Gordon Johnson, Subzero Engineering’s Senior CFD Manager, examines how air cooling remains critical for data center memory storage and containment systems

By Amber Jackson
As Published on DataCentreMagazine.com

Why liquid cooling is becoming essential for AI-driven data centers—while air cooling still plays a critical role

As AI drives immense compute demands, it is expected to be the main cause of data center power demands doubling worldwide between 2022 and 2026—unless operators are able to tackle sustainability head-on to decrease rising emissions.

One of the main topics of conversation is thermal management, as GPUs continue to draw significant power densities. Cooling servers quickly has become one of the most pressing challenges in a hyperscale data center environment, particularly as traditional air cooling systems can no longer keep up with the demands of AI workloads.

To dig into this further, Subzero Engineering’s Senior CFD Manager, Gordon Johnson, shares his analysis that liquid cooling has quickly become the new norm for data centers of the future—but that air cooling is still required.

“Direct Liquid Cooling (DLC), and specifically Direct-to-Chip (DTC), is now essential for controlling heat,” he says. “However, about 25% of the heat produced by IT equipment still needs to be expelled through the air, especially from secondary parts such as memory subsystems, storage, and power delivery circuits.

“It is impossible to overlook this heat residue, and that’s where traditional airflow strategies are still needed, albeit in a supporting role.”

Confronting hyperscale cooling challenges

Gordon explains that hyperscale operators are seeing a sharp rise in OPEX from both power and cooling, given that it has become one of their most significant challenges.

“In recent years, power and cooling have become strategic levers and margin killers in hyperscale operations,” he says. “If you’re operating at scale, your P&L is directly tied to your power and cooling intelligence.

“Those who get it right will widen their advantage. Those who don’t could find AI infrastructure becoming financially unsustainable.”

He argues that efficiency is no longer just best practice, as—despite the rise of renewable energy resources—AI is effectively slowing down decarbonization.

“Data centers’ energy usage is driven by the fact that advancements in AI model performance frequently result in larger models and more inference, raising energy costs and contributing to sustainability challenges,” he explains.

“AI needs to get more efficient, not just more powerful.”

Energy consumption remains a key concern as AI continues to boom. Once deployed, these models require an enormous amount of inference infrastructure to process countless queries every day.

“Modern AI GPUs are now drawing upwards of 500 watts per chip,” Gordon says. “Hyperscale data centers that once operated in the 10–30 kW/rack range are now pushing 80–120 kW/rack to support AI training and inference.

“With air cooling limited to about 30–40 kW/rack, the air just cannot carry the created heat quickly enough, even with optimal containment and supply airflow.”

Embracing air cooling in direct liquid cooling

Higher compute density, increased energy efficiency, and more reliable thermal control at the component level are all made possible for hyperscale operators by DLC—and specifically DTC.

Gordon says: “It is a practical means of maintaining the safe thermal working range of contemporary CPUs and GPUs. DLC also permits higher incoming air temperatures, reducing reliance on traditional HVAC systems and chillers.”

However, he argues that advanced DTC systems do not eliminate the need for air cooling.

“Air cooling is necessary even with the most sophisticated DTC systems,” he says. “The cooling of non-critical components, cabinet pressurization, and residual heat evacuation still require airflow.”

Additionally, hot and cold aisle containment systems have been proven to separate the hot exhaust air from the cold intake air effectively.

“Efficiency increases can result in a cooling energy decrease of 10–30%,” Gordon says. “Containment is essential for optimizing the performance of air-cooled systems in legacy settings.

“Raised flooring, hot/cold aisles, and containment systems are becoming progressively more crucial in environments that are transitional or hybrid (liquid + air cooled).

“These airflow techniques aid in the separation of AI-specific and older infrastructure in mixed-use data centers. However, in modern AI racks, air cooling is the supporting act rather than the main attraction.”

For operators managing large megawatts of IT load, hot/cold aisle containment is one of the most cost-effective and space-saving solutions available. Gordon explains that making slight improvements to airflow containment can ultimately result in large-scale energy savings in high-density settings.

“By stabilizing temperature zones and lowering fluctuation, this improves cooling system responsiveness while lowering chiller load and encouraging energy-reuse initiatives,” he explains.

“Hot/cold aisle containment is no longer just a best practice—it is becoming a critical optimization layer in tomorrow’s high-performance, high-efficiency data centers.

“For operators managing hundreds of megawatts of IT load, hot/cold aisle containment is still one of the most cost-effective, space-efficient tools available.”

Where does the industry go from here?

As the data center industry continues to transition toward liquid cooling adoption, Gordon is eager for operators to understand that air cooling will remain relevant.

He explains that air management in the cooling stack is changing from a primary to a supporting, yet essential, system—highlighting that the industry is improving and re-integrating traditional tactics alongside cutting-edge liquid systems rather than discarding them.

“Hyperscalers are under constant examination to meet net-zero targets. In addition to complying with energy efficiency regulations, the hybrid solution offers data center operators a way to transition from conventional air-cooled facilities to liquid-readiness without requiring complete overhauls,” he says.

“With high-density AI workloads, air cooling just cannot keep up. It’s a physical limitation. Hybrid methods that combine regulated airflow with DLC are now the engineering benchmark for scalable, effective, and future-ready data centers.”

Gordon Johnson, Senior CFD Manager for Subzero Engineering, outlines the need for sustainability in data center operations and highlights how optimized strategies can reduce energy waste and environmental impact.

By Gordon Johnson, Senior CFD Manager at Subzero Engineering
As Published in Data Centre & Network News

An Environmental Cost

As essential as data centers are to our increasingly digital lives, they come at a huge environmental cost to our planet.

It doesn’t help that much of the energy required to power them is still sourced from fossil fuels. It’s one of the reasons that the industry has been identified as a major contributor to climate change.

Given the growing environmental concerns, it is now an urgent necessity to transition to sustainable, renewable energy sources, energy-efficient technologies, and recyclable materials. To impose the importance of net zero, governments and regulatory bodies worldwide are seeking to implement stricter environmental policies to meet global climate goals.

The adoption of sustainable design ensures adherence to these regulations. Furthermore, as sustainability becomes a crucial component of corporate social responsibility (CSR) for many organizations, and more consumers and businesses are favoring companies with strong environmental commitments, a strong sustainability policy can yield a competitive advantage in a tough marketplace.

Transitioning from White to Green

White space, as it relates to data centers, is the space inside a building devoted to IT hardware, such as servers, storage, and networking components. It is a highly controlled environment with restricted access, monitored for temperature, humidity, and other factors critical to maintaining the health of IT systems.

Increasing demand for data center performance and capacity while at the same time reducing operating costs requires an efficient use of white space. What could the transformation from white space to green building offer? And can it still deliver on operational excellence?

Incorporating renewable energy sources and embracing natural power supplies, such as wind or solar, enables operational efficiencies to be raised, cooling requirements reduced, and CO₂ emissions to be significantly reduced.

In addition, construction using recycled and recyclable materials also supports global initiatives in combating climate change, reducing waste, and lowering greenhouse gas emissions.

Green Building Certifications

According to the US Office of Energy Efficiency and Renewable Energy, data centers are one of the most energy-intensive building types, consuming 10 to 50 times the energy per floor space of a typical commercial office building. This energy consumption is only expected to increase due to high intensity emerging technologies such as artificial intelligence (AI), blockchain and cryptocurrency.

Global green building certifications, such as Leadership in Energy and Environmental Design (LEED), are heralding a new era of environmentally sustainable practices. These certifications set a framework for integrating recycled and recyclable materials with measurable benchmarks for sustainability, energy efficiency and environmental stewardship.

Globally recognized green building certifications and standards that evaluate the environmental impact and performance of buildings are essential in promoting environmentally conscious design in contemporary infrastructure. Internationally recognized indicators give data centers the means to demonstrate their commitment to minimizing environmental impact, and set a bar for best practice in sustainable construction and operation. This encourages industry-wide adoption, opening the door for a more sustainable future.

Balancing Costs and Sustainability

Transitioning to greener materials and practices offers significant environmental benefits, but it also raises questions about cost. Does the investment in recyclable, green materials balance the return on investment?

Upfront costs of adopting green building practices are indeed high, particularly in legacy data centers, but the long-term financial benefits are indisputable. Over time, utilizing energy-efficient designs and systems can lead to a lower total cost of ownership (TCO) by reducing power and operational expenses.

Integrating renewables can also decrease organizations’ reliance on fossil fuels, helping them to better manage any future energy challenges. Additionally, data centers that actively pursue net zero initiatives can enhance their brand perception by complying with regulations, benefiting from a value that is difficult to quantify.

These benefits justify the initial investment. When evaluating costs concerning TCO, the argument for both financial and environmental sustainability is compelling.

The Power of Collective Responsibility

While data centers have an unavoidable influence on the environment, the industry is quickly establishing itself as a leader in environmental sustainability by implementing a variety of net zero strategies. However, all industry stakeholders need to play a role in the collective accountability for an environmentally friendly future.

Partnerships are integral to this collaborative approach. From operators adopting renewable energy sources to designers innovating with eco-friendly materials, investors funding sustainability projects to policymakers incentivizing green practices; we are all answerable in the acceleration of sustainable operation.

Setting an Example

Taking decisive action is the first step to sustainability. The choice of being a sustainability leader yields benefits beyond the environment; it brings about a positive change chain reaction, a ripple effect across all industries. Positive transformation inspires and influences all sectors and markets.

Adopting this role of responsibility leverages a legacy of accountability and investment in sustainability, with the long-lasting positive impact on the globe to be enjoyed by the next generation of technology entrepreneurs.

About the writer 

Gordon Johnson is the Senior CFD Engineer at Subzero Engineering, responsible for planning and managing all CFD-related jobs in the US and worldwide. 

He has over 25 years of experience in the data center industry which includes data center energy efficiency assessments, CFD modeling, and disaster recovery.  He is a certified US Department of Energy Data Center Energy Practitioner (DCEP), a certified Data Centre Design Professional (CDCDP), and holds a Bachelor of Science in Electrical Engineering from New Jersey Institute of Technology.   

We’re slowly beginning to see a shift toward purpose-built AI data centers, especially from hyperscalers like AWS, Google, and Microsoft. With the latest announcement from the White House in their AI Action Plan, the US Government has formalized exporting American AI across the world, and promoting the rapid buildout of data centers and AI-focused infrastructure, making this shift simpler to implement. 

These aren’t just scaled-up legacy setups, however. They’re designed from the ground up for AI workloads and require specific infrastructure, particularly in power delivery and cooling infrastructure. But what do they need that separates them from the ‘standard’ facility, and what makes it so challenging to try to retrofit legacy data centers for these?

By Gordon Johnson, Senior CFD Manager at Subzero Engineering

AI infrastructure demands

AI is quickly becoming the dominant consumer of compute, and traditional infrastructure just can’t keep up. The industry is shifting to fundamentally new architectures and data centers that don’t adapt will be left behind as the industry transitions to radically new designs, which is where the US White House announcement in unveiling their AI Action Plan is helping to accelerate the development of AI infrastructure across the country. But its main focus on the rapid buildout of AI-ready data centers, exporting AI technology, and a more targeted focus on the infrastructure required to support this shift, reflects what hyperscalers are doing.

It is hard to overlook the infrastructure constraints of traditional data centers as AI workloads grow in complexity and scale. We’re entering a new era in data infrastructure, one that legacy data centers weren’t built to handle. To address the specific requirements of large-scale artificial intelligence, top hyperscalers such as AWS, Google, and Microsoft are spearheading the evolution by building a new class of data center from the ground up: AI-native infrastructure.

Why can’t legacy facilities handle AI’s demands?

Legacy data centers were designed for general-purpose computing. They were built to account for predictable workloads, with moderate power usage and flexible hardware.

AI has different needs and many legacy data centers are unsuitable for the task due to the scope and intricacy of the criteria.

AI workloads are vastly more power intensive than traditional workloads. They require three to ten times as much electricity per rack, therefore merely adding extra GPUs to the same old racks is not an option. The extreme heat produced by CPUs, GPUs, and TPUs cannot be controlled using conventional air-cooling methods, so in meeting the requirements of contemporary AI, liquid cooling infrastructure such as direct-to-chip becomes necessary.

Unpredictable bursts of energy required for ultra-fast connections between thousands of nodes are necessary for AI training, and the requirement for densely populated, high-performance clusters conflicts with the sprawl of traditional data halls. Long cable runs and low-density racks can increase latency, reducing performance for large AI jobs. This kind of infrastructure concern is exactly why this AI Action Plan couldn’t have come at a better time, with targets to fund next-gen, resilient digital infrastructure and collaboration between public and private organisation on AI system reliability and performance.

Legacy Data Centers Built for Yesterday

Legacy data centers were built for yesterday’s workloads. AI isn’t just demanding more, it’s demanding different. Hyperscalers know it, and they’re not waiting around. The future of digital infrastructure is being redefined by the emergence of the purpose-built AI data center era.

Retrofitting an existing data center for AI isn’t easy.  Typically, data centers will need to leverage their existing investments in air cooling while selectively deploying liquid cooling where needed.  Although infrastructure can be reworked and redesigned, concessions will always need to be made. These compromises could come at the expense of performance and efficiency. Power availability (typically capped at the site level) and cooling capacity (particularly in raised-floor environments), while rack weight and floor loading, together with ceiling height restrictions that reduce airflow design, are physical constraints on most older sites. Add in the layout obstructions and interconnect distances that cause latency bottlenecks, and this could be a compromise too far.

What Makes AI Workloads Different

Traditional data centers tend to average 5–20 kW per rack, whereas the power draw per rack due to AI workloads can be significantly higher than traditional compute, pushing 30–100 kW per rack or higher. Infrastructure needs to be approached very differently to support this degree of power density, as on-site substations, busways, and high-capacity PDUs are increasingly the standard rather than the exception.

AI workloads are inherently unforgiving of infrastructure failures. While traditional workloads can often handle transient faults or recover from minor slowdowns, AI training that is running for days or weeks requires near-perfect uptime, clean compute environments, and dependable performance. Even small variations or inconsistencies in hardware, firmware, or thermal performance can be catastrophic. Not because it can’t recover, but because the cost of failure is so high, with every crash potentially hours (or days) of lost compute, wasted energy, and missed opportunity.

Designing for AI

To unlock the full value of AI, the AI data center infrastructure must evolve. AI demands infrastructure that’s not just fault-tolerant, but fault-predictive and self-healing.

When embarking on a new data center build, you must consider:

• High-Density Power and Cooling
  Custom power paths must be able to handle 80–100kW racks or more, while air cooling, the mainstay of legacy facilities will not be enough to cool these high-density racks. Advanced thermal strategies such as liquid cooling and direct-to-chip solutions must be integrated into the infrastructure of the AI data center.
• Architecture
  Physical CPU/GPU/TPU cluster layouts need to be optimized to ensure latency is minimized and training throughput maximized. Consolidated floorplans and thermal awareness allows for increased efficiency, faster deployment, and future expansion.
• AI-Centric Design
  Real-time predictive failure monitoring and telemetry should be used on every component from temperature to power draw. Machine learning-based fault prediction isn’t optional anymore. It’s how downtime can be preempted, and uptime can be optimized.
• Sustainability
  Carbon-neutral power resources, energy storage, recycling and reusing waste heat output and using alternative building materials can all assist with sustainability and environmentally friendly policies and strategies. Adherence to green strategies can not only improve the facility’s efficiency but can provide competitive advantage.

Legacy data centers were designed for flexible, general-purpose compute. However, AI clusters depend on ultra-low-latency interconnects between accelerators. That changes everything from the physical layout to how cable trays are built. New facilities need to be dense, compact, and often modular designed to reduce data movement friction.

Bigger and Better

AI data centers aren’t just bigger — they’re different by design. Larger footprints are not a luxury but rather a necessity to accommodate the density and specialized layout, thermal management and performance characteristics AI environments.

With the White House calling for more land and more power, hyperscalers are starting to plan and construct data centers in pod-based, modular designs that are tailored for AI workloads and optimized for independent cooling, powering, and scaling. Workloads are not distributed equally throughout the data center by AI clusters. Rather, concentrated compute pods (hundreds to thousands of GPUs in a tightly integrated fabric) are needed, calling for larger real estate to accommodate the GPU/TPU cages or liquid-cooled racks, zoning to isolate various workloads and effectively manage thermal loads, and a larger whitespace per cluster to accommodate power, cooling, and cabling routes.

Space is needed to manage the significant heat produced by high-density AI workloads, for heat exchange devices, immersion tanks, liquid cooling loops, and greater hot/cold aisle separations, often with isolated or enclosed cooling corridors.

Each pod needs short, direct power paths, larger substations, and dedicated power rooms. They also require extra room for redundant switchgear, transformers, and UPS systems, as well as increased floor loads and reinforced infrastructure to support denser, heavier racks.

A Fundamental Rethink

The White House’s AI Action Plan reinforces what industry leaders are starting to adopt, that companies that are already leading in AI-native infrastructure are paving the way forward for AI, and this transformation needs to be seen across the industry. Hyperscalers aren’t building these new AI driven data centers because they’re trendy or because they want the biggest facility. It’s because it’s necessary. AI is not an experimental upgrade cycle. It’s fundamental infrastructure. And with any foundational shift in computing, it demands a matching evolution in physical and digital architecture.

Companies that continue trying to run next-generation AI on last-generation infrastructure will find themselves bottlenecked in performance, efficiency and ultimately competitiveness. The AI-native future is rapidly overshadowing the computing era for which legacy data centers were constructed.

For this reason, hyperscalers are designing data centers that embrace and give priority to specially designed AI infrastructure. They are not merely scaled up facilities. They are precision engineered, and offer the performance, resilience, and AI acceleration that will define the next decade. 

About the writer 

Gordon Johnson is the Senior CFD Engineer at Subzero Engineering, responsible for planning and managing all CFD-related jobs in the US and worldwide. 

He has over 25 years of experience in the data center industry which includes data center energy efficiency assessments, CFD modeling, and disaster recovery.  He is a certified US Department of Energy Data Center Energy Practitioner (DCEP), a certified Data Centre Design Professional (CDCDP), and holds a Bachelor of Science in Electrical Engineering from New Jersey Institute of Technology.   

Direct Liquid Cooling (DLC), especially Direct-to-Chip Cooling (DTC), is now essential for high-density AI racks. DLC enables higher compute densities and better energy efficiency. 

Air cooling alone can’t keep up with the thermal output of modern CPUs and GPUs, but even with advanced DTC, approximately 25% of ITE heat still needs air cooling.

Cold and hot aisle containment is a tried-and-tested climate control strategy that separates the two airflows while improving energy efficiency. For energy savings that can’t be ignored, should hot/cold aisle containment be considered a necessity in hyperscale data centers?

By Gordon Johnson, Senior CFD Manager at Subzero Engineering

Introduction

AI workloads are driving unprecedented compute demand. Not only is the demand intensifying rather than decreasing, it is also altering the economics and structure of computing at all levels.

AI is expected to be the primary cause of the anticipated doubling of data center power demand worldwide between 2022 and 2026 and, unless offset, increased compute = increased emissions.

With GPUs drawing up to 700W each and power densities exceeding 80–100 kW per rack, thermal management has become one of the most critical challenges in hyperscale environments. Conventional air-cooling techniques can no longer keep up with the thermal densities of contemporary AI workloads and liquid cooling is no longer just a viable option for the future. It has become the new norm.

Direct Liquid Cooling (DLC), and specifically Direct-to-Chip (DTC), is now essential for controlling heat. However, about 25% of the heat produced by IT equipment still needs to be expelled through the air, especially from secondary parts such as memory subsystems, storage and power delivery circuits. It is impossible to overlook this heat residue, and that’s where traditional airflow strategies are still needed, albeit in a supporting role.

Challenges

Hyperscale operators are seeing a sharp rise in OPEX from both power and cooling, and it’s becoming one of their most pressing financial and operational challenges.

In recent years, power and cooling have become strategic levers and margin killers in hyperscale operations. If you’re operating at scale, your P&L is directly tied to your power and cooling intelligence. Those who get it right will widen their advantage. Those who don’t could find AI infrastructure becoming financially unsustainable.

Efficiency is no longer just best practice

Many hyperscalers have already hit PUEs of 1.1–1.2, limiting room for improvement and further efficiency gains. This suggests that absolute power usage is now rising even if relative efficiency stays the same. In high-density environments, even marginal improvements in airflow containment can lead to significant energy savings.

Despite the rise of renewable energy resources, AI is effectively slowing down decarbonization. Data centers’ energy usage is driven by the fact that advancements in AI model performance frequently result in larger models and more inference, raising energy costs and contributing to sustainability challenges. AI needs to get more efficient, not just more powerful.

Air-Cooling Limits

Millions of kWh of electricity are needed to train large-scale AI models like GPT-4, Gemini or Claude-class. The scale of this energy consumption is one of the key concerns of the modern AI era. Once deployed, these models require an enormous amount of inference infrastructure to process the countless number of queries every day, and this can exceed training energy usage.

Modern AI GPUs (like NVIDIA H100 or AMD MI300X) are now drawing upwards of 500 watts per chip. Hyperscale data centers that once operated in the 10–30 kW/rack range are now pushing 80–120 kW/rack to support AI training and inference. With air cooling limited to about 30–40 kW/rack, the air just cannot carry the created heat quickly enough, even with optimal containment and supply airflow.

Direct Liquid Cooling (DLC)

Higher compute density, increased energy efficiency, and more reliable thermal control at the component level are all made possible for hyperscale operators by DLC, specifically DTC. It is a practical means of maintaining the safe thermal working range of contemporary CPUs and GPUs. DLC also permits higher incoming air temperatures, reducing reliance on traditional HVAC systems and chillers.  

In addition, Direct-to-Chip (DTC) can reduce overall cooling energy (PUE impact) by up to 40%, when compared to traditional air systems, by targeting cooling directly to the hottest components. However, even the most advanced DLC/DTC systems do not eliminate the need for air cooling. Air cooling is necessary even with the most sophisticated DTC   systems. The cooling of non-critical components, cabinet pressurization and residual heat evacuation still require airflow.

Hot/Cold Aisle Containment

Hot/cold aisle containment is a proven architectural strategy that separates the hot exhaust air from the cold intake air. Through containment, the two air temperatures are kept from mixing, meaning colder air is ensured, servers are reached more directly, cooling load is decreased and thermal predictability is enhanced. Efficiency increases can result in a cooling energy decrease of 10–30%. Containment is essential for optimizing the performance of air-cooled systems in legacy settings.

Raised flooring, hot/cold aisles, and containment systems are becoming progressively more crucial in environments that are transitional or hybrid (liquid + air cooled). These airflow techniques aid in the separation of AI-specific and older infrastructure in mixed-use data centers. However, in modern AI racks, air cooling is the supporting act rather than the main attraction.

The Case for Containment

For operators managing tens or hundreds of megawatts of IT load, hot/cold aisle containment is one of the most cost-effective and space-saving solutions available.

Even with DTC intensive systems, containment is not obsolete. Modest improvements to airflow containment can result in large-scale energy savings in high-density settings. By absorbing and diverting leftover heat from partially liquid-cooled equipment, containment enhances airflow circulation to secondary components. By stabilizing temperature zones and lowering fluctuation, this improves cooling system responsiveness while lowering chiller load and encouraging energy-reuse initiatives.

Hot/cold aisle containment is no longer just a best practice; it is becoming a critical optimization layer in tomorrow’s high-performance, high-efficiency data centers. For operators managing hundreds of megawatts of IT load, hot/cold aisle containment is still one of the most cost-effective, space-efficient tools available.

Conclusion

As hyperscale operators transition to liquid-cooled infrastructure, the expectation might be that airflow strategies will become irrelevant. But the reverse is happening. In the cooling stack, air management is changing from a primary to a supporting, yet essential, system. The industry is improving and re-integrating traditional tactics alongside cutting-edge liquid systems rather than discarding them.

Hyperscalers are under constant examination to meet net-zero targets. In addition to complying with energy efficiency regulations, the hybrid solution offers data center operators a way to transition from conventional air-cooled facilities to liquid-readiness without requiring complete overhauls.

With high-density AI workloads, air cooling just cannot keep up. It’s a limitation of physical limitation. Hybrid methods that combine regulated airflow with DLC are now the engineering benchmark for scalable, effective, and future-ready data centers.

About the writer 

Gordon Johnson is the Senior CFD Engineer at Subzero Engineering, responsible for planning and managing all CFD-related jobs in the US and worldwide. 

He has over 25 years of experience in the data center industry which includes data center energy efficiency assessments, CFD modeling, and disaster recovery.  He is a certified US Department of Energy Data Center Energy Practitioner (DCEP), a certified Data Centre Design Professional (CDCDP), and holds a Bachelor of Science in Electrical Engineering from New Jersey Institute of Technology.   

Discover how optimized containment strategies and sustainable Composite AisleFrame (CAF) systems can reduce data center carbon emissions by up to 4,299 kg CO₂ per frame while improving operational efficiency and cutting costs.

By Andy Conner, Channel Director EMEAr at Subzero Engineering

Introduction

Environmental consciousness is not just a trend. We can’t rapidly mend the hole in the ozone layer and climate change concerns won’t be undermined by ever-evolving technology any time soon. Humankind has a collective responsibility to reduce our carbon emissions, to lower waste, and to change the way we create and use energy in our day-to-day lives. However, it’s a constant challenge for organizations to balance scalability, operational efficiency and power resourcefulness with sustainability objectives.

Data centers are hugely energy-intensive buildings. Handling an increasingly growing capacity and complexity of AI and high-performance computing (HPC) means they are consistently using an aggressive amount of power and energy. It’s imperative that we minimize the environmental impact of these buildings, reducing the power consumed while maximizing the energy that is used. New strategies need to be implemented, sustainable materials deployed, and a mindset change to get to net zero and stay there.

Goals and Objectives

Reduce, recycle, and reuse policies should be integrated in every organization’s core values; however, longer-term environmental goals that support a sustainable infrastructure built with energy-efficient technologies and renewable energy sources must be considered when redesigning or extending legacy data center facilities, or building new ones.

One of the best strategies to accomplish sustainability objectives in these buildings is by utilizing optimized containment. Optimizing containment is a vital first step toward achieving a sustainable data center and can significantly reduce unfavorable environmental consequences.

After the ITE, cooling is the biggest consumer of a facility’s energy resources. Containment strategies decrease energy waste and efficiently regulate airflow. This enables data centers to maximize and boost operational efficiency while minimizing their impact on the environment.

Utilizing containment helps maintain consistent thermal temperatures and increases cooling effectiveness by keeping the hot and cold air streams separate, enabling a regulated airflow environment and increasing the efficiency of the cooling systems. This way data centers can conserve energy, not consume more.

Traditionally, an AisleFrame containment system is made of steel. This provides an integral floor supported structure that physically separates the cooled and expelled hot air. With excellent sustainability credentials, steel is 100% recyclable and can be melted down and reused time and again without deterioration. Through closed-loop recycling, every ton of steel scrap recovered can replace one ton of primary steelmaking while keeping its integrity in terms of properties or performance. In addition, steel’s long lifespan and minimal maintenance requirements contribute to its overall sustainability attestation. On the flip side however, decarbonizing remains a challenge and a global priority, and steelmaking currently contributes around 8% of the world’s total carbon emissions.

The Alternative

Composite AisleFrame (CAF) is a system made from alternative and sustainable materials 100% and is a frame-based, floor supported structure for IT/HPC deployments. Used in the construction industry for more than 20 years in many proven applications, such as airplane tail structures, outdoor utility/telephone poles, and transportation bridges, this composite material has now been refined for specific use in data centers to be denser, stronger, and with additional fireproof properties. 

CAF has many benefits compared with a Steel AisleFrame system. Every element in a data center has an intrinsic cost that needs to be accounted for, and steel is a heavy material. This translates to high transit costs and increased installation times that must be factored into the build.

In comparison, CAF material is 50% lighter than steel alternatives. It can be installed swiftly without the need for powder coating and is easily reconfigurable as requirements change, offering more flexibility and easier scalability. It can be reused multiple times and has an extended lifespan over steel, supporting waste reduction and net-zero initiatives, leading to lower Total Cost of Ownership (TCO). It can be flat-packed, allowing more product to be shipped in the same physical footprint, and delivers on lower transportation emissions and costs, offering up to 4,299 kg CO₂ savings per frame compared with non-recycled steel and up to 429 kg CO₂ savings per frame compared with recycled steel.

CAF’s strength per linear meter and its seismic compliance enable multi-level data centers to have CAF systems running throughout each building floor without the additional financial risk of having to strengthen weight-bearing floors. Its higher tensile and flexural attributes, with a better compressive strength-to-weight ratio than steel, mean CAF is more efficient structurally.

While steel is resource-heavy, CAF is non-resource-heavy in implementation. This means the CAF system can be delivered and installed in a fast and time-appropriate fashion. A steel structure can potentially take months to be shipped, but CAF could conceivably be delivered in weeks.

Conclusion

Optimizing cooling by separating the hot and cold air can ensure stable and consistent temperature distribution. By improving energy efficiency and overall cooling effectiveness, this can deliver on significant energy savings.

As the industry shifts to using greener technology, the development of a sustainable infrastructure built with energy-efficient technologies and recycled materials continues to be a key strategy in the next generation of high-performance data centers.

Renowned for being hugely power-intensive buildings, data center operatives must constantly investigate strategies and technologies to lower their TCO. Whether restructuring, redesigning, or building from scratch, the cost savings accredited to CAF can contribute to a much quicker return on investment in data center infrastructure. And it’s a win-win when you’re lowering operational costs and optimizing the facility’s high performance and reliability at the same time as achieving long-term global environmental objectives.

About the writer 
Gordon Johnson is the Senior CFD Engineer at Subzero Engineering, responsible for planning and managing all CFD-related jobs in the US and worldwide. 

He has over 25 years of experience in the data center industry which includes data center energy efficiency assessments, CFD modeling, and disaster recovery.  He is a certified US Department of Energy Data Center Energy Practitioner (DCEP), a certified Data Centre Design Professional (CDCDP), and holds a Bachelor of Science in Electrical Engineering from New Jersey Institute of Technology.