Project Cuts Emissions By Putting Data Centers Inside Wind Turbines (cnn.com) 168
CNN reports on a new German-based project called WindCORES that operates data centers inside existing wind turbines, making them almost completely carbon neutral. "If you look at the sustainability pyramid, the highest form of sustainability is using things that already exist," said Fiete Dubberke, managing director of windCORES, which was founded in 2018. From the report: The concept uses existing wind turbines to power data centers on site, while fiber optic cables provide a constant internet connection. Planning for a project like this began 10 years ago, Dubberke said, when WestfalenWIND realized the electricity grid was too weak to handle the huge capacities of electricity being produced by its wind turbines during peak wind hours, resulting in their windfarms being switched off due to grid security issues. WindCORES estimates that the unused electricity generated during this period could power one-third of all German data centers.
Its solution was to bypass the "middleman" (the grid) altogether, and instead, power IT servers from directly inside the large concrete wind turbine towers. Each tower is 13 meters wide and could potentially hold server racks up to 150 meters high. As the area is mostly empty space, Dubberke calls the concept a "no-brainer." According to Dubberke, an average of 85-92% of the power needed to sustain a windCORES data center comes directly from the host turbine. When there is no wind, electricity is obtained from other renewable sources, including solar farms and hydroelectric power plants, via the electricity grid. "The German data center average is 430 grams of CO2 released per kilowatt hour," he said. "For windCORES, it is calculated at just 10 grams per kilowatt hour."
Since launching, windCORES has acquired around 150 clients through co-location and cloud solutions, from very small start-up companies to bigger, more established ones, such as Zattoo, a leading carbon-neutral Swiss TV streaming platform with several million monthly users. Zattoo joined windCORES in 2020, when it moved one of its six data centers into a wind turbine in Paderborn. Currently, 218 channels are encoded with windCORES, and by the end of next year, the company hopes to relocate more existing servers to the wind farm, making it Zattoo's main data center location. [...] WindCORES has recently opened a larger, second location called "windCORES II" at the Huser Klee windfarm in Lichtenau, Germany. Built for a new large automotive client from Munich (the name is yet to be revealed), it is over three levels and around 20 meters high.
Its solution was to bypass the "middleman" (the grid) altogether, and instead, power IT servers from directly inside the large concrete wind turbine towers. Each tower is 13 meters wide and could potentially hold server racks up to 150 meters high. As the area is mostly empty space, Dubberke calls the concept a "no-brainer." According to Dubberke, an average of 85-92% of the power needed to sustain a windCORES data center comes directly from the host turbine. When there is no wind, electricity is obtained from other renewable sources, including solar farms and hydroelectric power plants, via the electricity grid. "The German data center average is 430 grams of CO2 released per kilowatt hour," he said. "For windCORES, it is calculated at just 10 grams per kilowatt hour."
Since launching, windCORES has acquired around 150 clients through co-location and cloud solutions, from very small start-up companies to bigger, more established ones, such as Zattoo, a leading carbon-neutral Swiss TV streaming platform with several million monthly users. Zattoo joined windCORES in 2020, when it moved one of its six data centers into a wind turbine in Paderborn. Currently, 218 channels are encoded with windCORES, and by the end of next year, the company hopes to relocate more existing servers to the wind farm, making it Zattoo's main data center location. [...] WindCORES has recently opened a larger, second location called "windCORES II" at the Huser Klee windfarm in Lichtenau, Germany. Built for a new large automotive client from Munich (the name is yet to be revealed), it is over three levels and around 20 meters high.
Inside the turbine? (Score:4, Insightful)
The drives that rotate inside will sing with joy! https://youtube.com/watch?v=oG... [youtube.com]
Or maybe you need to proof-read your headlines?
I can't take Germany seriously on CO2 emissions. (Score:2, Interesting)
Germany is clearly not taking CO2 emissions seriously if they continue their policy of not building new nuclear power plants. Here's a statement published by nations taking CO2 emissions seriously: https://www.energy.gov/article... [energy.gov]
In that commitment to increase nuclear energy capacity is over 20 nations, and that list includes a few of the largest CO2 emitters like USA, Canada, South Korea, and Japan. A few nations that are large CO2 emitters that didn't sign on are currently building more nuclear power p
Re: (Score:2)
Germany is clearly not taking CO2 emissions seriously if they continue their policy of not building new nuclear power plants.
You might be right but you are suggesting there is only one way to reduce emissions. You provided a single data point. I provided you with a counter point and you went quiet. Would you like more data?
Re: (Score:2)
You might be right but you are suggesting there is only one way to reduce emissions.
First, I'm suggesting nothing, I'm giving a report from the US Department of Energy that they are joining 20+ other nations on a commitment to increase nuclear power capacity in an effort to lower CO2 emissions. Second, you apparently didn't read the link since it includes the words "partner well with renewable energy sources" so there is no suggestion to use nuclear power to the exclusion of all else.
You provided a single data point.
I'd think the IPCC reports are a big data point. There's also reports from the US DOE, the UK government,
Re: (Score:2)
Re: (Score:2)
You made an anti-reality example? Because coal use is increasing in Germany. This is factual and undeniable, in spite of claims by your types for over a decade now that mass wind as used in Germany would reduce it.
Remind me, what is Einstein's definition of insanity? Though in this case, I would question is this is insanity or malevolence.
Re: (Score:2)
by your types
What is 'my types'? Those who can use more than one data point?
Because coal use is increasing in Germany. This is factual and undeniable
https://www.statista.com/stati... [statista.com]. The long-term trend is down. This is a fact. The number of years of increase? Just the last two. If we see it increase for another two or three you might have a point. At th
Re: (Score:2)
Why hasn't Germany made a commitment to expand use of nuclear power?
Because unlike you they realise that getting a nuclear plant built in any reasonable timeframe to meet their emissions targets won't happen, and also realise, again unlike you, that there's more than one way to reduce emissions. Incidentally Germany is actually reducing emissions.
Germany doesn't care about you. On the other hand I hope you realise that few people here are taking *you* seriously.
Re: (Score:3)
Incidentally Germany is actually reducing emissions.
Germany electricity emissions from the last 3 years [nowtricity.com]:
- 2020: 314g CO2eq/kWh
- 2021: 355g CO2eq/kWh
- 2022: 380g CO2eq/kWh
This looks like a trend going up.
Germany overall emissions are decreasing though I will give you that. Mainly because they are losing a lot of [politico.eu] industries [ft.com] due to their energy policy. They are basically getting poorer, which means less emissions. Nice plan.
Germany doesn't care about you.
Nor do we care about Germany. Physical laws on the other hand, cannot be ignored, and Germany should care about them. They are the reason w
Re: (Score:2)
I don't think looking at the last three years tells you much about whether Germany in on the right track or not.
1990: 764 g/kWh, 2000: 644 g/kWh, 2010: 555 g/kWh, 2019: 401 g/kWh
(source: https://www.umweltbundesamt.de... [umweltbundesamt.de]
This is hardly a failure to decarbonize the electricity grid. One can argue that it is too slow, but not that it is not working.
I have this discussion since at least a decade with nuclear proponents on Slashdot. Every single time something gets worse for a brief period, they take it as clea
Re: (Score:2)
The main reason the emissions changed from 1970 to 2019 is because they switched from using coal to using natural gas. Although it's called "natural," gas still releases 50% to 80% of the CO2 that coal does. Some of these emissions aren't always counted accurately, making the numbers seem lower than they really are. Before the war in Ukraine, Germany got gas from Russia through pipelines. Now, they import a lot more liquefied natural gas (LNG) from the US, which is less efficient because the gas has to be t
Re: (Score:2)
Germany electricity emissions from the last 3 years [nowtricity.com]:
Anytime I see someone post statistics narrowed down to a very specific time of geopolitical turmoil, one that has created outliers in an otherwise different looking trend, I don't even consider them stupid, I consider them dangerous. You sir are a dishonest piece of shit and you know it.
Re: (Score:2)
I see someone post statistics narrowed down to a very specific time of geopolitical turmoil
Because you think the geopolitical turmoil is going to calm down anytime soon? This is a direct result of political instabilities caused by climate change. As climate change will increase, geopolitical instabilities will increase.
You are in for a rude awakening. Call me when Germany emissions per kWh are at France or Norway level. I won't be holding my breath.
You sir are a dishonest piece of shit
Coming from you, that's a compliment. What's with the recent personal attacks though? Ran out of bad arguments?
Re: (Score:2)
Germany electricity emissions from the last 3 years [nowtricity.com]:
- 2020: 314g CO2eq/kWh
- 2021: 355g CO2eq/kWh
- 2022: 380g CO2eq/kWh
This looks like a trend going up.
Cherry picking data is fun!
What happens if you look at the previous 7 years? Oh, emissions are down. And even including the previous 3 years where they have gone up, it's still down over 10 years, indicating a overall downward trend. And if you look at all available data, you'll see that Germany's CO2 emissions peaked in the 1970s and have been on a sharp downward trajectory ever since 1990 [ourworldindata.org].
You cherry-picked some noise out of the trend line, which can clearly be seen elsewhere on the overall downward tre
Re: (Score:2)
Then you look at the bottom line [umweltbundesamt.de], and see something else entirely - germany has steadily slashed its total CO2 emissions 30% since 1990.
Re: (Score:2)
Yes, and emission factor (gCO2/kWh) for electricity production by more than 50%.
Re: (Score:2)
Germany, like most progressive Western states understands that underlying reality is irrelevant. Only perception matters.
This is why German Energiewende solutions are about giving the masses a perception of a very green nation. Look at all the subsidies. Look at all the wind turbines being built. Look at all these quirky, interesting sounding solutions like one in the OP. It must be a green grid! Just compare it to those evil, archaic French with their weird culture, who stick to their archaic nuclear power
Re: (Score:2)
Germany can't use new nuclear to achieve their legally mandated goals.
Oh, really?
The law currently says that all coal plants must be closed by 2038, although the government is currently planning to have them all gone by 2030. Given that that is only 6 years away, or 14 years for the hard legal deadline, it's not enough time to build new nuclear plants to replace the coal ones. They have to do it with renewables and gas.
If that is their plan then I expect them to miss their goals. Are you familiar with the definition of insanity? It's not a serious definition but the joke is that insanity is doing the same thing and expecting a different result. Germany has been trying to replace coal with renewable energy for a very long time now and has been failing. They can miss their goal with nuclear power or without it, either way they will miss their goal. Best they can do is start building nuclear power plants now
Re: (Score:2)
Not going to fail. It already failed several goals, including that of maintaining the reduction. They've had to switch to mass lignite in last few years due to increasing demand for peakers to handle "not windy" periods in a nation with low wind intensity.
They're now at the stage where they no longer count CO2 emissions from coal based peakers when they're running and synced to the grid, but not actively producing (because that's how you must run them for them to be able to function as peakers due to slow s
Re: (Score:2)
So far, Germany was successfully replacing fossil fuels with renewables and this brought down CO2 emission from its level 1990 fulling its target from the Paris agreement, i.e. 40% by 2020. Reductions in the electricity sector played a major role here. All your continued shouting does not change that renewables replaces fossil fuels in Germany and will continue to do so until all fossil fuels are gone. As a physicist, I have no idea what laws of physics this is supposed to violate.
Re: (Score:2)
What part of "there isn't enough time before the deadline to plan, site, construct, test, certify, and achieve first criticality" did you not pick up on? Why would they abandon the plan they've already been working on, which is anticipated to achieve the stated goal near the stated deadline in order to start over with a "put all your eggs in one basket" plan that is incredible susceptible to years of delay and multiples of cost overrun? Because you said so?
Re: (Score:2)
What part of "there isn't enough time before the deadline to plan, site, construct, test, certify, and achieve first criticality" did you not pick up on?
I understand just fine. What you don't understand is that with or without nuclear fission Germany will miss their CO2 reduction deadline. They can keep on this path that will never result in net zero CO2 emissions, or they can get back on the path of nuclear fission and reach their goal of net zero emissions beyond their self imposed deadline.
Why would they abandon the plan they've already been working on, which is anticipated to achieve the stated goal near the stated deadline in order to start over with a "put all your eggs in one basket" plan that is incredible susceptible to years of delay and multiples of cost overrun?
Who is saying anything about putting all their eggs in one basket? If anything adding nuclear fission to their mix of energy is the opposite of putting all their eg
Re: (Score:2)
Germany can't use new nuclear to achieve their legally mandated goals.
And they can't achieve their legally mandated goals without nuclear either. They will miss their goal wether they start building nuclear or not. The only difference is that the sooner they start, the sooner they will reach netzero (again: after having failed to achieve their legally mandated goals).
And you can argue that they will have to pay fines and whatever. Doesn't matter, their emissions will still be above their legally mandated goals, and the climate historically doesn't cool down by using fines mon
Maintenance (Score:2, Insightful)
Re: Maintenance (Score:2)
Better than the recent story of a Chinese data center that is under water
Re: (Score:2)
Re:Maintenance (Score:5, Insightful)
And how easy is the maitenance of a server when you have to change a hard disk in a tower 150 m tall?
In the story they say "could potentially" which is media speak for "will probably never". If you look at some photos of turbines they are wide at the bottom and taper at the top in order to increase stability. They have lots of space low down and less higher up. So to begin with you will probably have simple standard height racks in the base. Then, once the concept is proven and you have lots of customers, you might have a two or three story construction inside the base of the turbine but with normal stairs and floors so that data service engineers never need to have working at height certification. Even now, wind turbines are big enough that they have lifts inside [youtu.be] so space for a data centre isn't a problem.
By the time you got to 150M high up inside, I'm not sure if there's enough space for all the access needed, though probably some of the new ones which can be up to about 300M high [euronews.com] have lots of space at that level.
This is so freaking stupid.
I really don't get the level of negativity here on Slashdot on everything related to renewable energy. I think this must be coming from massive levels of fossil fuel propaganda. This is a fairly clear and really interesting technology
When I began to read the news I thought that the innovation was that the turbine provided free refrigeration but that doesn't seem the case.
A wind turbine is an incredibly tall pipe with a hole at the bottom, for the door, and at the top (for the connection to the nacelle). That's otherwise known as a chimney. If you put a small amount of heat at the bottom of a tall chimney then you get a huge draw of air caused by the buoyancy of the heated air. They don't mention it but I'm almost 100% sure that they can't avoid using that.
Re: (Score:2)
>I really don't get the level of negativity here on Slashdot on everything related to renewable energy.
It comes from the fact that basic math and logic is understood by a larger percentage of users here vs general populace. Has to be, because much of IT is driven by basic math and logic.
It doesn't stop people like yourself who have been thoroughly propagandized to leave their math and logic skill behind when talking about "data centers in the clouds" (literally in this case!), and then project this propa
Re: (Score:2)
I really don't get the level of negativity here on Slashdot on everything related to renewable energy
There is little to no negativity here on Slashdot when it comes to renewable energy. Most here are proponents of it and recognize that our dependence on fossil fuel is a dead end. However, most "renewable energy" products tend to be gimmicks and there is strong negativity towards them.
As an analogy, there is no negativity here on Slashdot when it comes to medicine. But if most "medical innovations" were snake out, you would see a lot of negativity in articles about supposed medicine.
Re: (Score:2)
A wind turbine is an incredibly tall pipe with a hole at the bottom, for the door, and at the top (for the connection to the nacelle). That's otherwise known as a chimney. If you put a small amount of heat at the bottom of a tall chimney then you get a huge draw of air caused by the buoyancy of the heated air.
Re: (Score:2)
Very easy. See, it's a tower, and towers have a bottom sitting on the ground. So you put your servers down there. And if you want go higher, you put a ladder and have floors inside it.
They also get pretty darn big, so things can be pretty roomy inside.
Re: (Score:2)
Go to their website. There is literally a short video showing that the hardware is installed at the base of the tower.
Re: (Score:2)
And how easy is the maitenance of a server when you have to change a hard disk in a tower 150 m tall?
Like those underwater data centres, they won't bother. The machines will either be diskless or just have some high reliability solid state storage. If they break they will be retired, not replaced. The failure rate will be calculated to be low enough that over the expected lifetime of the product that there is no need for on-site maintenance.
When the hardware becomes obsolete and of no use to anyone, it will have already made a profit and will simply be abandoned until the turbine itself reaches EOL.
Re: (Score:2)
And how easy is the maitenance of a server when you have to change a hard disk in a tower 150 m tall?
Simple, you walk up the internal stairs or climb the internal ladder. I know, I know, actually getting to a place without a mobility scooter is a scary thought for an American.
Re: (Score:2)
I see you are stuck in the past. Modern servers do not have hard disks.
Re: (Score:2)
Perhaps the data center is located at the bottom of the windmill?
Re: (Score:2)
I know this is a hard one to visualize, but maybe put the higher maintenance equipment in racks at the bottom so it's easier to get to them?
Like, say, all your storage racks with mechanical devices that have a reasonable expectation of failure? And then you know what you put at the top, which is the least accessible? The shit you never have to physically touch.
Man, that was a hard solution to come up with, wasn't it?
Workaround for poor planning (Score:2)
So either Germany has a poor electric grid that cannot handle the currents needed, or the wind farm was so poorly planned that they forgot to check if the transmission lines can handle their power when the location was chosen. The workaround is to try to move the load into the farm to avoid the electric grid altogether.
Perhaps they could plan better the next time when they choose a location for any kind of power generation? Maybe make sure the transmission lines can handle the power generated first?
Re: (Score:2)
The electric grid in Germany is having problems. I found one report on this as an example: https://www.bloomberg.com/news... [bloomberg.com]
I'm not a fan of Bloomberg as they like to hide things behind a paywall but that was a link that came out on top of my search.
I can recall this going back years with Germany trying to get more power lines running north-south. I recall it was something like they had an abundance of wind power in the north but needed that electricity in the population centers to the south, or something
Re: (Score:3)
A problem with this plan is what to do when the wind isn't blowing. Do they shut down all the computers?
Right there in TFS: "According to Dubberke, an average of 85-92% of the power needed to sustain a windCORES data center comes directly from the host turbine. When there is no wind, electricity is obtained from other renewable sources, including solar farms and hydroelectric power plants, via the electricity grid."
The grid connection is bi-directional. Who knew? ;-)
Re: (Score:2)
It's almost like grid is there for a reason, and instead of putting a small scale data center in the clouds, it makes sense to just put it in an existing large data center with economies of scale, and use that same bidirectional connectivity that exists across the grid to feed that?
Re: (Score:2)
Re: (Score:2)
Six degrees of separation to blame Trump is real. A mind of a fanatic is truly a fascinating thing.
Re: (Score:2)
Six degrees of separation to blame Trump is real. A mind of a fanatic is truly a fascinating thing.
Nope, the Bavarian Christian Social Union is slightly smarter than MAGA (for example they don't think Jeebus wrote the King James Bible in English) and they have fewer guns but they are basically not that far removed from being a German version of MAGA.
Re: (Score:2)
Bavaria is governed by a party that is very similar in mind to the MAGA republicans. They've been blocking building new power lines or wind turbines for decades, classic NIMBY tactic.
I'm getting the impression that you don't know what it means to be a "MAGA Republican". I don't know of any Trump era policy that was blocking power lines or windmills, though there's plenty of examples of Democrats doing so. Trump made an effort to improve America's energy independence, though arguably that would have been diminished with support for natural gas pipelines from Canada. Democrats don't like pipelines.
Their neighbors aren't exactly better. Ironically, the same states also categorically rules out the placement of any long term nuclear waste storage facilities, even for the waste they produced locally.
Trump appointed Rick Perry as Secretary of Energy and Perry said to put the nuclear waste
Re: (Score:2)
I'm not aware of any MAGA initiatives to block expansion of the grid, and am darn sure the farmers in my area (mostly MAGA country) don't mind leasing their land for windmills. They're everywhere here.
Re: (Score:2)
The electric grid in Germany is having problems. I found one report on this as an example: https://www.bloomberg.com/news... [bloomberg.com]
That's grid management of the type that has been offered in most Western nations for five or more decades, such as the UK's Economy 7 which has been going at least from the 1970s.
I can recall this going back years with Germany trying to get more power lines running north-south. I recall it was something like they had an abundance of wind power in the north but needed that electricity in the population centers to the south, or something like that.
That's planning rather than a problem per se.
Re: (Score:2)
A problem with this plan is what to do when the wind isn't blowing. Do they shut down all the computers?
Why do you have such a problem understanding that electricity can flow both ways through a wire, and bi-directional metering exists? You repeatedly bring up this "problem" even though it's been solved for literally decades.
Re: (Score:2)
You overbuild generation capacity. Energy sources often come with the potential for excess energy, but unlike conventional power sources, for renewable sources that excess energy doesn't come with a fuel cost, so you might as well use it. It hardly matters where the servers are. Other energy intensive industries also colocate with power sources, because building long distance transmission lines is expensive.
Re: (Score:2)
Other energy intensive industries also colocate with power sources, because building long distance transmission lines is expensive.
Only when you have no other choice. Transporting materials in and products out is also expensive, so is trying to hire people to work in far away locations. Anything you need takes longer and more expensive due to the extra transport.
When you setup your factory or whatever in a remote location just to stay near power source, everything else got more expensive and you lose out to your global competitors who can locate where it makes the most economic sense without worrying about getting power (i.e. in coun
Re: (Score:2)
Re: (Score:2)
building long distance transmission lines is expensive.
For wind generation they will be put in place anyway. Yes, it's an issue on a continental scale but on a national scale those interconnections are going to be planned anyway. Sure, sometimes building of wind farms might exceed the transmission lines for a while but you are going to put those lines in anyway to ensure your country has the flexibility to place industry relatively straightforwardly. Yes, there might come a point when you think you are always going to have excess in a particular spot and transp
Re: (Score:2)
So either Germany has a poor electric grid that cannot handle the currents needed, or the wind farm was so poorly planned that they forgot to check if the transmission lines can handle their power when the location was chosen.
Neither. It's someone pitching a business proposition.
Re: (Score:2)
Reduce (Score:2)
"the highest form of sustainability is using things that already exist"
Wrong, the highest form of sustainability is not using them in the first place.
Re: (Score:2)
True, but telling people that they have to decrease their standard of living is a tough sell. So, for the same amount of input energy, you often have a more beneficial impact on the overall system by focussing on 'reuse' and 'recycle' than you do on 'reduce'. It's similar to the way that Just Stop Oil is having a negative impact on the willingness of people to consider doing _anything_ to fix the problem.
Very cool, but... (Score:2)
How do you make a power generating facility an even more attractive terrorism target? Oh, wait, I know! Co-locate a bunch of server farms!
OTOH I guess this could count as an increase in resiliency, if the servers would otherwise be located with others in a bigger cluster elsewhere.
Re: (Score:2)
Blowing up wind and/or server farms are unpopular approaches to terrorism because it doesn't result in obvious physical suffering or death. To a first approximation, nobody would be terrified by such an attack.
Nuclear power plants are a special case because of the radiation and the possibility of going critical. Other power plants would just be an economic inconvenience.
Re: (Score:3)
Nuclear power plants are a special case because of the radiation and the possibility of going critical.
Nuclear plants are special, because of dumb claims like this one (either said on purpose, or throug sheer ignorance). In a nuclear power plant, the term "critical" doesn't imply a dangerous situation; rather, it refers to a stable state of nuclear fission reactions within the reactor core. A nuclear reactor operates in a critical state when there is a balance between the rate of neutron production (from fission reactions) and neutron absorption in the reactor core. This critical state allows for a sustained
Re: (Score:2)
Let's take an even worse situation, what actually happens during a meltdown in the middle of a Tsunami with waves 10m height? Well, this is what happened in Fukushima in 2011. Guess how many deaths resulted from the meltdown? 1.
That was one suspected death that was paid insurance as if caused by radiation from the meltdown. It's difficult to track the cause of any cancer so it's as easy to blame radiation as secondhand smoke, assuming the victim wasn't a smoker. I believe it was wise to pay out, disputing the cause would not likely do well for the insurance company costs and would only add fuel to the fire of how heartless and/or dangerous the nuclear power industry is in Japan. Had there been more cancer deaths blamed on the m
BMW? (Score:2)
Built for a new large automotive client from Munich (the name is yet to be revealed), it is over three levels and around 20 meters high.
Seems that BMW fit that description.
Re: (Score:2)
Re: (Score:2)
That's part of the clever bit. The other part is because they have a large on-site customer for the power they are generating, they can continue generating when the grid otherwise wouldn't be able to accept the energy they are over-producing. So they essentially get to operate the "data center" for free under those conditions.
warning wind to low server shutdown underway! (Score:2)
warning wind to low server shutdown underway!
Re: (Score:2)
warning slashdot poster can't be bothered to read the summary; nonsense post underway!
Copper wires don't give a shit which direction electricity flows.
Re: (Score:3, Interesting)
I think a detail was probably missed here that would make sense.
If they were to build the windfarm towers with cooling holes in them to take advantage of the colder air, then it pretty much cools itself when powered by wind.
That said, I feel this has the same argument of the "baseload" that is better solved with batteries. Just stick battery stacks inside the tower to charge when it can't put the power on the grid, and draw down the power from the battery stacks when wind isn't operating at 100%
Remember tho
Re: (Score:3)
Also, compute kit can be very capital intensive not sure how the cost of making it secure enough stacks up.
Maybe with the AI/ML boom, the energy-hogging training workloads could be suspended when the wind doesn't blow cheap energy your way, but that would leave your chip investment unproductive for those periods:
Re: (Score:2)
The opposite is obviously the hot days, which reduces active days further. I.e. not only does your heavy investment in a data center in the clouds (literally, the irony is real) turn off when it's not windy enough, but it also turns off when it's windy and hot.
Data centers are not free. Not in terms of investment and not in terms of carbon emissions. They're not purely operating costs and nothing else.
Re: (Score:2)
Battery back-up (or other "green" back-up where possible) may become a necessity in a reduced CO2 world, but it's not cheap and simple.
A typical newer wind turbine is around 3 MW. Let's say you have one of those and it outputs 1.5 MW
Re: This is so fucking stupid (Score:2)
Google runs a data center in the Arctic Northern Sweden in an old mine. How is this worse?
Re: (Score:2)
Re:This is so fucking stupid (Score:5, Insightful)
"High speed internet interconnects" are available anyway, because these turbines are managed remotely. The big ones have redundant connections, and since they're fairly new, they're not legacy phone lines but obviously fiber, because it's the cheapest non-mobile technology. They had to put an electricity grid connection there, so obviously throwing a couple fiber strands in there too hardly registers as an extra cost.
They're not "bypassing the grid, period", just bypassing it for the times when they can't use it because of a glut of wind-generated electricity that exceeds grid capacity. Powering their servers from the grid at times when the wind doesn't blow is not a problem.
Re:This is so fucking stupid (Score:5, Interesting)
They had to put an electricity grid connection there, so obviously throwing a couple fiber strands in there too hardly registers as an extra cost.
It's even better than that. Modern power cables (mostly made of solid copper) are so expensive and special that it's actually a default to put quite a number of optic fibres inside each one. Normally the only change required is to actually terminate these fibres into internet transmission equipment.
Data batteries are 100% efficient (Score:4, Insightful)
There's lots to this if the other practical of this don't sink it.
Lots of times, say at night, the wind blows but there's no demand for the electricity. You could store it in batteries but then you pay an efficiency cost of conversion into and out of the battery. If the battery isn't local to windmill or eventual use point you pay the transmission line cost. And the batteries have to be large ( expensive).
The servers convert the surplus electricity to useful products directly with no inherent inefficiency over any other server doing the job that is powered off the grid. So in that sense it's perfectly efficient. The data product is inexpensive to transport over Internet.
At times when the windmil is not producing power at all the data center could in principle be powered by sending power in the opposite direction back to the windmill site from the grid. You'd need a transformer there I suppose but not a new set of transmission lines. So lower maintainence and fixed costs for the power distribution system to the data center
But you do need a data product whose production can shut down when the windmill is producing power for the grid at max capacity. That's easily conceivable as computers can also be gridded around the world and their computations moved from center to center ( eg night on other side of planet ). Or you pick a data commodity that can be episodically produced. I shudder to mention Bitcoin but at least for the time being that's actually practical --- module your feelings about whether bitcoins wasteful work model makes sense. Another one would be high compute loads for jobs like designing proteins, scheduling airlines, training AI models, massive fluid dynamic simulations , weather modeling, etc.... and all sorts of scavenger computing jobs.
A bonus for this model is that you could build out wind power stations in advance of power demands ( like before you decommission your coal and nuke plants ) and have a use for them . If you built out enough excess capacity then you also solve one of the reasons we have these traditional plants -- the need to have enough power when winds are insufficient. With enough excess capacity and well geographically distributed then wind fluctuations won't ever need much base load backup power from coal
Re: Data batteries are 100% efficient (Score:4, Interesting)
and with the heat from the servers concentrated conveniently you could consider what to recycle the heat for.
Examples might be chemical processes that need heat like desalination or conversion of methane to Syngas or making concrete or greenhouses and shrimp farms.
Re: (Score:3)
All large windfarms will have batteries soon...
Batteries it can double a windfarm's usefulness so they pay for themselves very quickly.
Re: (Score:2)
All large windfarms will have batteries soon...
Batteries it can double a windfarm's usefulness so they pay for themselves very quickly.
I could say that all nuclear power plants will have batteries soon, that would address the problem of ramping big slow steam turbines up and down with changing demand while lowering CO2 emissions. Batteries at nuclear power plants can double their usefulness, allowing them to pay for themselves very quickly. We can park large electricity consumers close to nuclear power plants too, and consumers of heat and steam. Perhaps we can put nuclear power plants and desalination plants close to each other so plac
Re: (Score:2)
"High speed internet interconnects" are available anyway, because these turbines are managed remotely. The big ones have redundant connections, and since they're fairly new, they're not legacy phone lines but obviously fiber, because it's the cheapest non-mobile technology. They had to put an electricity grid connection there, so obviously throwing a couple fiber strands in there too hardly registers as an extra cost.
They're not "bypassing the grid, period", just bypassing it for the times when they can't use it because of a glut of wind-generated electricity that exceeds grid capacity. Powering their servers from the grid at times when the wind doesn't blow is not a problem.
High speed interconnects, on the level of entire wind farms, are required. They are not necessarily required for individual turbines, however. I've relevant experience in the area. And yes, they tend to provide interconnects along with the transmission lines.
However, putting servers in the turbines themselves doesn't make much sense as that makes maintenance of the servers complex. You are better off having a DC for a wind farm but that doesn't obviously work for onshore. Ideally you put DCs where it ma
Re: This is so fucking stupid (Score:2)
I agree that colocating other people's hardware in a wind turbine makes little sense, unless they are all cloud servers. In that case you spend your maintenance money overprovisioning and just abandon failed hardware until you have enough to make a site visit worth it.
Re: (Score:2)
Re: This is so fucking stupid (Score:2)
>cloud servers
>60m in the air
I see what you did there
Re: (Score:2)
>"High speed internet interconnects" are available anyway, because these turbines are managed remotely
Didn't expect this level of ignorance from slashdot. How much data do you think remote management interface of a turbine requires?
How much data throughput do you think a typical data center requires?
You stopped reading too soon. The key point was in the next sentence:
For runs of more than a few hundred meters, fiber does quickly get to be cheaper than copper. If it's true that the economics of wiring motivate the use of fiber rather than copper for the remote management interface (which seems possible) or even that using fiber doesn't cost much more (which seems likely), then t
Re: (Score:2)
What, do you think people installing state-of-the-art wind turbines are running a 40+ year old POTS 2-wire line out there for management and control, and using some 40+ year old dogshit analog modem or ISDN?
They are already installing fiber to sufficiently large wind farms because it's cheaper than any other solution except maybe LTE, but I doubt that they are using LTE for anything other than a backup to the wired connection due to the need for positive control over the production of power to keep the grid
Re:This is so fucking stupid (Score:4, Interesting)
It's worse than that, and completely in line with current German "innovations" in power grid things. When wind doesn't blow, your "data center" either turns off, or needs electricity supplied to it from the grid that has been "bypassed".
This is badly explained in the summary but the thing about wind turbines is that it makes sense to build them bigger than the capacity of the power connection to them. That's because grid connections are very expensive so if you build more capacity then you can ensure that even in light wind situations that grid connection is kept completely full. At that point, you have an excess of available electricity at the wind turbine most of the time. Your data center can run on "free" electricity most of the time.
Yes, when the wind is not blowing where you are, you have to import electricity from elsewhere (remember the wind is always blowing somewhere) ans so yes, you're right you don't bypass the grid. The opposite in fact, you have a free grid connection which can easily run in reverse and handle the load of the data center. That and the fact you get fiber optic connections for free, because all large scale power cables come with them integrated nowadays, means that you are overall reducing transmission costs.
This is the same logic which means it's sensible to have a solar farm on the roof of a data centre or a small number (but not lots - almost all places I know of limit home solar) of solar panels at home. As long as the load or generation you are adding to the system is smaller than the opposing generation or load that you are siting it next to, you basically get your transmission for free.
Re: (Score:2)
The actual logic behind solar panels on data center is the massive dumping done by Chinese manufacturers combined with hilarious subsidy regimes in places where sun doesn't shine. Without those two, the idea of CO2 and CO emissions disaster that is solar in regions where sun doesn't shine would have never been adopted outside high intensity solar areas.
Re: (Score:2)
That and have you ever seen a wind turbine have a structural/stress failure?
You really want to park an expensive data center under one of these?
https://www.youtube.com/watch?... [youtube.com]
Re: (Score:2)
Re: (Score:2)
That and have you ever seen a wind turbine have a structural/stress failure?
You really want to park an expensive data center under one of these?
You could put it 1/4 mile away. The chances of impact will be incredibly low. If you look at onshore wind farms they tend to have some sort of buildings on the highway and then the turbines are quite spread out, but there can often be at least 1/8 or maybe 1/4 mile from the buildings to the turbines as you don't want a turbine so close to the road such that if one fails it blocks the road. So you'd just be looking at more buildings near the road. I still don't think it makes sense, though, as by tying your
Re: (Score:2)
So you need a separate housing, separate interconnects for power and data, and none of the advantages of a centralized data center.
Like I said above, this is the level of brilliant German grid solutions, the kind that blew their CO2 grid emissions up after long history of reducing them. When you start with a stupid end point, and then try to cheat the system and the numbers to pretend that it makes sense.
And then you have to obfuscate the emission numbers at the tail end so you don't have to answer for your
Re: (Score:2)
So you need a separate housing, separate interconnects for power and data, and none of the advantages of a centralized data center.
Yes, it makes slightly more sense in a DC/farm but even then not a lot.
Like I said above, this is the level of brilliant German grid solutions, the kind that blew their CO2 grid emissions up after long history of reducing them.
https://www.statista.com/stati... [statista.com], https://www.statista.com/stati... [statista.com], https://www.nowtricity.com/cou... [nowtricity.com], . It's still going down, apart from the last two years. Two years isn't long enough to say if there's a long-term trend up, but the long-term trend for a long time has been down.
Re: (Score:2)
Re: (Score:2)
"Project Cuts Emissions By Putting Data Centers **Inside** Wind Turbines"
Otherwise, this "news" is nothing of the sort. It's just the same way we currently do EVERYTHING.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
It's worse than that, and completely in line with current German "innovations" in power grid things. When wind doesn't blow, your "data center" either turns off, or needs electricity supplied to it from the grid that has been "bypassed".
This in addition to needing high speed internet interconnects to each such wind turbine now.
Here we go again, When the wind stops blowing the lights go out!! ... Mr Trump? ... is that you? ... As is explained in the summary, the date centre uses energy that otherwise would go to waste when the wind farms are switched off due to grid security issues during peak wind hours. When the wind is not blowing the data centre can be powered from the gird using solar or hydro power. In fact, you could even co-locate the wind park with a solar plant and grid battery facility that can at least cover the data-c
Re: (Score:2)
>As is explained in the summary, the date centre uses energy that otherwise would go to waste when the wind farms are switched off due to grid security issues during peak wind hours.
That's cute. Data center that is only on a small fraction of the time, while requiring full investment. Why not just have it anywhere else in the grid, if it basically blacks out every time wind is too much or too little, since there you can get benefits of centralization and much easier maintenance?
Again, this is the level o
Re: (Score:2)
>As is explained in the summary, the date centre uses energy that otherwise would go to waste when the wind farms are switched off due to grid security issues during peak wind hours.
That's cute. Data center that is only on a small fraction of the time, while requiring full investment. Why not just have it anywhere else in the grid, if it basically blacks out every time wind is too much or too little, since there you can get benefits of centralization and much easier maintenance? .... blah blah asinine crap.
What's cute is that you seem to think that the shit you post is a really brilliant and subtle trolling or ... you are actually stupid enough to believe that crap. I'm going to assume the latter. They are co-locating the data centre with the power supply to take advantage of a periodic superabundance of extremely cheap power and they are getting the space to house their data centre basically for free. The data center is still connected to the grid, it draws power from the grid whenever the cheap wind power f
Re: (Score:2)
You know, there are some places where the wind effectively blows 95% of the time. And perhaps the data center is for someplace like Google where the data is replicated and if its offline there are other options. But for that x% it is available it can run off much cheaper power source.
I'm not saying it's worse or better. But you need to look at the requirements and costs for everything before you judge whether this is stupid or genius.
Re: (Score:2)
Both of you are propagandist morons, and I'll wager you don't even have petrochemical stocks.
What part of 85%-92% of all the power needed do you not get?
Re: (Score:2)
Why would it be any more work to install a fiber line while you're running the big ass electrical transmission lines to the wind turbine? Sure, there's technically one more thing to do, but it can be done at the same time, in the same utility trench or poles.
And why would you expect a company that operates wind farms to not know that wind can be intermittent? People keep trotting this out and it makes no sense - do you think that everyone at this company missed the fact that wind doesn't constantly blow a