI factories behave like one system and render the mega data centre unnecessary for a substantial class of workloads.

Why distribution matters

Closer proximity can improve responsiveness.
Many smaller sites can improve resilience.
Distributed deployment helps reduce dependence on a few hyperscale nodes.
Existing commercial locations may provide faster pathways than greenfield campuses.

Why conventional data centres face resistance

The public sees the water, power, land and waste-heat problem.

Large centralized data centres can concentrate electricity demand, cooling-water demand, land use and rejected heat in one place. Cloud Wash is designed around the opposite proposition: distribute compute into compatible existing businesses, recover useful heat first, and use water only where it materially improves rejection.

Water

The current Los Angeles screening model uses 74 m³/year of direct evaporative water in its central case. A dry-only captured-loop sensitivity uses zero direct cooling water, while a 0.6× rated-IT hybrid cooler rejects all captured heat with at most 217 m³/year of recycled host water in the screened cases.

Useful heat instead of waste heat

Under the optimistic fully temperature-compatible bound, 50.38 GWh/year becomes useful process heat: 79.4% of IT electricity and 93.4% of modeled host thermal demand. Lower temperature compatibility materially reduces that benefit.

Existing sites, not giant greenfield campuses

Car washes and laundromats already have recurring thermal demand. The current model treats that demand as the siting and capacity constraint rather than assuming arbitrary compute density can be accommodated.

Resilience target

Near-zero downtime is a network architecture goal, not a single-site promise.

99.999%*Multi-site availability design target through geographic failover.

~5.3 min/yr*Implied annual downtime at 99.999% availability.

Many sitesIdentical virtual machines can be placed closer to users and shifted around outages or maintenance.

Validation gateThe audited thermal paper excludes the earlier availability module until a reliability-block model and field data exist.

*Target derived from the business/network design architecture. It is not presented as a validated scientific result.

Open science downloads

Current papers, methods and model limits.

First page of the current heat-sink-constrained distributed AI infrastructure manuscript

Heat-sink-constrained siting of distributed AI infrastructure: energy, carbon and water impacts in Los Angeles

The current deterministic hourly study places 6,160 B300-equivalent accelerators across 120 representative car-wash and laundromat hosts and tests useful heat, carbon, water, climate and metropolitan-network constraints. Its network programme uses NVIDIA Spectrum-XGS as the scale-across layer over dedicated optical fibre and specifies fibre-spool and multi-site experiments to test whether distributed sites can match co-located clusters — the key experiment behind the project’s objective of proving that mega data centres can become obsolete as the default architecture for many workloads.

Current paper PDF Model + files

First page of the current mechanism-resolved fanless photovoltaic-thermal heat-rejection manuscript

Mechanism-resolved modeling and adaptive control of a fanless photovoltaic-thermal heat-rejection array

The current 170 m² roof-array study resolves passive and wetted mechanisms separately, quantifies the water-capacity trade-off, rejects array-wide ionic wind on energetic grounds and tests roof-area matching from micro facilities to hyperscale.

Current paper PDF DOCX Model + files

The goal: prove the mega data centre is no longer necessary.

Cloud Wash Laundry is being designed around a direct technical challenge to the conventional mega-campus model: place compute where useful heat already exists, then connect those heat-sink-sited AI factories with dedicated dark fibre and NVIDIA Spectrum-XGS scale-across networking so the distributed sites can operate as one metropolitan compute fabric.

The dark fibre is the dedicated optical transport; Spectrum-XGS is NVIDIA’s scale-across networking layer. The project’s proposed fibre-spool testbed and multi-site field pilot are intended to determine whether distributed sites can match co-located clusters on workload-level measures such as time-to-result and energy per token.

Measured next. Modeled now. Heat-sink constrained.

Current design-stage results combine the 120-host Los Angeles heat-sink-siting model with the updated fanless heat-rejection screening figure. They are screening outputs, not field-certified performance. In parallel, the network programme tests whether NVIDIA Spectrum-XGS scale-across networking over dedicated dark fibre can make geographically separated AI factories behave as one compute system. The mega-data-centre conclusion remains an experimental objective, not an already validated result.