If there’s one point of agreement regarding a proposed data center campus in Shalersville Township, it’s that the installation would increase noise.
How much, and what kind, gets complicated.
Bitdeer Technologies Group, the Singapore-based bitcoin mining company, claims that the noise level at the property line of its proposed Shalersville campus would be under 50 decibels.
Shalersville Township trustees want...
Read morePutting my biases up front, I am against data centers generally.
As we consider the potential impacts of data center noise it is important to keep in mind that noise quality is as important as noise quantity. A lot of industrial installations (including data centers) produce elevated levels of infra sound which we can’t hear with our ears but can still have effects on our bodies and mental health. For those who haven’t seen Ben Jordan’s video(s) on infrasound, in the one linked below he highlights the levels of infrasound produced by data centers.
Thanks a bunch for providing this link! I was totally unaware of this infrasound issue.
I’ll put my own biases up front. I don’t know enough about the full picture to say that I am “against data centers generally.” In fact, I have doubts that anyone knows enough to make such a broad claim with complete confidence.
To have a truly informed opinion, you would need to understand a long list of complicated issues. And simply saying, “Let China do it,” doesn’t seem like a serious answer either. There is a very real risk that falling behind in this area could have serious consequences for the United States.
The main concerns about data centers seem to fall into three categories: power use, water use, and noise. But I think both supporters and opponents often miss an important question: Can we use the problems created by data centers as an opportunity to solve other problems?
Power use may be the best example.
Data center developers would obviously prefer not to pay for all the problems their projects might cause nearby. Their argument is basically that the benefits to the larger economy and society outweigh the burdens placed on a relatively small number of people. So, in effect: “Leave us alone.”
Opponents can take the opposite position. They may argue that the burdens placed on nearby residents should be enough to stop a project, even if doing so harms the larger economy or society.
But why do we have to choose between those two extremes? Is there a way to balance the benefits and burdens? Simply refusing to build data centers does not seem like a realistic answer.
So, now what?
One thing that has bothered me about the power issue is that our electric grid already has serious weaknesses. You don’t have to be an expert to know that. Before data centers became such a hot topic, problems with the reliability of the power grid were regularly in the news.
Remember the Texas power crisis in 2021? Or what happened in Gary, Indiana, during the past couple of weeks?
If data centers are going to create a huge new demand for electricity—and attract huge amounts of investment—could we use some of that investment to make the electric grid stronger?
That is where this gets interesting.
As my ChatGPT discussion linked below explains, much of the noise from a data center, including infrasound, can be greatly reduced if large-scale power generation is not located at the data center itself.
But the electricity still has to be generated somewhere. Moving the generators away from the data center may simply move the noise problem somewhere else. It becomes the same problem affecting different people. Why should one group of people be expected to carry that burden instead of another?
Water use may also be manageable. Where water consumption is a real concern, closed-loop cooling systems can greatly reduce the need for water.
The power issue may offer an even bigger opportunity.
Interestingly, the ChatGPT discussion suggests that a data center could help make the local electric grid more reliable even without generating its own electricity. A data center could operate as part of a microgrid and use large battery systems for backup power.
Those batteries could potentially help support important local services during a short-term power outage.
So we might think about it this way:
Plan A: The data center normally gets its electricity from the grid.
Plan B: The data center has large battery backups that can also help support critical local services during short-term outages. That could make the community more resilient and might even save lives during an emergency.
Plan C: The data center also has its own power generation for a long-term or catastrophic outage. Except for testing and maintenance, those generators would rarely operate. But if the larger grid went down for an extended period, they could provide significant benefits to the surrounding community.
And maybe we should go one step further.
What if some of that power generation were designed to operate regularly rather than only during an emergency?
Could a data center then become more than a large consumer of local resources? Could the project also produce electricity, strengthen the local grid, and provide enough economic value to compensate the relatively small number of people who bear most of the project’s burdens?
Could we use the enormous amount of money now flowing into data centers to create lasting community wealth?
That seems to me like a more useful discussion than simply asking whether we are “for” or “against” data centers.
Here is a link to my ChatGPT discussion. It starts with the issue of infrasound from power generation and then moves into the larger questions of power generation, microgrids, and grid resilience.
I would be very interested in hearing what others think.