358
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 19 May 2026
358 points (96.4% liked)
Technology
85080 readers
4409 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
Wouldn't maintenance be a lot easier if they just placed it near the sea and pumped the water through from there? Or used a heat exchanger. All water going in is sent back out at a higher temperature.
By placing the heat sinks directly into the water there's no electricity needed for a pump, and tides, weather, and heat convection will move the water around.
Having the entire facility underwater also means less exposure to the elements.
Saltwater is one of the harshest terrestrial environments for a data center other than maybe lava. Pressure, oxygen, and sodium ions make the ocean extremely corrosive to metal structures.
You could be right about the first part, but I take issue with the second. Ships in seawater usually need sacrificial anodes so corrosion eats those instead of the hull and fasteners. I'm not sure how that would affect heat exchange, maintenance, or long-term reliability. It would definitely limit the materials you could use.
You know that those sacrificial anodes are simply zinc, right? One of the most plentiful metals on the planet?
Why would anyone on earth not make those simply swappable, like on boat motors, and on ocean going vessels.
We have been building ocean going vessels out of metal for over a century now. I think those so-called engineering challenges are solved.
Ever notice that the vast majority of oxidation actually occurs ABOVE the waterline?
Care to guess why?
Here's a hint, look at the first 3 letters of the word "oxidation."
Edit to add: Plus there's different metal choices for the actual heat exchangers, such as stainless 2507 or even titanium, which is extremely resistant to such corrosion, and for the parts of the building that do not need heat exchange, an insulating coating will mitigate nearly all oxidation issues.
First off, there's no need to be that combative. Second,
There's plenty of oxygen dissolved in water.
Combative?
You're reading way too much into words without emotion.
They're words.
And while there's oxygen dissolved in water, it's still less likely to cause oxidation than the oxygen in air.
There's study after study on this, and you would do well to look those up.
You clearly know more about this than I do. I understand the anodes are zinc, but you have to use materials that are going to feed on the zinc rather than visa versa. I honestly don't know if that's a meaningful limitation or not. I just know about boats and how to delay the water from destroying the hull.
What maintenance?
If the server is in the same room with you, you don't need to over build. If it's at the bottom of the ocean, you build it not just with redundant internals, you package reduncancies in full moduales.
If module 1 has a problem, switch to back up, and swap module 1 for a new full module.
They're not opening shit up and exposing it to sea water, everything will be in water proof containers, and you just hook up a couple connectors.
The only "maintenance" underwater will be unplugging something to raise it and connecting the replacement you sank.
As far as heat exchange, it happens all around it. Not just natural circulation, hot water rises and cold water sinks.
Even in a lake with no currents, if it's deep enough that produces flow on its own.
This solution probably doesn't need pumps. Pumping seawater is tricky, parts corrode, salt deposits build up, algae grows and clogs, regular maintenance is needed.
I'll bet what they're doing is a sealed system with large radiators, the seawater cools the radiators, the radiators cool the air in the internal closed-loop system which never actually touches seawater.
At least, if I were designing the system, that's what I would try to do.