A Japanese consortium led by shipping giant NYK Line is testing moored, quayside data centres as a way round scarce land and power — and it raises a genuine question for UK buyers stuck in planning queues. The idea is real; the capacity and dates are not. Talk to a Servnet engineer about what's actually deployable now.
View the data behind this chart
| Phase | Starts (week) | Duration (weeks) |
|---|---|---|
| Yokohama port demo testing | 0 | 20 |
| Design & site assessment | 15 | 60 |
| Samsung Heavy target launch | 100 | 20 |
What's actually been announced
The consortium pairs four very different specialisms: NYK Line, the shipping and logistics operator, is overseeing development; NTT Facilities is assessing designs, construction and operation; renewable energy developer Eurus Energy is exploring power supply; and MUFG Bank is looking at financing for related ventures.
Rather than starting with open-ocean facilities, the group's near-term focus is keeping the units moored safely alongside a quay, where they can tap existing port power and infrastructure rather than building everything from scratch. The consortium has previously signed a memorandum of understanding with the city of Yokohama for a renewable-powered demonstration at the port, with testing that began in March this year.
Why this matters beyond Japan
UK buyers will recognise the underlying problem even if the solution is unfamiliar: scarce development land and constrained grid connections are pushing data centre planners to look at sites they wouldn't have considered a few years ago. A quayside facility that borrows port power and telecoms, rather than competing for greenfield land and a fresh substation connection, is an attractive concept on paper for exactly the markets where the UK is squeezed hardest.
That said, nothing in the current announcement points to a UK deployment, a committed capacity figure, or an operational date. When asked directly what capacity the facilities might offer and when they could go live, the consortium had not responded at time of publication. Buyers should treat this as an early-stage concept test, not a procurable option.
The opportunity: ports as ready-made infrastructure
The logic of staying alongside the quay rather than heading out to open water is sound. Ports already have grid connections, fibre routes and physical security, which sidesteps three of the hardest problems in any new-build data centre project. It's a pragmatic first step before the harder goal of facilities well beyond the coast, which the consortium is also exploring alongside renewable sources like offshore wind.
Others are chasing the same prize from different angles — Oregon-based Panthalassa is exploring wave power for floating compute, and Samsung Heavy Industries is working towards its own facility, which it expects to have operational by the second calendar quarter of 2028. For UK buyers, that spread of parallel efforts suggests the concept has credibility, even if none of it is buildable here yet.
The risks that keep this offshore of a buying decision
The consortium itself acknowledges that true open-ocean deployments still face unresolved practical hurdles: maintenance access, high-capacity telecoms links, and independent power supplies all need solving before facilities can move away from the coast. Those are not trivial engineering problems, and they're exactly the kind of thing that turns a promising concept into a multi-year programme.
For anyone planning server configuration and capacity for the next two to three years, the sensible read is that floating data centres are a watch-item, not a line item. Buyers should keep tracking the Yokohama demonstration and the Samsung Heavy timeline for signal, while making real estate and power decisions based on what's buildable today.
Quayside, open-ocean, or onshore: comparing the options
It's worth laying the three broad site strategies side by side, because they carry very different risk profiles for a UK buyer weighing alternative sites against a conventional build.
What UK buyers should do in the meantime
While floating concepts mature, the practical levers for easing a UK space-and-power crunch remain the ones already available: extending the life of existing estate, right-sizing new deployments, and financing capacity in a way that doesn't commit capital to unproven sites. Running a server end-of-life checker against current estate can often free up rack space and power headroom faster than any new site strategy.
Where new capacity genuinely needs to be added, model the costs properly before committing. An IT finance calculator can help compare the economics of refurbishing or extending current infrastructure against a wait for alternative sites that may be years from operational. For buyers wanting a second opinion on whether to hold, extend or build, talk to a Servnet engineer before locking in a strategy around a technology that's still being pilot-tested at a single Japanese port.
- 01The Register — Japanese consortium bets on building bobbing bit barns · 15 October 2026