Computing Power Used to Heat Public Pools

A Devon swimming pool is utilizing waste heat from a small, on-site data center to maintain its water temperature. The startup Deep Green provides the computing unit for free, helping the leisure center avoid massive energy bill increases. This innovative 'digital boiler' model is now being expanded to other pools across England to prevent further facility closures.

Key Points

• Deep Green's 'digital boiler' repurposes waste heat from AI computing to warm swimming pool water.
• The leisure center receives the heating system for free and is reimbursed for the electricity used to run it.
• This technology helps public facilities survive 'astronomical' energy price hikes that have threatened many pools with closure.
• The decentralized model solves the expensive cooling problem for data centers by moving them to locations where heat is needed.
• Seven additional pools in England have already signed up to implement this heat-sharing technology.

Sentiment

The overall sentiment is cautiously supportive. Hacker News largely agrees that the article describes a plausible and useful application of waste heat recovery, especially for facilities with constant heating needs, but the community is pragmatic rather than celebratory. The strongest pushback is not against the pool-heating concept itself, but against thin reporting, limited scalability, infrastructure costs, and any implication that heat reuse cancels out broader data-center impacts.

In Agreement

• A public pool is a good target for compute waste heat because it needs steady warmth and often faces tight operating budgets.
• Modern GPU systems can produce enough heat in a compact footprint to help maintain an indoor pool when paired with a proper heat exchanger.
• Using the heat directly is more sensible than trying to convert low-temperature waste heat back into electricity.
• District heating and similar community heat-reuse systems are mature enough that data centers should consider them as part of local infrastructure planning.
• Visible public benefits such as cheaper heat, warmer facilities, or subsidized energy could reduce local opposition to data centers.

Opposed

• The article leaves out important technical and economic details, so readers had to rely on other sources to understand the scale and savings.
• The approach only works where heat demand is nearby and the plumbing, maintenance, and operational complexity are worth the cost.
• If heating is the main goal, conventional heat pumps may be cheaper and simpler than buying or colocating compute hardware.
• Data centers may still create environmental and community harms through grid demand, fossil-fueled generation, noise, and rushed construction.
• Heat reuse should not be used to portray large data-center growth as environmentally harmless.