Lost Password

News

Company
Video

Video Tour of Subzero Engineering’s Massive Industry-Leading Facility

Video Overview

Welcome to Subzero Engineering’s groundbreaking facility in the heart of Salt Lake City, Utah—an impressive 155,000 square-foot powerhouse of progress that’s set to redefine the industry. Join us on a journey through our state-of-the-art facility, where innovation meets immersion in our expansive product demo room showcasing cutting-edge data center and cleanroom solutions. Focused on collaboration, our cross-product knowledge-sharing approach brings together teams for comprehensive support on custom projects and new innovations. With a dedicated research and development area, we’re pushing the boundaries of engineering solutions. Our enhanced inventory and materials storage, coupled with a larger workforce, ensure efficient operations and improved customer support. Positioned strategically in Salt Lake City, a central shipping hub expedites deliveries nationwide. Hear from Shane Kilfoil, our President, as he reflects on our journey from a garage startup to this impressive facility, underscoring our commitment to value-added solutions. Join us in pioneering the future of turn-key engineering solutions for critical environments. Visit our website and schedule a tour to witness the Subzero Engineering difference firsthand.

Further highlights include:

  • Tour our expansive product demo room featuring our complete Data Center and Cleanroom product line
  • Learn how cross-product knowledge sharing contributes to the turn key solutions we provide for critical environments
  • Learn about the value of our dedicated Research and Development area
  • See our robust material and inventory storage capabilities
  • Gain confidence in our improved customer support department, ensuring your needs are met promptly and efficiently
  • Details about our centralized shipping hub which guarantees faster delivery times and smoother logistics
Data Center
Product Spotlight

Why Industry Collaboration Matters in the Next Era of Data Center Infrastructure

By Shane Kilfoil, President, Subzero Engineering

The data center industry is moving faster than ever. AI adoption, higher rack densities, hybrid cooling strategies, and increased power demands are reshaping how critical infrastructure is designed, built, and deployed.

Operators are being asked to do more on nearly every front: deploy capacity faster, improve energy efficiency, support more complex cooling environments, and meet increasingly ambitious sustainability goals. No single product or company can solve all of those challenges alone. That is why collaboration across the industry has become so important.

At Subzero Engineering, we have always focused on helping customers understand their airflow, containment, and cooling challenges, then engineering practical solutions that support the specific demands of their environment. For more than two decades, we have worked with data center operators around the world to improve thermal performance, reduce energy consumption, and support scalable infrastructure growth.

As the market evolves, the need for integrated solutions becomes even more important. Containment can no longer be viewed as an isolated component. It must work in coordination with the broader facility design, including ceiling systems, structural elements, airflow pathways, and evolving cooling strategies.

That is what makes our strategic alliance with USG Mission Critical Solutions an important step forward.

USG brings deep expertise in building systems, construction materials, and large-scale project execution. Subzero Engineering brings specialized experience in data center containment, airflow optimization, and critical environment infrastructure. Together, we can help customers approach containment and structural ceiling integration in a more coordinated, efficient, and scalable way.

This type of collaboration is especially important as AI continues to influence data center design. Higher density environments, direct-to-chip liquid cooling, hybrid cooling architectures, and faster deployment schedules all require solutions that are flexible, practical, and engineered for real-world application.

By combining areas of expertise, partnerships like this help simplify procurement, improve coordination, and support faster movement from design to deployment. Customers benefit from a more complete solution ecosystem, while each company continues to focus on the specialized knowledge it brings to the industry.

There is also a sustainability benefit. Effective containment remains one of the most impactful ways to improve cooling efficiency and reduce unnecessary energy consumption in the white space. When containment is thoughtfully integrated into the larger building system, it can help reduce cooling demand, improve operational performance, and support long-term environmental goals.

For Subzero Engineering, this alliance reflects a broader belief: the future of data center infrastructure will be shaped by collaboration. As the demands placed on data centers continue to grow, the industry will need companies willing to combine knowledge, solve problems together, and deliver practical solutions at scale.

Our partnership with USG is one more step in that direction.

Data Center
Press ReleaseProduct Spotlight

Subzero Engineering Announces Strategic Alliance with USG to Deliver Integrated Data Center Containment Solutions

New alliance combines Subzero Engineering’s industry-proven containment technology with USG Mission Critical Solutions’ building systems expertise to simplify data center design and construction.

SALT LAKE CITY, UT — November 1, 2026 — Subzero Engineering, a global leader in turnkey data center containment and mission-critical infrastructure solutions, announces a strategic alliance with USG Mission Critical Solutions, a division of leading building materials manufacturer USG Corporation.

Effective immediately, the alliance introduces USG DataConstrux™ Containment powered by Subzero Engineering, a portfolio of modular aisle containment solutions designed to deliver speed, performance, and scalability for modern data center environments.

As demand for data center infrastructure accelerates, operators, contractors, and developers face increasing pressure to deploy high-performance facilities quickly and efficiently. The alliance brings together Subzero Engineering’s specialized containment systems, engineering expertise, and installation capabilities with USG’s building materials solutions and technical resources.
Together, the companies aim to simplify the design-to-construction process by offering integrated systems that address the evolving requirements of modern data centers.

Combining Proven Containment Technology with Integrated Building Solutions

Data center containment requires the coordination of multiple systems and components, including doors, panels, structural ceilings, framing, and HVAC infrastructure. Traditionally, integrating these elements can introduce complexity during specification, procurement, and installation.

Through this alliance, Subzero Engineering and USG Mission Critical Solutions will provide integrated containment solutions, including both ceiling-mounted and floor-mounted systems, to support a range of data center applications.

From traditional hot aisle containment to infrastructure designed to accommodate emerging liquid cooling technologies, the collaboration will help customers address increasingly complex cooling and infrastructure requirements.

“We couldn’t be more excited to join forces with USG in this crucial initiative,” said Shane Kilfoil, President of Subzero Engineering. “We’ve heard loud and clear from our customers that they need high-performance materials to build strong containment systems, but they also need support and technical know-how to maximize their success. Together, we can offer everything they’re looking for in one place.”

Simplifying Data Center Design and Deployment

A primary objective of the alliance is to make it easier for customers to specify, source, and deploy integrated containment solutions.

By combining the technical expertise, product capabilities, and resources of both organizations, the alliance seeks to streamline project coordination, reduce installation complexity, and support faster deployment.

“What really excites me about this collaboration is the benefits it will yield for customers,” said Diane Earll, President of the Ceilings Division at USG. “USG Mission Critical Solutions and Subzero Engineering are joining forces at a critical time for our industry, when contractors need data center containment systems built right, built quickly, and built to last.”

Cheryl Han, Head of USG Mission Critical Solutions, emphasized the importance of simplifying project delivery.

“Our main goal in collaborating with the great team at Subzero Engineering is to simplify the design to construction process,” said Han. “Access to the expertise and materials of both our organizations, with a streamlined point of contact, means they can simplify the specification process and reduce installation time and rework.”

Supporting the Next Generation of Data Center Infrastructure

The alliance reflects a shared commitment to helping customers navigate the growing complexity of data center construction.

By combining Subzero Engineering’s specialized containment expertise with USG’s established building systems capabilities, USG DataConstrux™ Containment powered by Subzero Engineering offers an integrated approach to infrastructure designed for efficiency, flexibility, and long-term scalability.

The collaboration reinforces Subzero Engineering’s commitment to delivering precision-engineered solutions that simplify deployment, optimize performance, and support the evolving needs of the global data center industry.

About Subzero Engineering

Subzero Engineering is a global leader in turnkey engineering solutions for data centers and mission-critical environments. Since 2005, the company has specialized in precision-engineered containment systems that optimize airflow, improve energy efficiency, and support scalable, high-performance infrastructure worldwide.

Trusted by leading colocation, web, social media, and enterprise operators, Subzero Engineering delivers rapid-deployment, customized containment solutions backed by global manufacturing capabilities. From design through installation and lifecycle optimization, Subzero Engineering simplifies complexity and delivers measurable performance, reliability, and sustainability outcomes.

Website: www.subzeroeng.com

About USG Corporation

USG Corporation is an industry-leading manufacturer of building products and innovative solutions and the parent company of CGC and USG LATAM. Headquartered in Chicago, USG serves construction markets around the world with wall, ceiling, flooring, sheathing, and roofing products.

Website: www.usg.com

Data Center
Educational Article

Sustainable Data Center Containment: The Hidden Benefits

As AI-driven power and cooling demand puts data center operators under growing environmental scrutiny, this article argues one of the more concrete sustainability wins available to them is going almost entirely untold: the material used to seal or enclose containment systems.

The carbon case is where the disconnect becomes clearest. Sealing containment properly carries real, quantifiable emissions savings at around 333 kg of carbon credit potential over five years per 100,000 sq ft, and this scales significantly across large portfolios. Yet major operators already running facilities sealed with this material have never connected that fact to their sustainability reporting. The savings are happening; they’re just not being counted.

Often decisions that are made by facilities and procurement are made independently to sustainability mandates, even though they may ultimately affect environmental results. While compliance drives adoption in strict-enforcement markets, it might be that it’s the sustainability messaging that has the stronger lead in more environmentally focused ones. 

The material behind Subzero Engineering’s DAB (Dissolvable Air Barrier) panels is a certified, already-installed material with a real environmental impact. We ask, why is nobody reporting it?

Article

Unlocking the Green Story Behind Containment

The growth of AI-driven power and cooling demands is putting data center operators under much harsher scrutiny when it comes to environmental impact. Sustainability mandates mean more questions are being asked of operators’ carbon footprints, PUE gradings, and long-term net-zero commitments. While ambitious green roadmaps, solar power purchase agreements, and liquid cooling systems all check the carbon-reduction box, one of the simplest sustainability wins is already happening in many data halls. Yet the material used to seal and fully enclose either horizontal or vertical containment, a product already installed and doing its job, barely gets mentioned.

A Sealed Story, Left Untold

Hot aisle and cold aisle containment is standard practice in modern data centers. The containment roof panel for cold aisle containment and vertical panel for hot aisle containment, and the material used, is often treated as a minor detail within a much larger cooling strategy.

There is often one detail that goes unaddressed: openings above the cold aisle containment due to a lack of roof system, and openings above the hot aisle due to containment not going up to the plenum ceiling, and it’s been driving up cooling costs since the day the facility opened.

It’s not an oversight or a design flaw. It’s a direct byproduct of adhering to fire code. Sprinkler heads are engineered for the widest possible dispersal pattern to maximize coverage in a fire event and fully sealing or enclosing the top of the cold aisle with a roof system or with vertical panels above the hot aisle and up to the plenum ceiling can take the facility out of compliance with the fire code.

That leaves facility operators with two options:

  • To leave the roof system off of cold aisle containment or to leave at least 24” of containment off the top of the vertical hot aisle containment up to the plenum ceiling, preserving the sprinkler’s coverage pattern at the cost of a fully sealed containment system; or
  • Install sprinkler heads in every single aisle, which solves the dispersal problem but comes with a catch — a layout that made sense at commissioning rarely stays fixed for the life of the facility, and a sprinkler system built around a specific aisle plan doesn’t adapt easily when that plan changes.

Most facilities take the first option. It’s simpler, cheaper, and satisfies fire code, but it also means that either the cold or hot aisle containment system is not fully sealed and contained, allowing the hot air to rise, drift across the gap, and mix back into the cold aisle. This design undoes some of the benefits that the containment system just performed, and the HVAC system compensates by working even harder.

This is a constant, low-grade condition and rarely gets flagged as an issue. Instead, it shows up as elevated energy costs and lower cooling efficiency, spread across the operational life of the facility in a way that’s easy to normalize and tough to trace back to its origin.

However, closing the openings outright isn’t as simple as adding a horizontal roof or vertical panels for the physical containment seal. Doing that just reintroduces the original problem of fire code compliance. If the suppression system can’t detect a fire fast enough because the containment is fully sealed, it becomes a real hazard.

Why DAB Panels Are the Overlooked Asset

DAB panels solve a problem that has limited containment for years. Traditional roof panels can obstruct sprinkler coverage, force relocation of sprinkler heads, or get rejected outright by the local Authority Having Jurisdiction (AHJ). DAB panels are currently the only approved product that seals the horizontal gap over the cold aisle without triggering those concerns. The same is true with hot aisle vertical panels up to the plenum ceiling.

DAB panels that are made from an EPA-certified, plant-based cellulose composite provide a continuous seal across the top of a contained aisle. Cooling equipment runs closer to design conditions instead of continuously compensating for leaked airflow, and panels can be easily cut, configured and fitted on-site around cable trays, pipes, and other overhead obstructions, ideal for both new builds and retrofits.

Paired with a purpose-built passive airflow flap, a small mechanism fitted with a melting fuse plug, DAB panels keep the containment seal closed and intact under normal operating conditions, letting the HVAC system run at full efficiency. If a fire starts and aisle temperature rises, the fuse plugs melt, the flap opens, and hot air is immediately released upward for cold aisle containment and outward for hot aisle containment to the sprinkler activation system. The sprinklers dissolve the DAB panels within 30 seconds, removing the risk of any falling-object hazard that might restrict egress, so response time stays fast and AHJs stay satisfied.

This is a design that fire protection engineers have reviewed and signed off on directly. A fully sealed cold or hot aisle, paired with airflow flaps where needed, meets fire code requirements without compromising fire suppression response times.

Why the Disconnect Exists

DAB panels are being adopted because they solve a fire-safety-versus-efficiency conflict that had nothing to do with ESG reporting. They weren’t adopted because someone was chasing a carbon number. That’s why the sustainability benefit has gone unclaimed. It’s a side effect of a sound engineering decision.

Containment sealing is typically a facilities and procurement decision, evaluated on cooling performance, installation cost, and code compliance, not on its contribution to a corporate carbon disclosure. Sustainability teams are rarely able to voice an opinion on the mechanical and construction details of a build-out. Therefore, two functions inside the same organization end up shaping the same environmental outcome but have no shared line of sight between them.

What’s missing is the connectivity between the facilities team that specified the panels and the sustainability team that reports on emissions.

Closing the Gap

Properly sealed containment delivers real emissions savings: roughly 333 kg of carbon credit potential over five years for every 100,000 square feet sealed. Applied across a hyperscale portfolio spanning millions of square feet, it becomes a line-item that’s worth reporting. But if operators running facilities sealed with DAB panels haven’t linked that fact to their sustainability reports, then the savings are happening but aren’t captured or publicized on carbon credit calculations or customer-facing disclosures.

For an industry under constant pressure to justify progress on net-zero programs, operators are leaving value unclaimed. The material is installed, the certification exists, and the savings are accruing.

Unlocking the potential doesn’t require new capital or a change in engineering standards. It requires operators to:

  • Audit existing containment installations to identify where certified sealing materials are already in place.
  • Quantify the carbon credit potential across the full portfolio, not just per site.
  • Build a reporting bridge between procurement/facilities and sustainability teams, so decisions made for performance reasons are also captured for disclosure.
  • Lead with the business case, not just the compliance case, in markets where sustainability credentials are a competitive differentiator.

The carbon case for fully sealed containment may be the most quantifiable sustainability win sitting inside a data center today, precisely because it’s already been made. The panels are up, and the seals are in place. The only thing left to do is claim its value.

Data Center
Educational Article

Technical and Operational Barriers to True Data Center Sustainability

Even the most well-designed modern data centers experience energy loss in small and often invisible ways that can add up to massive costs over time. Headline initiatives may be about the big strategic shifts, but the real challenge is often in the less glamorous technical details that require ongoing diligence.

Retrofitting legacy facilities with efficient models to upgrade cooling units can significantly reduce energy waste, but it is technically complex and capital intensive. Containment strategy is especially important but relies on accurate airflow design.

One underappreciated hurdle is rack hygiene, the ongoing task of finding and sealing air leaks. While small gaps around rails, floor tiles, and rack openings may not cause immediate damage, over time they add up to significant wasted energy. This is easy to deprioritize, but expensive to ignore.

Physical inefficiencies are compounded by strategic ones. Replacing inefficient hardware requires identifying consolidation candidates and committing migration resources.

None of these barriers are solved with a single upgrade. In this article, we investigate how prioritizing the small things with sustained investment and follow-through, not just at implementation but as routine practice, can help to pave the way to achieving true data center sustainability goals.

The Hidden Costs: Why Data Center Sustainability Fails Without Operational Discipline

Even the most efficiently designed modern facilities experience energy loss in small, often invisible ways. Individually, these losses are insignificant and rarely worth writing in a sustainability report. Headline initiatives may be about the major strategic shifts, but the uncomfortable truth is that data center sustainability is not about the big transformation; it is about the follow-through.

For boards overseeing data center investments, sustainability conversations tend to gravitate to the initiatives that are easier to present and promote: capital projects with clear before-and-after metrics, declarations and commitments to renewable energy, and efficiency targets. But they are not where most of the avoidable costs lie.

The bigger financial exposure is often in the less glamorous, less visible details: the ongoing operational discipline required to keep a data center efficient after it’s been built. Without it, even a well-designed, well-capitalized facility will bleed money, unevidenced as a single line item that anyone can point a finger to.

This is a different kind of risk than a cost overrun or an outage. It doesn’t announce itself. It accumulates. 

Risk Profiles

Capital initiatives: Replacing aging cooling infrastructure and redesigning airflow containment are conspicuous, fundable, and accountable initiatives. These go through a capital approval process, have a defined ROI case, and produce a measurable result. Organization is generally good at executing this kind of work because it fits existing governance structures.

Operational discipline: The continuous, dull work of maintaining what’s already been built does not have an approval gate. There often isn’t an obvious accountable owner, and, without a dramatic failure that forces it onto the agenda, it is most likely to be underfunded and under-governed. But it can represent a comparable share of total energy waste.

Where the Losses Lie

Physical waste: While small gaps around rails, floor tiles, and rack openings may not cause immediate problems, over time they add up to significant wasted energy.

There’s no failing equipment, so no incident report gets filed. However, left unmanaged, this energy waste can compound, forcing cooling systems to work harder for the same output. This is easy to deprioritize, but expensive to ignore.

Strategic waste: Strategic inefficiencies exacerbate the physical ones. Zombie servers draw power without delivering value, and outdated hardware continues to run well past the point where consolidation or replacement would be cheaper. But to avoid these inefficiencies means the component candidates need to be identified and the resources committed – projects often deprioritized amid daily operations.

Sustained Investment

In both cases, the financial exposure isn’t caused by a bad decision. It’s caused by the absence of a decision. A gap that nobody is responsible for closing.

Capital expenditure gets one moment of intense scrutiny, and then it’s done. Operational discipline doesn’t work that way. It’s a continuous process with no single sign-off, no individual decision, and no moment where someone needs to ask, “Is this still efficient?”

One large, visible cost is easy to capture. What’s not easy is to catch a hundred small ones, none of which individually justifies an investigation. The result is a facility that looks the same on paper as the day it was commissioned while performing materially worse in practice.

The Big Issues

1. PUE Decay

A facility commissioned at a PUE of 1.3 rarely stays there. Modern hyperscale facilities are routinely engineered toward a PUE near 1.1–1.2. In 2026, enterprise data centers typically operate between PUE 1.6–2.0 (higher for legacy sites), while modern colocation facilities generally achieve PUE in the 1.4–1.6 range and that gap often isn’t a design flaw; it’s accumulated drift. Common contributors are:

  • Containment Breaches. Missing blanking panels, gaps around cable cutouts, and unsealed floor tiles that let hot and cold air mix. Each individual gap is tiny. Dozens of them, collected over years of moves, adds and changes, are not.
  • Cooling Systems. Reduced airflow and heat transfer efficiency creep in gradually, and the cooling system compensates by working harder rather than failing outright, so there’s no alarm, just a slowly rising kWh.
  • Sensor Drift. Miscalibrated temperature and humidity sensors cause CRAC/CRAH units to overcool, undercool, or fight each other across zones.
  • Orphaned Adjustments. Changes made during an incident (a hot spot, a vendor visit, a one-off load event) that are never reverted once the original condition resolves.

None of these triggers a ticket on its own. They show up as anomalies that are easy to attribute to “weather” or “load” or “just how the building runs.”

2. Stranded Capacity

Operators frequently overprovision power and cooling per rack to be safe, then never revisit allocations as the real draw becomes clear. The result is that the facility reads as fully allocated, i.e., power and cooling spoken for, while the actual utilization sits at 40–60% of that allocation.

Every additional kW of usable capacity therefore ends up costing far more than the original design model assumed. It can also distort expansion decisions, with operators concluding more space might need to be found when the real problem is that existing capacity has never been reclaimed.

3. Keeping the Lights On

Industry surveys have repeatedly found large shares of physical servers are “comatose” – drawing power while doing little or no useful work. An asset goes idle when a project ends or a migration completes, and nobody formally retires it. This results in both wasted power draw and wasted rack space that could otherwise be reclaimed as usable capacity. This is a tracking and decommissioning process problem.

4. Deferred Maintenance Masquerading as Savings

Skipping or stretching maintenance cycles can produce a visible short-term lower maintenance spend but can become an invisible long-term liability. At best, nothing bad happens for years. The deferral looks like a good decision, until a single failure event such as an unplanned outage or an emergency vendor callout presents a loss that dwarfs every dollar saved.

The dangerous part of this pattern is that it’s self-reinforcing. Because the early years look fine, deferred maintenance gets read as evidence the schedule was conservative to begin with, rather than as a risk that hasn’t come due yet.

5. Vendor/Contract Creep

Service contracts, licensing, and managed-service agreements go unreviewed. Costs can escalate via autorenewals, by increased pricing that hasn’t been renegotiated against market rates, or because the subscription is just not needed anymore.

Once again, no single renewal is the problem. It’s the absence of a periodic audit.

The Common Thread

All these issues share the same consequences: small, individually defensible decisions or omissions that are repeated or left unaddressed over time. They amalgamate into a real and substantial cost. This is not a failure of engineering. It is a failure, or absence, of an ongoing process designed to catch drift before it compounds.

This is also why it’s organizationally hard to fix. Capital projects have natural owners and natural review cycles built into how they’re funded. Operational discipline doesn’t. These are metrics that don’t show up until something has already gone quietly wrong for a long time. 

The Bottom Line for the Board

Sustainability commitments are typically judged by the visible initiatives a company announces. The actual cost discipline that ascertains whether those commitments translate into real savings is determined by something much quieter: whether routine operational maintenance is treated as a governed business function or an informal activity.

The most effective oversight the board can provide here isn’t by approving the next capital project. It’s ensuring the organization has clear and durable ownership of the unexciting operational diligence so that any efficiency gains achieved through capital investment aren’t eroded by the absence of follow-through.

The one big question any data center needs to ask itself: is there a named owner for ongoing efficiency, with a recurring review, independent of incident response?

If the answer is no, drift isn’t a risk. It’s the default state.

Data Center
Educational Article

Composite AisleFrame® (CAF): Delivering a Custom Data Center Solution

As data centers are designed to facilitate higher-density IT and HPC environments, operators must rethink their infrastructure to support our ever-evolving industry.

Steel infrastructure, once seen as standard, provided the structural foundation for data center deployments for decades. But as facilities become more complex and focused on sustainability, its weight, installation, transportation requirements, and environmental benefits introduce additional considerations into a new building for operators.

This created an opportunity for Subzero to rethink the aisle infrastructure itself and shone a spotlight on our Composite AisleFrame® (CAF), launched last year as a sustainable frame-based support structure support customers for today and future digital demands.

We now have our first official deployment of CAF with one of the leading colocation data center providers in the northeastern United States, which allowed for their facility to enhance their high performance and reliability at the same time as accomplishing long-term environmental objectives with our expert team.

What you’ll learn from this case study:

  • Designing around client requirements: What happens when a containment idea meets the realities of an existing data center? This new launch showcases the structural and site-specific considerations needed during deployment and how Subzero Engineering’s expert team approached the challenge in designing around an established raised-floor environment without losing sight of the customer’s wider infrastructure requirements.
  • Why our CAF solution was chosen: Choosing the right containment framework involved more than evaluating the aisle itself. Weight, structural support, existing site conditions and the limitations of traditional installation methods all had to be considered.
  • Design to Deployment: A strong strategy still has to work when it reaches the data center floor. Our case study looks into how Subzero Engineering responded as the project moved from initial design to installation, including the challenges our teams encountered along the way and the thinking that helped turn a complex set of requirements into a practical deployment.

To explore the project in more detail, read the full Subzero Engineering case study.

Data Center
Educational Article

Direct Liquid Cooling Optimization: The Containment Advantage

As AI workloads push rack densities beyond 100 kW, Direct Liquid Cooling (DLC) is becoming an increasingly important part of the data center cooling strategy. But liquid cooling doesn’t eliminate the need for effective airflow management.

In this new white paper, Subzero Engineering Senior CFD Engineer Gordon Johnson explores why containment remains a critical foundation for efficient, reliable hybrid cooling—and how operators can prepare their facilities for the next generation of high-density compute.

AI Is Changing the Data Center Cooling Equation

Modern CPUs, GPUs, and AI accelerators are generating thermal loads that increasingly exceed the practical limits of traditional air cooling. As rack densities move beyond the 25–35 kW range and approach or exceed 100 kW, liquid cooling provides a scalable way to remove heat directly from high-power components.

But the transition isn’t simply air cooling vs. liquid cooling.

For many facilities, the future is hybrid cooling, which combines both air and liquid cooling to manage the heat generated by low-density and high-density compute loads.

That makes the management of hot and cold air more important, not less.

Liquid Cooling Doesn’t Eliminate Airflow.

It Changes How You Need to Manage It.

Direct-to-Chip cooling removes heat at the source, but supporting equipment continues to release heat into the surrounding environment. Without containment, bypass airflow, recirculation and unstable inlet temperatures can undermine the performance of an otherwise advanced cooling strategy.

Containment provides the critical bridge between today’s air-cooled infrastructure and tomorrow’s high-density liquid-cooled environments.

Inside the White Paper

Gordon examines the technologies, challenges and infrastructure decisions shaping next-generation data center cooling, including:

  • What’s driving liquid cooling adoption as AI and HPC push power densities higher
  • How hybrid cooling works and why air and liquid cooling increasingly operate together
  • Direct-to-Chip vs. immersion cooling and how the DLC market may evolve
  • Why containment remains non-negotiable in modern thermal management
  • How containment improves thermal reliability and cooling efficiency
  • How AisleFrame® supports AI and HPC environments with scalable, ground-supported infrastructure ready for air and liquid cooling

The Foundation for a Hybrid Cooling Future

Containment physically separates supply and exhaust air streams, helping maximize the effectiveness of both air- and liquid-cooling infrastructure.

For existing facilities, it can optimize today’s air-cooled environment while creating a more scalable path toward future DLC deployment. For new AI and HPC facilities, containment provides a foundational thermal strategy that supports higher densities, predictable inlet temperatures and more efficient cooling performance.

Subzero’s AisleFrame® System takes that strategy further by integrating containment, DTC cooling piping, racks, data and power cabling within a modular, ground-supported structural platform designed to scale as infrastructure requirements evolve.

Data Center
Educational Article

The Green Opportunity in Data Centers

Why Your Data Center’s Green Strategy Might be Missing Half the Equation

Data centers now consume nearly 2% of global energy. As the mainstay of the digital economy, that level of consumption is a problem but also an opportunity at the same time. If you push efficiency forward here, even modestly, the outsized environmental payoff scales accordingly.

So why isn’t it happening faster?

Promise v Performance

Many corporations pledge zero-carbon goals, a target year, and a whole load of optimistic language, only to produce rather underwhelming results. This disparity is rarely about technology; there’s plenty of technology available to enhance efficiencies. So, if technology isn’t the issue, what is it?

It’s a lack of leadership commitment. True sustainability necessitates visible and ongoing commitment from the C-suite through budget cycles and shifting priorities. And that’s where it often breaks down. Green initiatives get deprioritized the moment something else becomes more pressing, or funding dries up before gains materialize.

Design Gets You Started, Not Finished

For those leadership teams that do commit, the natural starting point is design. When building a new facility or retrofitting an old one, Computational Fluid Dynamics (CFD) modeling is a powerful tool. It simulates how air will move, where hot spots will form, and how cooling and containment strategies will perform before a single rack is installed.

But CFD has a blind spot: it’s a prediction, not a guarantee.

Most data centers are designed once and then run for a decade or more. The cooling strategy, airflow pathways, and containment setups are typically locked in during the design phase. They’re modeled carefully, validated, and then largely left to run on the assumption of the model data. But the moment a facility goes live, real life gets in the way. Not all variables are known in a data center before it becomes operational; workloads shift, and racks get added, removed, or reconfigured. Hardware begins to age and begins to run hotter than it did when it was new. Containment seals degrade, and hot aisle and cold aisle separation – perfect on a CFD model – slowly erodes through tiny, unremarkable changes such as a blanking panel left off here and a cable tray rerouted there.

None of these changes show up as an alarm. They develop as gradual inefficiencies, such as a CRAC unit working harder than it should, or a slightly elevated inlet temperature on one rack, or a few extra percentage points on the cooling bill. Individually, they’re invisible.

But the simulation that justified your design decisions on day one has become far less accurate on day 500, even if nothing has visibly changed.

The Case for Continuous Monitoring

Tools like CFD modeling are essential in the planning phase, but the moment real-world variables move – and they always do – the model stops describing reality. Without a way to check actual performance against the original plan, operators are left managing a facility based on an assumption that’s been silently going stale.

This is where continuous monitoring earns its place. Facilities need real-time visibility into what’s happening now: actual airflow, actual cooling performance, and actual containment integrity. This needs to be tracked day to day rather than presumed from a simulation run months or years earlier.

What Real-Time Visibility Actually Provides

A monitoring system built for this purpose should give operators the following:

  • Day-to-day performance data, not periodic snapshots, so that any drift is caught early. It’s already too late when discovered during an annual audit.
  • Granular insight into airflow and containment, down to the rack or row level, where small leaks and inefficiencies originate.
  • A feedback loop, not just a dashboard. Visibility is only useful if it leads to adjustment.

The value isn’t just in detecting problems. It’s in the ability to make ongoing adjustments to keep a facility tuned to its current conditions rather than its original blueprint.

Monitoring isn’t valuable as a passive record. It needs to enable operators to act before those small inefficiencies compound into real energy and cost penalties.

Two Tools, One Strategy

The takeaway here isn’t that CFD is outdated or that monitoring makes design irrelevant. It’s the opposite: no data center can claim to be truly green on the strength of either approach alone. CFD gives you a strong starting position, while continuous monitoring keeps that position reliable and trustworthy.

No data center can be genuinely green without the leadership team pairing strong upfront design with robust, ongoing monitoring. CFD and monitoring aren’t competing approaches — they’re complementary. They help to close the imbalance between what companies promise and what they deliver. The facilities that perform best over time aren’t necessarily the ones with the most sophisticated design. They’re the ones that treat optimization as continuous rather than finished. Real-time visibility turns green thinking from a one-time engineering decision into an operational discipline. It becomes one that adapts as the facility itself changes.

In a sector where even small efficiency gains carry outsized environmental and financial weight, that kind of ongoing responsiveness isn’t a nice-to-have. It’s the difference between a data center that was designed to be efficient and one that stays that way.

And closing that gap is, ultimately, a leadership decision as much as a technical one.

Data Center
Product Spotlight

Subzero Engineering and USG Strategic Partnership: Why do alliances like this matter to the digital Infrastructure?

From Shane Kilfoil, President, Subzero Engineering

The data center industry is rapidly evolving, driven by AI adoption, power demands and innovative cooling strategies coming into the market. Operators are being asked to deploy capacity faster, support higher rack densities, and meet increasingly ambitious sustainability targets, all whilst managing cost and growing complexity. But collaboration is vital to success.

Subzero Engineering’s latest strategic alliance with USG Mission Critical Solutions represents a
transformative step forward that strengthens the ability to serve customers and support the
next generation of data center infrastructure.

This New Alliance

At Subzero Engineering, our focus has always been on the foundation of understanding what
the customers are trying to achieve and engineering containment solutions that help them
get there. For more than two decades, we have specialized in containment systems that
support mission-critical environments around the world.

USG brings a complementary set of strengths. As one of North America’s most established
building materials manufacturers, USG provides expertise in construction materials, project
execution, and large-scale market reach.

This collaboration ensures delivery in a more integrated approach that simplifies
procurement and deployment, strengthening the ability to deliver containment solutions at
scale and positions both organizations to support the evolving needs of AI-driven
infrastructure.

Meeting the Demands of the AI Era

AI is reshaping data center design, with higher rack densities, hybrid cooling architectures,
direct-to-chip liquid cooling, and accelerated deployment schedules all requiring
containment solutions engineered for flexibility and scale.

As the industry evolves towards more hybrid cooling environments, industry collaboration is
vital. Technological advancements happen best when organizations combine expertise and
experience to develop practical solutions that can be deployed at scale.

Subzero’s alliance with USG strengthens its capabilities. By combining structural building
expertise with containment engineering, customers can move quickly from design to
deployment, accelerate innovation, improve deployment methodologies, and respond faster
to evolving infrastructure requirements.

Driving Sustainable Growth Through Partnership

Strategic partnerships can expand reach while remaining focused on the engineering
expertise that differentiates us. Customers benefit from a broader solution ecosystem while
Subzero continues advancing containment technologies, airflow optimization, and critical
environment infrastructure.

The partnership also supports a shared commitment to sustainability. Containment remains
one of the most effective strategies for improving cooling efficiency and reducing energy
consumption, and with better airflow management, it lowers cooling demand and
environmental impact.

By bringing together USG’s expertise in building systems and Subzero Engineering’s
containment capabilities, we can help customers implement integrated solutions that
support both operational performance and environmental responsibility.
Looking to the Future

For Subzero Engineering, this alliance is a natural extension of our commitment to helping
customers build more efficient, scalable, and sustainable critical environments. The future of
data center infrastructure will be shaped by collaboration. This is one more step in that
direction.

Data Center
Product Spotlight

Full vs. Partial Containment: How DAB Panels Improve Efficiency and ROI in the Data Center

Data centers consume tremendous amounts of energy, yet a surprising proportion of cooling capacity is squandered on air that never reaches the equipment it’s designed to cool. As a result, managing airflow presents both a monetary and thermal problem.

Full aisle containment has long been the gold standard in data center thermal management: isolating hot and cold airstreams that reduces cooling demand and lowers PUE values. Unfortunately, the real world gets in the way of things.

Many operators have felt compelled to use partial containment systems due to fire regulations, sprinkler head placement, and the physical constraints of existing facilities. Although such configurations provide some of the positive effects of full containment, they leave an efficiency gap that is reflected in higher energy consumption and operating costs.

A solution to that trade-off is provided by Dissolvable Air Barrier (DAB) panels. In locations where full containment was previously impractical, DAB panels can completely seal the aisle without moving the overhead fire safety sprinkler systems. This has significant effects on ROI, energy consumption, CapEx and OpEx.

Why Containment Matters

Efficient data center cooling is essentially about separating hot and cold air. When hot and cold air mix, if the supply air arriving at the server inlets is warmer than the intended temperature, the cooling systems need to work harder to compensate, wasting energy.

That waste is significant. Studies conducted by ASHRAE and the Green Grid consistently show that
uncontained or partially contained data halls are known to operate at significantly higher PUE values than facilities with full aisle containment.

Containment creates a physical barrier between supply and return air paths. Hot/cold air mixing is
eliminated by physically enclosing the end-of-aisles with doors, and over-rack roof systems or over-rack containment up to the ceiling. The tighter the barrier, the less mixing occurs, and the greater the efficiency gain.

Partial containment will always result in some supply and return air mixing, higher than necessary cooling demand, and lower cooling efficiency than full containment could provide. In retrofits, the targeted sealing of the most problematic airflow gaps is often considered. This can include the use of curtains, end-of-row doors, and blanking panels.

However, even full containment often faces a major implementation obstacle: compliance. Most fire codes insist that sprinkler heads must be placed within a certain distance of the ceiling, and many containment structures, like overhead roof panels, obstruct this relationship. In addition, over-rack containment up to the ceiling may also not be possible if sprinkler heads are not correctly located in both the cold and hot aisle after containment. Relocating sprinklers to fit into a containment structure can be complex, expensive, disruptive, and, in some facilities, practically impossible without a full infrastructure overhaul.

The result has been a divided market: facilities that can achieve full containment do, and facilities that are unable to do so make do with partial containment solutions that reduce, but do not eliminate, airflow mixing.

DAB panels change that calculation.

The Role of DAB Panels

DAB panels are purpose-engineered vertical over-rack panels or horizontal roof panels designed to meet fire code requirements by completely sealing the top of a contained aisle without obstructing the sprinkler system. The key difference is the material from which they are constructed. DAB panels are composed of a unique cellulose-based composition that, when exposed to sprinkler water, dissolves completely in less than 30 seconds.

They function as a solid, continuous air barrier that seals the aisle as effectively as any rigid containment panel, but in the event of a fire suppression activation, DAB panels effectively disappear. They clear the way for the sprinklers to do their job, thus maintaining full compliance with fire code requirements.

Maintaining the thermal isolation between hot and cool aisles while adhering to fire regulations has grown more difficult as data center fire safety compliance codes change. DAB panels meet these requirements, by offering containment that functions within current fire suppression systems without posing the safety risks associated with falling panels.

Calculating the ROI of DAB Panels

The ROI calculation also considers the installation footprint. If sprinkler relocation is avoided, the capital cost of attaining complete containment can be significantly decreased. DAB panels are installed inside the aisle seals and the current containment structure. The financial argument for DAB panels operates on two levels.

First, CapEx reduction comes from avoiding sprinkler relocation. A full containment retrofit that requires sprinkler relocation can add significant cost and project complexity – in some facilities, it has been the reason full containment projects have been put on hold indefinitely. DAB panels remove that cost.

Secondly, OpEx reduction follows from the efficiency gains. Reduced cooling demand results in lower electricity use, which directly lowers monthly operational costs. These potential annual savings from switching from partial to full containment are considerable, particularly in those facilities with high rack density or substantial cooling infrastructure expenditure.

Design and Implementation

Instead of requiring modifications to current fire suppression systems or changing the physical structure of the facility, DAB panel installation follows the same process as other containment systems. For horizontal roof containment (typically CAC deployments), DAB panels are fitted across the top of the contained aisle to create a continuous seal. For vertical over-rack containment (typically HAC deployments), DAB panels are easily fitted up to the ceiling of the contained aisle to again create a continuous seal.

Ongoing maintenance considerations are minimal. The maintenance profile of the panels is like that of other passive containment components, since they are not moving elements and can be easily visually inspected to eliminate gaps and openings that result in unnecessary containment leakage.

Efficiency Without Compromise

The choice between full and partial containment has never really been a question about performance. Operators have always recognized that full containment is more effective. The question has always been whether full containment was possible under practical limitations.

DAB panels provide the affirmative. Facilities that have previously been restricted to the limitations of partial containment designs can now reach the efficiency ceiling of full containment by completely closing the aisle while still adhering to fire suppression regulations. Better thermal performance, effective energy use, lower capital and operating expenses, and a quicker return on investment are the outcomes, without forcing operators to choose between efficiency and safety.

For data center operators weighing their next infrastructure investment, the question is no longer whether full containment is possible. The question is how quickly it can be implemented.

Data Center
Product Spotlight

The Future of Data Center Cooling: Insights from Gordon Johnson, Senior CFD Manager at Subzero Engineering

An Interview with Gordon Johnson, Senior CFD Manager at Subzero Engineering

As artificial intelligence, high-performance computing, and increasingly dense IT environments continue to reshape the data center industry, operators are facing new challenges in cooling, airflow management, efficiency, and long-term sustainability. Technologies that were once considered emerging are now becoming essential components of modern infrastructure strategies.

To help make sense of these changes, we sat down with Gordon Johnson, Senior CFD Manager at Subzero Engineering. With more than three decades of experience across data centers and cleanroom environments, Gordon has helped organizations around the world optimize cooling performance, improve energy efficiency, and prepare their facilities for the next generation of computing demands.

In his role at Subzero Engineering, Gordon leads Computational Fluid Dynamics (CFD) modeling projects across North America, EMEA, and APAC while providing training and guidance on containment, cooling strategies, airflow management, and data center best practices. He is also a frequent industry speaker, sharing insights on the technologies and trends shaping the future of mission-critical facilities.

In this interview, Gordon discusses the evolution of data center cooling, the growing importance of hybrid air and liquid cooling strategies, common misconceptions surrounding airflow management, and why cooling has become one of the most strategic infrastructure decisions operators face today.

Whether you’re designing a new facility, planning for AI workloads, or looking to improve the efficiency of an existing data center, Gordon’s insights offer valuable perspective on where the industry is headed—and how organizations can prepare for what comes next.

What is your role today at Subzero Engineering, and how did your journey into the data centre industry begin?

As Senior CFD Manager, I manage all of our CFD modeling projects across the US, EMEA, and APAC regions, as well as deliver our internal and external training on data center design, containment, cooling strategies, and CFD best practices. My role focuses on improving efficiency and operational performance in legacy and next-generation data centres. In addition, I work closely with our global sales and engineering teams to ensure customers receive accurate, highperformance containment solutions that are tailored to their facilities.

Also, I regularly speak at industry events on data center cooling, containment, and emerging trends shaping the future of our industry.

You’ve spent more than 30 years working across data centres and cleanroom environments. What changes have had the biggest impact on how facilities are designed and operated today?

When I initially entered the industry, efficiency and energy savings were largely seen as an afterthought in data center management and operation. As recently as 10–15 years ago, the prevailing mindset for data center cooling was simply ‘the colder, the better.’ Beyond driving massive energy bills, this approach wasted natural resources and expanded our carbon footprint.

Fortunately, the industry has since shifted its perspective, with many facilities now operating supply temperatures near the upper end of ASHRAE’s recommended serverinlet guidelines, ensuring efficiency and a closer connection to the industry’s ambition for a more sustainable future.

Looking back across your career, what have been some of the most important lessons you’ve learned about designing resilient infrastructure?

Looking back on my career in the data center industry, one of the most important lessons I’ve learned is that simplicity is often what drives reliability and performance.

Data centers constantly evolve, and the demand for high-performance computing densities are growing, so those in the industry must design their facilities to adapt to changes without requiring complete redesigns. Containment is a perfect example of simplicity at play, which is why I’m so passionate about the topic. Rather than consuming energy, effective containment reduces energy use and makes every other efficiency initiative more impactful and costeffective. It reinforces a principle I’ve come to fully embrace – the best energy saved is the energy you never consume in the first place.

As AI workloads push rack densities higher, how is the industry’s approach to cooling and airflow evolving?

As recently as 2020, a 20 kW rack qualified as HPC (High Performance Computing). Today, AI and machine learning workloads routinely drive rack densities to 100 kW and beyond, with demand continuing to climb. This rapid escalation is a key reason hybrid cooling, combining air and liquid systems to handle both low and highdensity racks, has become the industry’s evolving standard for efficient cooling regardless of the workloads deployed.

The industry is recognising that future-ready data centres must combine liquid cooling with highly disciplined airflow management systems to support the evolving needs of the competitive market.

There’s growing conversation around liquid cooling, immersion, and direct-to-chip technologies. How do you see these approaches fitting together within future data centre environments?

Future data centre environments must consider using a hybridised approach into their cooling technologies, so I expect to see this in next generation environments. It will be common to see directtochip, immersion, and air-cooling solutions operating side by side within the same facility depending on the density requirements, operational priorities, and sustainability goals. What is also important to note is seeing a more deliberate separation between highdensity and lowdensity ITE, which is ultimately how operators will maximize efficiency.

What do you think operators still misunderstand about airflow management and thermal efficiency in modern facilities?

Some people don’t realize that the most common form of liquid cooling—directtochip (DTC) — still relies heavily on air cooling strategies. DTC cold plates remove heat only from the highestload components such as CPUs and GPUs. The remaining server and rack heat must still be removed by traditional air systems. Since roughly 75% of the heat is captured by the liquid loop, a 100 kW rack still requires enough airflow to remove the remaining ~25 kW. To do this effectively, proper coldaisle or hotaisle containment is also essential.

At what point does cooling become a strategic infrastructure issue rather than simply an operational one?

Cooling has now become a strategic infrastructure consideration, directly impacting scalability, operational cost, sustainability ambitions, and the competitive market advantage. With AI and HPC, cooling is now shaping the entire data center infrastructure, including power, layout, sustainability, and total cost of ownership, moving away from being seen as an operational issue the moment it limits what can be deployed. We’ve hit the point where cooling is no longer something we operation, but rather it’s something we must strategically design for the future industry.

How can operators balance rising compute demand with growing pressure around sustainability, energy efficiency, and resource use?

The key to balancing rising computing demand with growing pressures is to first, pause and take a breath – not every rack moving forward will be running AI or ML workloads. At the same time, we must design and engineer nextgeneration data centers to be sustainable, energyefficient, and relevant for the next 10 to 15 years. Constant threeyear ripandreplace cycles of both hardware and cooling aren’t viable. We need a balanced approach that prioritizes smart, longterm TCO decisions.

What role do modelling and simulation technologies like CFD now play in improving operational efficiency and long-term planning?

CFD has evolved from being a design validation tool into a strategic operational tool. It is still very much needed, even in data centers that deploy direct to chip liquid cooling. In many cases, this has become more important, not less.

For example, CFD is needed to validate the airflow is still delivered correctly and that no hotspots form around non liquidcooled components since CFD is the only tool that can accurately model how that remaining heat behaves in the white space. CFD is also the only way to simulate transient and failure scenarios safely.

As infrastructure becomes more complex, CFD modelling becomes more important as operators can no longer rely on previously used design approaches alone.

Looking ahead five years, what’s one bold prediction you would make about the future of cooling and thermal management in the data centre industry?

Data center cooling and thermal management will continue to evolve over the next five years and well beyond. We expect ongoing adoption of hybrid cooling, through integrating air and liquid systems, to increase capacity, efficiency, and sustainability. Moving forward, data center design and cooling infrastructure will no longer follow a one-size-fits-all model, so all of us in the industry need to evolve and adapt to this way of thinking.