Maintaining optimal operating conditions is increasingly critical for modern data centers. Todays and tomorrow’s server infrastructure is tasked with delivering more computing power—and with that power comes heat that must be mitigated.

But when it comes to cooling, data centers have options, and it may not always be easy to determine the right solution to meet your needs. How to determine the right data center cooling solution? In finding the answer, it’s worth starting by examining a few critical requirements for your data center operations:

Designing for Resiliency

Data center resilience is an important metric for any data center. As defined by the U.S. Department of Energy’s (DOE’s) Office of Energy Efficiency & Renewable Energy, resilience is a data centers ability to “prepare for and adapt to changing conditions and withstand and recover rapidly from disruptions.”

Though resiliency is commonly framed around major external factors—things like severe weather events, cyberattacks, physical assault, or fuel supply unavailability—data center resiliency can be contingent upon a number of internal factors, too. These include the ready availability of power and cooling for IT operations, per DOE.

It means you need a cooling system you can trust to provide continuous service for the lifetime of your server systems. System failure can lead to costly downtime for your operations and critical outages for your customers. Reliable cooling systems are essential for total data center resiliency.

Greenfield or Brownfield?

Your choice of cooling systems may depend on whether your project is a greenfield deployment or a brownfield deployment.

In a greenfield deployment—in which the design, installation, and configuration of your data center is all brand new—it’s worth evaluating state-of-the-art options. And not just for cooling, but for every part of the data center infrastructure. Utilizing outmoded systems may put your new data center behind the times before it is even fully operational.

In a brownfield deployment—in which you’re utilizing and upgrading an existing structure or systems—it may make sense to maintain a legacy cooling system to save on upfront costs. But it’s worth considering an upgrade in this scenario, too, in order to best prepare for the future. Traditional server rack configurations reached 10 kilowatts of power were considered high-density, but tomorrow, extreme-density configurations can top out at 40 kilowatts. That number is fundamentally untenable to maintain optimal temperatures with traditional air-cooling methods.

Maximizing Efficiency

For data centers focusing on efficiency, power usage effectiveness (PUE) is a critical metric. The National Renewable Energy Laboratory (NREL) defines PUE as “the ratio of the total amount of power used by a computer data center facility to the power delivered to computing equipment.”

According to NREL, data centers report a wide range of PUE figures, but the overall average tends to be around 1.8. Data centers focusing on efficiency, however, commonly target PUE values of 1.2 or less—and your cooling system can have a significant impact on that figure. PUE calculations account for the power used by fans and pipe trace heaters associated with outdoor cooling equipment. Heating, ventilation, and air conditioning (HVAC) are also considered, calculating the power consumption of fan walls, fan coils that support the data center electrical rooms, and the makeup of the air unit. These things are worth serious evaluation when selecting a cooling solution.

Equipment Protection

Your server infrastructure represents considerable investment, and data center owners and managers are tasked with protecting and maintaining IT equipment to maximize its longevity and generative output.

Temperature control and equipment protection go hand in hand. The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) recommends that servers and storage hardware operate in the 64.4 F to 80.6 F (18 C to 27 C) range in data processing environments. Temperatures exceeding those ranges have the potential to impact the longevity of server and storage hardware.

Data Center Cooling: A Rundown of Your Options

Keeping these critical needs in mind, let’s take a look at some of the options data centers must choose from:

Air cooling. Air-cooling systems are the most conventional method for maintaining environmental conditions. They release cool air into a cold aisle, where it is drawn through the front of the IT hardware and across the components before being ejected into the hot aisle. That hot air is then rejected into the hot aisle, returned to the HVAC unit, cooled and delivered back into the data center. But these methods increasingly struggle to efficiently handle greater power densities and the resulting heat.

Indirect cooling. Indirect cooling, also called cold plate cooling, works by absorbing and dissipating heat through closed-loop liquid-cooling systems. More efficient and compact than traditional air-cooling systems, they offer some clear advantages. However, they can be expensive and rely on a steady supply of water. Tubes and water pumps have the potential to corrode over time, increasing the risk of leaks in spaces filled with electronics.

Immersion cooling. Immersion cooling applications submerge computer components in a nonconducting dielectric fluid, which convects heat away as the fluid flows through a heat rejection mechanism. And by eliminating the limitations of air or indirect cooling, they can deliver some big benefits for data centers.

Forward-thinking data center owners and operators are already in on immersion cooling, especially when it comes to cloud computing. Immersion cooling brings a number of benefits to the table on its own, including better thermal performance, which can lead to less equipment throttling or downtime. Immersion cooling also enables better PUE for data centers by eliminating the need for power-hungry fans, allowing facilities to divert more kilowatt consumption directly to hardware. Finally, it can help extend equipment life by reducing server design complexity and reducing operating temperatures to prevent thermal degradation of server components.

Interested in learning more?

CompuZol™ immersion cooling fluid solutions from Lubrizol represents an ideal cooling solution for data centers, helping to keep your IT infrastructure resilient, efficient and operating at peak performance.

At Lubrizol, we work collaboratively with our data center partners to identify specific needs, enabling us to deliver a complete solution along with ongoing support. Contact your Lubrizol representative to find out how CompuZol can transform your operations today.

Evaluating Your Cooling Options