DIGITAL INSIGHTS: Real Estate or Infrastructure?
Viewed through the lens of real estate, we get a two-dimensional look at data centers. Viewing them as infrastructure shifts the perspective to three dimensions, leading to better discussions.
By Martin Davis
EDITOR-IN-CHIEF
Email Martin
Digital Insights is a weekly feature usually appearing on Thursdays that explores the role of data centers in our region. These columns will focus on four areas: tracking the development of data centers in our area, exploring projected and actual tax revenue trends, explaining what data centers are and how they affect our daily lives, and reporting on research and emerging trends in the industry. These columns are made possible, in part, by a grant from Stack Infrastructure.
For those old enough to remember, there was something revolutionary about sending an email for the first time. It wasn’t just that one could instantly communicate a message without having to find a postal drop box and then wait a couple days for a response by phone or return letter.
The whole process was so “green.” No need to cut trees to make paper. No burning fossil fuels in postal trucks to run letters from one place to another.
That feeling was intensified when electric vehicles (EVs) began gaining popularity.
With few moving parts, EVs don’t require motor oil or gasoline, and they don’t spew exhaust that leads to climate change. They represented green solutions to the problems of greenhouse gases — and they’re a heck of a lot of fun to drive.
However, with time we’ve come to appreciate that neither emails nor EVs are perfectly “green.”
Most people are aware that while emails may save on trees and fossil fuels, they depend upon data centers to function. And people are becoming more aware of the resources — power and water — that it takes to drive them.
What may be less clear is that EVs aren’t particularly green, either.
While EVs don’t directly produce carbon emissions that worsen climate change, they are adding to the strain on the electric grid. Currently, only about 10% of new car sales are EVs, but projections are that this will continue to rise. (Sales have dropped slightly in early 2026 owing to the rollback of federal tax credits, but market watchers remain bullish on EVs.) The expectation is that about 2% of grid energy will be required to power EVs by 2030. The percentage is substantially higher in Europe and China.
If that 2% figure doesn’t sound significant, consider that it only takes into account what’s required to charge an EV. The future of EVs, however, is not simply finding more power to charge electric vehicles that people drive.
The future is in automated driving.
Self-driving vehicles are gaining traction across the country and show no signs of slowing. The potential for automated vehicles to reduce deaths due to accidents, speed up travel by allowing computers to communicate with cars and traffic lights to maximize drive time while minimizing stopping time, and make travel more convenient and accessible in general is the great upside of the EV market.
Autonomous driving doesn’t work, however, without data centers. Self-driving automation requires that vehicles be able to “talk” to one another — those communications are managed through data centers. (It’s worth noting that non-EVs also depend on data centers, which make over-the-air updates to key computer components possible.)
Knowing this changes the calculus for how we think about data centers and the technology they drive.
Moving from Two- to Three-dimensional Thinking
Over the course of the past two years, the debate over data centers has turned hot, with increasing numbers of communities pushing to keep them out, or significantly reducing their footprint.
The concerns that communities are raising are not without merit — data centers do consume enormous amounts of energy (estimated to reach about 10% of all U.S. energy production by 2030) and depend on water for cooling. There are also concerns about building heights and the impact it has on a community’s aesthetic.
All of these concerns, however, are born from looking at data centers through a real estate mindset. How much land will the buildings take? How many raw resources will be required? And what is the benefit to the community, which is often discussed in terms of tax revenues.
This is a two-dimensional understanding of a three-dimensional reality.
The drivers of data centers — energy and cooling — aren’t simply lost resources. They are infrastructure, driving something tangible and potentially beneficial to communities. Looked at this way, data centers are part of a more-complex three-dimensional reality.
To understand how, let’s return to EVs. In addition to the benefits already mentioned, EVs are being looked to as a way to strengthen the electric grid.
Through a program called Vehicles to Grid Integration, EVs can actually sell excess energy back to the grid by staying connected throughout the day.
The idea behind V2G was created to solve the problem of electric vehicles overstressing the grid system. Unlike data centers, which place constant demands on the grid, EVs can be charged at any time. People tend to charge in the evening, however, after work. When the majority of cars are being charged at the same time, it puts a concentrated pressure on the grid.
V2G spreads out that demand, lessening grid strain. But it does more.
As described by Ideal Energy, “The vehicle-to-grid … concept is a system in which ‘gridable’ electric vehicles interact with the electric grid in more sophisticated ways than just charging. V2G systems could charge intelligently at times of low cost and low demand, provide ancillary grid services like load balancing and frequency regulation, and offer vehicle owners emergency backup power or even a source of income.”
The catch, however, is that V2G depends on data centers to manage the complex interactions between EVs and the grid. In short, without the infrastructure, many of the potential benefits of EVs — automated driving, selling energy back to the grid — are not attainable.
This is one example that highlights the difference between thinking of data centers in terms of real estate vs. infrastructure.
It’s far from the only one.
While data centers rightly receive attention for their water consumption, agriculture is considerably thirstier. According to UNESCO, agriculture consumes about 70% of the world’s freshwater withdrawals.
The application of artificial intelligence to agriculture, however, has the potential to significantly alter that.
The blog ForwardFooding explains the varied ways that artificial intelligence is working to making farming more efficient.
Here’s a look at some of the key technologies that make it possible:
IoT-Based Systems: Internet of Things (IoT) sensors collect data on weather, soil, and air quality, sending it to a cloud platform for analysis. The system then adjusts water flow and irrigation schedules in real-time, maximizing efficiency and reducing waste.
Artificial Intelligence (AI) or Machine Learning (ML)-Powered Solutions: Analyze historical and real-time data to predict water needs based on weather patterns, soil moisture, and crop requirements.
Sensor Technologies: Monitor soil moisture, temperature, and nutrient levels, ensuring that crops get exactly what they need.
Precision Irrigation Systems: Deliver water directly to the roots, reducing evaporation and ensuring minimal waste.
Weather-Based Controls: Use local weather data to adjust watering schedules so plants get just the right amount of moisture—even on a cloudy day.
Widespread adoption of these technologies is just now starting to take hold. The potential to not only reduce water use, but nutrient use as well, while producing greater yields is significant.
Finding Balance
None of this is to say that communities shouldn’t think about data centers as real estate. It is to say that communities need to also consider the infrastructure side of the equation.
What data centers take in terms of land, energy, and water isn’t a simple net loss. This infrastructure is what makes possible solutions that offset those loses in other areas — transportation and agriculture to name but two.
It’s an example of how a loss on one side of the ledger produces gains on the other side — the question becomes are the plusses enough to justify the losses?
As with any emerging technology, it will take time to get a handle on this. After all, sometimes things go sideways.
As mentioned above, moves by the Trump Administration have temporarily slowed EV adoption in the US, though it is surging across most of the rest of the globe.
This has put builders of batteries for EVs in a bind. A report recently out in the New York Times, however, talks about how companies are pivoting to make large batteries for data centers and utilities.
Batteries power electric cars, eliminating the need for a gas tank. But they can also be mounted on the wall of a home with rooftop solar panels, storing energy when the sun is shining and releasing it at night.
At a larger scale, shipping-container-sized batteries do something similar for power companies, which charge batteries during the day when electricity is cheap and discharge them onto the grid when prices spike in the evening.
In short, even if the potential upside to EVs proves to be a bust — unlikely, but one can never rule that out — it may lead to better ways to power data centers that result in reductions to traditional energy sources.
Such is the nature of disruptive technologies.
Thinking about data centers in terms of both their real estate component and their infrastructure component provides a more-balanced understanding of what data centers cost, and what they make possible.
It can also take some of the heat out of data center discussions and help communities better balance the tensions between data centers as real estate and as infrastructure.
Local Obituaries
To view local obituaries or to send a note to family and loved ones, please visit the link that follows.
Support Award-winning, Locally Focused Journalism
The FXBG Advance cuts through the talking points to deliver both incisive and informative news about the issues, people, and organizations that daily affect your life. And we do it in a multi-partisan format that has no equal in this region. Over the past year, our reporting was:
First to break the story of Stafford Board of Supervisors dismissing a citizen library board member for “misconduct,” without informing the citizen or explaining what the person allegedly did wrong.
First to explain falling water levels in the Rappahannock Canal.
First to detail controversial traffic numbers submitted by Stafford staff on the Buc-ee’s project
Our media group also offers the most-extensive election coverage in the region and regular columnists like:
And our newsroom is led by the most-experienced and most-awarded journalists in the region — Adele Uphaus (Managing Editor and multiple VPA award-winner) and Martin Davis (Editor-in-Chief, 2022 Opinion Writer of the Year in Virginia and more than 25 years reporting from around the country and the world).
For just $8 a month, you can help support top-flight journalism that puts people over policies.
Your contributions 100% support our journalists.
Help us as we continue to grow!
This article is published under Creative Commons license CC BY-NC-ND. It can be distributed for noncommercial purposes and must include the following: “Published with permission by FXBG Advance.”
Martin is right that our data center fights are stuck in “ones and twos” — project-by-project battles over traffic, noise, and tax breaks — while the real issue is that we have no shared standard for how to manage impacts up front.
My argument is that Virginia doesn’t need more studies on data centers. We need a Project Manager Guide that every data center developer has to use from day one. That guide would spell out, in plain language, what must be analyzed during design: water demand and local supply, grid impacts and who pays for upgrades, noise and air impacts on neighbors, realistic tax benefits minus infrastructure costs, and the cumulative burden of multiple sites. It would also lock in early community engagement — sharing the analysis at least a year before approval, presenting design options, and committing to monitor and fix problems if impacts exceed projections.
But a guide written by each company isn’t enough. Every community can’t become an expert in global data center practices just to check a developer’s work. So I propose a Virginia Data Center Standards Council — a mixed public‑private body (state agencies, utilities, universities, local governments, industry, and community groups) that maintains a single statewide standard and issues a simple “standards met / not met” opinion on each project’s impact analysis, posted online.
Localities would still decide yes or no on land use. The council wouldn’t approve projects; it would give everyone a common yardstick and a neutral-ish referee.
In commenting on Martin’s piece, you could say: we should keep his “real estate vs. infrastructure” lens — but pair it with a concrete process reform. A statewide Project Manager Guide, backed by a trusted standards council, is how we move from ad‑hoc outrage to predictable, transparent, three‑dimensional decision‑making.
From this morning's email from the Virginia Conservation League Call to Action Rapid Response Team:
Virginia shelled out more than $1.6 billion in tax breaks last year to energy-hungry DATA CENTERS, and what have we gotten in return? Dirtier air, higher energy bills, and out-of-control construction in communities across the state.
Not only are we subsidizing the wealthiest industry in the world, but we’re also undermining our clean energy transition in the process, as these facilities demand more and more energy and make it harder and harder to meet our state’s clean energy laws.
Act Now: Tell Your Lawmaker to Stop Tax Breaks for DIRTY DATA CENTERS
If any subsidies continue, they should only go to DATA CENTERS willing to be good partners in our clean energy transition and to commit to industry-leading energy efficiency and environmental stewardship benchmarks, and those that aren't inundating communities with pollution from dirty, backup generators or onsite fossil fuel infrastructure.
House Bill 897 would go a long way in securing a more sustainable industry, while also protecting our grid, lowering energy demand, and cutting energy bills. Your lawmaker needs to hear from you: Weigh in now and urge them to stop handouts to DIRTY DATA CENTERS and chart a more sustainable path forward!
- Virginia LCV Rapid Response Team