 |
| Chris Carreiro |
The rapid growth of data centers across the United States is attracting attention and significant community concern. How much electricity will these facilities require and what will be the impact on consumers’ utility bills? How much water will they use? What will they mean for local power grids, land use and surrounding communities?
These are all legitimate questions. Policymakers in many states are actively working on this issue even as statewide moratoriums on data center construction continue to pop-up across the U.S.
The International Energy Agency estimates that data centers consumed around 485 terawatt-hours of electricity globally in 2025 and expects that figure to roughly double by 2030, with the growth of artificial intelligence changing the scale of the challenge.
Aside from the heat of the debate, what does all that computing infrastructure actually do?
 |
Most of us rarely think about the physical infrastructure behind the internet. We stream movies, check our bank balances, ask AI assistants questions, and store photographs ‘in the cloud,’ expecting everything to be there whenever we need it. Even the word ‘cloud’ makes computing sound almost limitless. The reality is very different. Almost every digital interaction depends on physical infrastructure—servers, storage, networking equipment, power and cooling.
Take an ordinary morning. You might check the weather on your phone, read your email, look at traffic conditions, check whether a flight is delayed, reorder a prescription, transfer money between bank accounts or see when an online order will arrive.
Each of these interactions ultimately depends on physical computing infrastructure, most likely in a data center.
The same is true at a much more consequential level. Hospitals depend on digital systems to access patient information. Banks process millions of transactions. Manufacturers increasingly rely on connected systems to manage production. Governments deliver public services digitally. Businesses depend on applications, communications platforms and data to operate.
Behind all of this are physical servers, storage systems and networks. Power keeps those systems running, while cooling technology manages the heat that servers generate. AI is adding significantly to that demand, with training and running AI models requiring substantial computing power, often far more than traditional workloads. As AI becomes embedded in more of the services people and businesses use every day, the infrastructure required to support it will grow too. As a result, the physical infrastructure behind our digital lives is becoming more important, but also more resource-intensive.
Getting more from the infrastructure we already have
That growing demand brings the environmental and community impact of data centers into sharper focus. The question shouldn't simply be whether society needs more computing infrastructure. Our increasing reliance on digital services suggests that it does. The more useful question is how we provide the computing capacity we need as efficiently and responsibly as possible.
Part of the answer will come from new technology. More efficient processors, advances such as liquid cooling and heat reuse, greater use of renewable energy and improvements in data center design can all help reduce the resources required to deliver growing computing capacity. But innovation is only one side of the equation. There is another opportunity that is easily overlooked and that is to make better use of the infrastructure we already have.
Organizations have traditionally managed technology around relatively predictable replacement cycles. Hardware reaches a particular age, the manufacturer's support period ends, and the equipment is replaced. But as demand for computing capacity grows, organizations should rethink whether age alone is a good reason to replace infrastructure.
The end of a support contract does not necessarily mean the end of an asset's useful life. Often, servers, storage, and networking equipment can continue to meet an organization's performance, reliability, and security requirements well beyond the original support timeline. Keeping suitable equipment in service for longer can help organizations get more value from infrastructure they have already invested in.
That doesn't mean keeping every piece of equipment running indefinitely. Newer hardware can deliver significant improvements in performance and energy efficiency, particularly for intensive AI workloads. The important thing is that the decision is based on what the organization needs and how the infrastructure is functioning, rather than simply its age or the date on a support contract.
Managing infrastructure more intelligently
Better monitoring gives organizations a much clearer picture of how their hardware is performing and whether or not it still has useful life left in it. Predictive analytics can help spot problems before they lead to disruption, while remote monitoring means teams can often identify and resolve issues without needing to be on site.
AI is putting far more computing power into individual server racks, producing levels of heat that many traditional data centers weren't designed to handle. Technologies such as liquid cooling offer a more effective way to manage that heat as computing becomes more powerful and densely packed. These may sound like operational details, but together they point to a much bigger principle.
And that applies beyond the world's largest data center facilities. Computing infrastructure lives in corporate data centers, colocation facilities, offices, factories, hospitals and edge locations. Improving how those assets are monitored, maintained, cooled and ultimately replaced can contribute to a more efficient technology ecosystem.
The responsibility that comes with growing demand
As demand for computing grows, so will the questions around the resources needed to support it. Meeting that demand responsibly isn't just about building new data centers. It means thinking more carefully about how we use technology throughout its lifecycle, from how it is maintained and operated to when it genuinely needs to be upgraded or replaced.
AI is making those choices more urgent. Organizations will need more computing capacity, and new technology will undoubtedly be part of the answer. But the focus should be on making the right investment in the right place, rather than assuming that more, or newer, is always better.
***
Chris Carreiro is Chief Technology Officer for Park Place Technologies, a leading global IT infrastructure services firm headquartered in Cleveland with $1.2 billion in annual revenue and 3,000+ employees.
PR has more data than ever. The challenge now is turning it into clarity, context and better decisions.
For the first time, in-house teams and agencies have the opportunity not to just monitor narratives as they emerge, but to understand how they are forming.
Most conversations about AI search today focus on format: clear headers, simple definitions, easy-to-scan comparisons. That work matters, but it assumes the underlying content is still true. It often isn't.
Agencies are playing catch-up when it comes to drawing up the ground rules for AI. Getting up to speed starts with clearly defining how AI should be used.
Proprietary opinion research is one of the most effective ways PR teams can strengthen brand visibility in AI search because LLMs prioritize original insights supported by hard data over recycled information.
