This is just a rehash of the map already available at datacentermap.com (it’s literally cited as their source). Not sure what KXAN is adding here.
Also worth noting that a very loose definition of data center is used here that doesn’t fit the current colloquial usage of the term. One of the “data centers” listed in the middle of Austin, for example, is barely the size of a house and is actually just the switching infrastructure for a radio tower. Not every dot on this map is an energy-guzzling hyperscaler boogie man.
Totally agree. The story is similar for all the dots in the middle of nowhere and small towns. That's half the map!
If people are so curious, they should head on out there and see what the fuss is about. They're going to find a lot of sheds and random closets with a single rack at most.
Knowing how this state operates, I wouldn't doubt a lot of these ended up on the list for small business tax purposes or subsidies. That's a different and much less interesting news story, though.
A lot of these are decidedly not in the middle of nowhere. The corridor from San Antonio to Dallas, passing through Austin, is almost a continuous city, and it's becoming that.
Despite the lobbying effort, it's still true that over 90% of data centers are not closed loop. Per-acre data centers use much more water.
More important politically, is that your water bill will absolutely go up faster with a new data center in town than with a new golf course. Dc's usually draw from a single pipe while golf courses and farmland spread it out. Also a golf course usually uses primarily greywater or recycled water.
The grievances about water usage are absolutely well-founded. Industry sockpuppets often like to push statistics about closed-loop data centers (which are less than 10% of all data centers) to fight back. Closed-loop data centers have their own drawbacks like drastically increased electricity usage and added pollution concerns due to the toxic sludge that needs to be bled out every month.
The golf course line has never been convincing. Find something new.
>> Industry sockpuppets often like to push statistics about closed-loop data centers
>> So I'm going make my own slew of claims, yet provide absolutely zero sources for said claims, I'll just present my opinions as if they're known fact
I responded to a lazy comment with a higher effort comment. I think I left the conversation better than I found it. Not sure which part you'd like sourced, but here's a paper I think does a good job of covering some of these topics
> While data center electricity use has been the subject of much investigation, data center water consumption is a less studied topic that is becoming increasingly important in the context of water scarcity and freshwater conservation, especially at local levels. Many data centers can consume a significant amount of water directly for cooling and space conditioning
> For example, Google's data centers in The Dalles consumed 355 million gallons of water in 2021, representing 29 % of the city's total water consumption
> Additionally, data centers consume water indirectly through the water used for electricity generation. This type of water use can be significant at the local or national level depending on the power generation technologies, power mixes, and electricity usage of the data centers.
I predict "data centers" are going to start being built but they'll say it's a distribution warehouse for motorcycle parts or a factory or something. There'll be a token amount of floor space set aside to manufacture pen cartridges but then there'll just be an incidental 200000 sq ft of server racks.
One of the heaviest water consumers and a massive user of land (especially when considering the land needed to grow feed). And data centers aren’t creating lagoons of shit.
And that’s even before you get to the moral atrocity of dairy.
It's a little dated but I'm sure it hasn't changed much since 2024.
Tulare County (CA), the Panhandle region (TX), and the Magic Valley (ID) consistently rank as the highest-density clusters for the largest individual dairy herds in the country.
Government exists to serve corporations not individuals in this country. They can pretend these are confidential and have various cover to not release info as a result.
One question I've got is why Texas? It gets very warm there and DCs need cooling. Wouldn't like Wisconsin or Minnesota or Michian be cheaper. I mean Michigan's whole schtick is land of 1000 lakes, and that's not even considering Alaska which is massive, unpeopled, and colder than a nun's ruler?
I'm not saying for or against, I'm just curious why they are choosing Texas?
It's not even taxes. It's the fact that municipalities and cities are making deals with these companies to give them, e.g., 10 years of tax breaks to build there. If you live in Texas, you are subsidizing these companies
It's easier to get permits in Texas than most other states, which is especially important for the power generation that some of these data centers are building.
My takeaway from this is apparently you can build a datacenter in a place where the weather reaches 110 degrees in the summer. AC has come a long way. I wonder what cost feasability of building essentially a biodome that is actively climate controlled that spans an entire town or city.
Why is that a pipe dream? It’s BTM for at least one of the sites you are talking about. The only constraint, which is real, is availability of turbines but other than a delayed timeline i am not sure what’s a pipe dream.
are you off by a K? from the article - 'As of early August, ERCOT was considering more than 474 gigawatts of requests to connect to the Texas electric grid, about 90% of which came from data center projects.'
This is just a rehash of the map already available at datacentermap.com (it’s literally cited as their source). Not sure what KXAN is adding here.
Also worth noting that a very loose definition of data center is used here that doesn’t fit the current colloquial usage of the term. One of the “data centers” listed in the middle of Austin, for example, is barely the size of a house and is actually just the switching infrastructure for a radio tower. Not every dot on this map is an energy-guzzling hyperscaler boogie man.
KXAN is adding eyeballs. Maybe.
Totally agree. The story is similar for all the dots in the middle of nowhere and small towns. That's half the map!
If people are so curious, they should head on out there and see what the fuss is about. They're going to find a lot of sheds and random closets with a single rack at most.
Knowing how this state operates, I wouldn't doubt a lot of these ended up on the list for small business tax purposes or subsidies. That's a different and much less interesting news story, though.
It's almost like MOST OF TEXAS is 'in the middle of nowhere and small towns.'
lol
A lot of these are decidedly not in the middle of nowhere. The corridor from San Antonio to Dallas, passing through Austin, is almost a continuous city, and it's becoming that.
I much prefer the KXAN presentation of the data and the map layer styles.
Did this claim to be a 'hyperscaler' map?
And here's a map of major water users in Texas: https://golfcoursebrowser.com/
Despite the lobbying effort, it's still true that over 90% of data centers are not closed loop. Per-acre data centers use much more water.
More important politically, is that your water bill will absolutely go up faster with a new data center in town than with a new golf course. Dc's usually draw from a single pipe while golf courses and farmland spread it out. Also a golf course usually uses primarily greywater or recycled water.
The grievances about water usage are absolutely well-founded. Industry sockpuppets often like to push statistics about closed-loop data centers (which are less than 10% of all data centers) to fight back. Closed-loop data centers have their own drawbacks like drastically increased electricity usage and added pollution concerns due to the toxic sludge that needs to be bled out every month.
The golf course line has never been convincing. Find something new.
>> Industry sockpuppets often like to push statistics about closed-loop data centers
>> So I'm going make my own slew of claims, yet provide absolutely zero sources for said claims, I'll just present my opinions as if they're known fact
Well, you surely convinced me.
I responded to a lazy comment with a higher effort comment. I think I left the conversation better than I found it. Not sure which part you'd like sourced, but here's a paper I think does a good job of covering some of these topics
https://doi.org/10.1016/j.resconrec.2025.108310
Some highlights:
> While data center electricity use has been the subject of much investigation, data center water consumption is a less studied topic that is becoming increasingly important in the context of water scarcity and freshwater conservation, especially at local levels. Many data centers can consume a significant amount of water directly for cooling and space conditioning
> For example, Google's data centers in The Dalles consumed 355 million gallons of water in 2021, representing 29 % of the city's total water consumption
> Additionally, data centers consume water indirectly through the water used for electricity generation. This type of water use can be significant at the local or national level depending on the power generation technologies, power mixes, and electricity usage of the data centers.
I predict "data centers" are going to start being built but they'll say it's a distribution warehouse for motorcycle parts or a factory or something. There'll be a token amount of floor space set aside to manufacture pen cartridges but then there'll just be an incidental 200000 sq ft of server racks.
Actually I'm sure they're already doing that.
Now do dairy operations…
One of the heaviest water consumers and a massive user of land (especially when considering the land needed to grow feed). And data centers aren’t creating lagoons of shit.
And that’s even before you get to the moral atrocity of dairy.
https://storymaps.arcgis.com/stories/cc02f97b1129447db2420eb...
It's a little dated but I'm sure it hasn't changed much since 2024.
Tulare County (CA), the Panhandle region (TX), and the Magic Valley (ID) consistently rank as the highest-density clusters for the largest individual dairy herds in the country.
Also: https://ever.ag/map-downloads
How can locations be labeled secret (for 29 data centers)? Aren't there hearings or filings for proposed data centers?
Government exists to serve corporations not individuals in this country. They can pretend these are confidential and have various cover to not release info as a result.
Border patrol, law enforcement, FBI, etc. There's a few military bases as well. I'm not sure what else you're expecting in Texas.
If you look at google maps, the "secrets" will become a lot less mysterious.
One question I've got is why Texas? It gets very warm there and DCs need cooling. Wouldn't like Wisconsin or Minnesota or Michian be cheaper. I mean Michigan's whole schtick is land of 1000 lakes, and that's not even considering Alaska which is massive, unpeopled, and colder than a nun's ruler?
I'm not saying for or against, I'm just curious why they are choosing Texas?
It's not geography, it's politics. Low taxes, general amenability to big businesses, and low ecological restrictions.
It's not even taxes. It's the fact that municipalities and cities are making deals with these companies to give them, e.g., 10 years of tax breaks to build there. If you live in Texas, you are subsidizing these companies
It's easier to get permits in Texas than most other states, which is especially important for the power generation that some of these data centers are building.
Some of these are just office buildings with multiple companies. Are we really calling a rented space with servers a data center now?
My takeaway from this is apparently you can build a datacenter in a place where the weather reaches 110 degrees in the summer. AC has come a long way. I wonder what cost feasability of building essentially a biodome that is actively climate controlled that spans an entire town or city.
How does this compare to the crowdsourced data included on https://www.brockovichdatacenter.com/ ?
Some of these are pipe dreams. A Microsoft datacenter at 2.5MW, another at 2MW. There's no way they're going to get generation capacity for that load.
Why is that a pipe dream? It’s BTM for at least one of the sites you are talking about. The only constraint, which is real, is availability of turbines but other than a delayed timeline i am not sure what’s a pipe dream.
2.5 MW / 2 MW are tiny. I think you are confused.
The office building I’m sitting in has a 2 MVA transformer feeding a 2500A 480V three-phase service, 2 MW is peanuts.
are you off by a K? from the article - 'As of early August, ERCOT was considering more than 474 gigawatts of requests to connect to the Texas electric grid, about 90% of which came from data center projects.'
Now I'm curious what the other 10% is.
The entire rest of society? Schools, hospitals, homes, businesses, etc