Meta buys a nuclear power plant (more or less)(techcrunch.com) |
Meta buys a nuclear power plant (more or less)(techcrunch.com) |
The usecase of "cloud provider buys power plant to power data center" does not exist: All current power plants are completely unsuitable as single power source for a data center because uptime/reliability is way lower than what you would need; decoupling datacenters from the grid is just a losing move, which is why no credible operator is even trying.
Are we gonna see more vertical integration between power generation/datacenters operation in the future? Maybe. But I'm very confident that we're not gonna see datacenters "leave the electrical grid" to be powered directly by nuclear plants, not now and not within 30 years either.
> All current power plants are completely unsuitable as single power source for a data center
I don’t think this is a fair assessment.
For example, there is a data center being built in New Zealand that will be grid connected but the power will be supplied from a huge hydro dam that has to have generators shutdown if the load is not great enough. Its primarily purpose is to power an aluminium smelter and there is not enough transmission capacity to transfer all the electricity elsewhere.
Another counter example is home solar. If a house is grid connected it’s still producing green energy even if it still sips gas generation at night.
This is IMO just a CO2 accounting trick, basically, and does not really achieve anything, because all that happens is that the average electricity user in that area gets a bit "dirtier" (on paper) while Meta becomes "zero emission" (on paper), meanwhile nothing actually changed.
Don't get me wrong, this is not harmful, but building new emission free power or replacing fossil plants is much more useful than paying some cash to basically shift blame around.
> For example, there is a data center being built in New Zealand that will be grid connected but the power will be supplied from a huge hydro dam
Sure-- but you still really want that grid connection, both to sell power when you have too much and also to buy power when the turbines are being maintained or water is running low. My point here is just that it almost never really makes sense to couple plant and datacenter directly and skip the grid connection.
Taking this approach would also basically lock your datacenter power use to the exact output power of the reactor, preventing you from scaling either side of the setup freely.
I think looking at moves like this from a power perspective is wrong, and I strongly believe that this is just minimum effort hedging against increasing CO2 costs (both monetary and reputational), i.e. greenwashing.
Datacenters and especially ML training hardware is highly capital intensive and depreciates at basically constant rate regardless of utilization.
I see currently no scenario where you wwould be willing to idle this expensive infrastructure just to save pennies on the dollar on a grid connection; carbon credits would have to be nonsensically expensive for this to happen.
Creative bookkeeping is a thing when it comes to carbon neutrality. They are still using electrons produced with gas power. But others in a place where Meta doesn't require much power get to use nuclear power now.
It's interesting that they feel like they need to spend on carbon offsets. But I don't think carbon offsets are a long term solution. The core issue for companies like Meta is that they need to get their hands on large amounts of preferably clean power.
Another interesting thing here of course is that this nuclear plant is hopelessly dependent on subsidies to stay operational. That's quite common with aging nuclear plants. It's apparently a deeply unprofitable business to be in. There's no such thing as a profitable nuclear plant that doesn't get heavily subsidized. Plenty of people in the nuclear sector are getting rich but very few under their own power (pun intended).
This achieves almost nothing: Emissions stay the same, no additional clean power is generated and there is no reduction in power consumption either.
The only notable result is that Meta is now "blameless" on paper.
As long as you can get carbon credits cheaply from peers that don't really need/profit/care about them, the whole scheme is rather useless.
"Focus on impact."
Every Facebook/Meta corp slogan works.
Matching base load with base production in this way is actually sensible move. And for maintenance periods you can just plan for and buy the power from market.
> Meta and its Big Tech peers have fallen head over heels for nuclear power lately. Meta announced earlier this year that it was soliciting proposals for new nuclear power plants that would generate between 1 and 4 gigawatts of power. Today, the company said it has received over 50 qualified submissions for sites in more than 20 states.
I would like to see one of those submissions. Is Meta getting into the utilities business? Have they figured out how to store the waste yet? How to insure the power plant? How to make a dime of nuclear power without relying on massive public subsidies?
This nuclear deal isn’t just PR or “greenwashing” — it’s a serious bet on power availability, which is quickly becoming the real bottleneck in AI.
Compared to some of their peers, Meta seems more willing to take technically grounded, long-term positions rather than chasing hype.
Hows that metaverse thing that they pivoted into so hard they rebranded going?
Not the power plants problem, as per law the government is responsible for it and the utilities have to pay them. The govt is in breach of this by not opening Yucca Mountain.
For contrast, in Finland the utilies are responsible for their waste and they have Onkalo
https://en.wikipedia.org/wiki/Onkalo_spent_nuclear_fuel_repo...
> How to insure the power plant
Thats what insurance pools like the "American Nuclear Insurers" are for
> How to make a dime of nuclear power without relying on massive public subsidies?
German commercial nuclear power plants for example never received ANY subsidies and they still made a lot of money.
> Not the power plants problem, as per law the government is responsible for it and the utilities have to pay them. The govt is in breach of this by not opening Yucca Mountain.
Strange, we assigned blame, yet the problem persists. What are they going to do with nuclear waste until the government acts? [1] I guess you wouldn't be willing to put a barrel or two into your garage, would you?
> Thats what insurance pools like the "American Nuclear Insurers" are for
I have heard of those. Let's check how much they disbursed for Fukushima. Apparently Fukushima was insured by the German Nuclear Reactor Insurance Association. And apparently the Association did pay out... nothing [2]. Furthermore, liability is often capped for those insurance associations, just like in the USA [3]. In the end the public will pay anyway.
> German commercial nuclear power plants for example never received ANY subsidies and they still made a lot of money.
The science service of the German Bundestag seems to disagree [4]. A more comprehensive study is here [5].
[1] And "act" could just mean to revoke the law and institute something like in Finland. Proper compensation of risk could certainly speed up finding suited locations. I'm all for it.
[2] https://en.wikipedia.org/wiki/German_Nuclear_Reactor_Insuran...
[3] https://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear...
[4] https://www.bundestag.de/resource/blob/877586/4e4dce913c3d88...
[5] https://foes.de/publikationen/2020/2020-09_FOES_Kosten_Atome...
EDIT: typos, grammar
So, in the most unrealistic scenario, a 1GW power plant could power 140.000 households. The grid may have to handle those kinds of peaks by using other sources of energy, but let's assume all we have is this one power plant.
Let's pretend we don't need to handle peaks. Assuming average yearly power usage of 10kWh, average power draw could be around 1.15kW. In that scenario, a 1GW plant could power 870.000 houses.
They generate cash at a ridiculously insane rate (literally few bucks off everyone connected to the internet growing by 10-15 cents every quarter).
What happens when you can produce cash faster than you can cultivate capacities to use it, and Wall St is constantly pushing you do something with it?
https://sustainability.google/operating-sustainably/stories/...
> To ensure that Google is the driver for bringing new clean energy onto the grid, we insist that all projects be “additional.” This means that we seek to purchase energy from not-yet-constructed generation facilities that will be built above and beyond what’s required by existing energy regulations.
Unless the nuclear plant would shut down without their money, they're just taking carbon credits from the wider grid. Amazon had one of their nuclear plans rejected for this exact reason.
See e.g. this post from Urs Hölze, one of the fathers of hyperscale computing: https://www.linkedin.com/posts/urs-h%C3%B6lzle_rethinking-lo...
Assuming ~$10M of capex (to buy the datacenter) per MW of electrical power (required by the datacenter), and hardware that is obsolete after 5 years (or even 10!), turning that datacenter off for an hour just to save like ~50$/MWh (or whatever spot price is) seems extremely counterproductive, because your hardware running for that additional hour is worth multiple times that (you spent like >$100 per operating hour on the hardware alone assuming 10 year lifetime).
It seems much more attractive (and credible) to just install more batteries (or even a gas turbine), instead of chasing demand-side-regulation pretensions.
edit: thx for the link though, that is a very interesting study/data even if I disagree with that conclusion!
You are wrong. It's not the science service of the Bundestag quoted in your link, but the exact same junk-science study by the anti-nuclear "Forum Ökologisch-Soziale Marktwirtschaft" that you quote below
See here: https://dserver.bundestag.de/btd/14/080/1408084.pdf
Antwort des Parlamentarischen Staatssekretärs Siegmar Mosdorf vom 15. Januar 2002 > In Deutschland sind bisher in Leichtwasserreaktoren ca. 3 225 Mrd. kWh erzeugt und in öffentliche Netze eingespeist worden. Subventio- nen für die kommerzielle Stromerzeugung aus Kernenergie gab es nicht. Allerdings wurde die Forschung auf dem Gebiet der Kern- energie durch öffentliche Mittel unterstützt.
Translated:
Answer from Parliamentary State Secretary Siegmar Mosdorf dated January 15, 2002 In Germany, around 3,225 billion kWh have been generated in light water reactors and fed into the public grid. There are no subsidies for commercial electricity generation from nuclear energy . However, research in the field of nuclear energy has been supported by public funds.
Anyway, nuclear power probably is not economical, but it also is not an existential threat, like climate change. So there are better hills to die on than this topic.