[fowie]

It's a fail-in-place data center. If a hard disk fails, you don't do anything. They used to replace telephony equipment all the time too, now it's just a locked room in a building. This is pushing data centers to be more hands-off.

[49bc]

So they don’t do anything? Eventually you’re just maintaining stacks of errored and broken computers.

[jedberg]

By then you've replaced the datacenter and you can shift the workload and then salvage it.

In theory you'd only leave it down there for three years anyway before everything in it is worth zero, at least to the IRS.

[loeg]

Yeah. Fail in place over the service lifetime. When enough stuff is failed that it isn't worth keeping down there, you pull it up and refill it.

[49bc]

Based on the picture, "pulling it up" looks like a fairly intensive task, requiring boats, persons, etc.

That kind of thing eats directly into the ROI for a datacenter. I doubt it competes with a static building with a bunch of solar panels on top.

[loeg]

Might just fail-in-place, then. Although that looks pretty bad from an environmentalist perspective.