Why Water Stewardship Belongs in the Data Center Conversation

Key Takeaways

• As data centers use water for cooling and other operations, the industry has an opportunity to prioritize responsible water use, system efficiency and reuse.
• Dissolved oxygen, corrosion, scale, suspended solids and biological growth can reduce heat-transfer efficiency in closed-loop cooling systems. Managing these conditions can help keep the water already in a system performing effectively.
• Dissolved oxygen is a key contributor to closed-loop corrosion. Filtration, dissolved oxygen removal, and no-salt conditioning are potential treatment approaches.
• Third-party testing, corrosion measurements, water-quality testing and real-world case studies can help data center operators evaluate whether a solution delivers measurable results.
• Capturing, conditioning, and reusing water that might otherwise be discharged could help reduce demand for additional freshwater and support a more circular approach to data center water management.
• Communities such as OCP give engineers, owners, contractors, and water experts an opportunity to test ideas, share results and develop better approaches to water management together.

The conversation around data centers is increasingly a conversation about resources. As facilities support greater computing demands, the industry is being challenged to think carefully about the water and energy required to keep critical infrastructure operating.

That makes water stewardship an important part of the discussion.

For Stirling Boston, Director of Sales - EasyWater, participating in the Open Compute Project (OCP) offered a firsthand look at how the industry is approaching these questions. What stood out was not a room full of companies competing to promote their individual technologies, but engineers, owners, contractors and other specialists sharing challenges and considering how existing approaches could be improved.

“It’s really a community of cooperation and collaboration,” Boston said.

That kind of collaboration is particularly valuable when it comes to water. Data centers exist and require resources to operate, which creates an important question for the water industry: How can the water these facilities use be managed more responsibly?

Why does water quality matter in data center cooling?

Water quality matters because corrosion, scale, particles and biological growth can reduce the performance of closed-loop cooling systems and make efficient heat transfer more difficult.

Liquid cooling systems share some fundamental characteristics with closed-loop HVAC systems. Water or glycol circulates through the system to transfer heat, with much of that fluid being continuously recirculated rather than consumed.

As Boston explained, however, calling a system “closed” does not mean nothing changes inside it.

“For the most part, you’re recirculating the same water over and over and over in that system,” he said. “So what happens in there is dissolved oxygen that’s in the water will actually cause corrosion to take place.”

Maintenance, leaks, faulty equipment, vents, and fluid replacement can all affect conditions within a closed loop. In the systems discussed during his presentation, Boston said dissolved oxygen concentrations can typically fall around 3 to 7 parts per million.

Corrosion is only part of the concern. Scale, particles, and biofilm can also affect the ability of the system to transfer heat efficiently.

“If we have scaling, if we have biofilms, that’s things growing in the water that stick to the walls of the pipe,” Boston said. “Any of that will inhibit heat transfer from taking place.”

From a stewardship perspective, that creates an opportunity to look beyond simply how much water a facility uses and consider how well the water already within its systems is being maintained.

How can closed-loop water be managed more effectively?

Closed-loop water can be managed more effectively by addressing the conditions that contribute to corrosion, fouling, scale, and biological growth rather than relying on a single treatment method.

Traditional HVAC systems have commonly used chemical treatment to slow corrosion. But Boston argues that managing the source of the problem, particularly dissolved oxygen, deserves greater attention.

“The real root problem, however, is the dissolved oxygen,” he said. “That’s what we really need to attack. If we can get rid of the dissolved oxygen, we can stop corrosion from taking place.”

During OCP, Boston presented one potential approach that combines submicron filtration, dissolved oxygen removal and no-salt conditioning.

Separate media within the system address contaminants and dissolved oxygen, while the conditioning technology is intended to help prevent hardness minerals from forming scale.

The specific technology is only one part of the larger conversation. The concept raises an industry-level question about whether lessons learned through decades of HVAC water treatment can help inform the way water is managed in newer liquid-cooling applications.

“We have a tremendous amount of experience on the HVAC side,” he said. “But the data centers, the hyperscalers, if you think about them, they’re moving so fast that they cannot afford to slow down and say, ‘Do I want to take a risk putting something in that’s never been tested in our environment?’”

That makes validation especially important.

How can the industry know whether a water-treatment approach is working?

Water-treatment performance should be validated with measurable operating data rather than relying solely on technology claims.

During his presentation, an attendee asked for corrosion data before treatment so it could be compared with the post-treatment results. Boston had the information available through an existing case study, but the exchange reinforced how important accessible data is when discussing new approaches with this audience.

“The bottom line for all of these customers is the more data that we put into this, the more believable our story is,” he said.

That expectation extends beyond manufacturer testing. According to Boston, data center stakeholders want third-party testing, case studies, and other independent evidence demonstrating that a technology works within their operating environment.

There are several ways to build that evidence. In his presentation, Boston identified dissolved oxygen testing, corrosion coupon sampling, bacterial testing, total suspended solids, and turbidity as methods for evaluating water conditions and treatment performance.

Technologies that promise more responsible or efficient water management need measurable results behind them.

Why is industry collaboration important to better water stewardship?

Industry collaboration can improve water stewardship by bringing different disciplines together to test ideas, challenge assumptions, and establish better practices based on real-world results.

“There were a lot of panel sessions where you would have, let’s say, an engineer, maybe a biologist, a data center owner. and a contractor,” he said. “And they would all get up there and talk about one specific thing with the mindset of ‘how do we work together to get to the next level?’”

From there, participants might propose a different approach, test it, and return to a future gathering to explain what worked, what did not and what they would change.

This level of collaboration is particularly relevant to water stewardship because there is unlikely to be one technology that addresses every water-related challenge.

Instead, progress requires questions. Where is water being used? Where is it being lost? How long can water remain useful within a system? What effect does water quality have on overall performance? What treatment methods work, and what data proves it?

Can data centers do more to reuse water?

Water reuse could provide an opportunity to reduce waste by treating water as a resource that may serve more than one purpose before it is discharged.

This is an area where Boston believes the broader conversation should go next.

One example is the water used to flush mechanical systems before they are brought online. Boston referenced an estimate shared by a mechanical contractor that a data center project can use approximately 300,000 to 500,000 gallons of water during the system-cleaning process. He sees an opportunity to ask whether water from applications like this could be reconditioned rather than simply discarded.

“How do we help them with reconditioning the water so they’re not wasting water?” he said. “How do we reclaim water, like rainwater harvesting off of these huge facilities’ roofs? We can take that water and put it into the system.”

His larger point is about creating a more circular approach to water.

“We can turn that back into clean water by conditioning it, and we create a cycle where that water gets used over and over,” Boston said.

The goal is not simply finding enough water to support an application. It is understanding whether water can be maintained longer, reused, recovered, or replaced with an alternative source before additional freshwater is required.

What role should the water industry play in the future of data centers?

The water industry should help ensure that responsible water use remains part of the technical conversation as data center infrastructure evolves.

That does not require taking a position for or against the continued expansion of data centers. It means recognizing that these facilities exist and that wherever significant water use occurs, water professionals have a role to play in helping determine how that resource can be managed more responsibly.

The open, collaborative environment Boston encountered at OCP provides one model for accomplishing that. Ideas can be brought forward. Engineers and operators can challenge them. Solutions can be tested. The resulting data can then be shared so the industry can decide what deserves to become a best practice.

“They would put out a hypothesis: ‘Here’s how we think we would do it different next time,’” he said of the OCP discussions. “And then when we get together in another year, we’re going to come back and…tell you if our hypothesis was right, what we learned, what we’re going to do different again on the next one.”

As data center technologies continue to develop, water stewardship should develop alongside them. That means questioning established practices, applying lessons from other applications, validating new approaches with data, and continually looking for ways to use less water, waste less water, and get more value from the water already being used.

About the Author

Stirling is a leadership executive with 30+ years of experience in the heating, cooling, plumbing, and water quality industries. His background started as a manufactures' representative and has spent most of his career working with manufacturers developing, marketing, and providing solutions to customers that create comfort, energy efficiency, and long life mechanical and plumbing systems. Stirling hold a master’s degree in business from Dallas Baptist University and is Director of Sales – EasyWater – A Watts Brand.