Getting the most out of buildings—cost, speed to market, sustainability, user experience, wellness, and more—is the focus of “The Buildings Conference 2023” sponsored by BuiltWorlds. ESD President Kurt Karnatz joins a panel of professional peers as part of the two-day event starting May 18 in Chicago. (Learn more about BuiltWorlds conference.)
The complex and critical nature of data centers requires thorough analysis to ensure IT and support equipment is adequately cooled. One tool that proves particularly beneficial in data center applications is computational fluid dynamics.
Computational fluid dynamics (CFD) utilizes numerical analysis and underlying principles of fluid mechanics and heat transfer to model behavior of fluids. For data centers, this means that CFD can be used to model data halls and understand airflow and heat transfer as it pertains to cooling critical equipment – essentially aiding in design validation, troubleshooting and retro-commissioning efforts.
The value in CFD is the ability to predict the effect of air distribution on thermal performance prior to construction. While a design engineer can properly size cooling equipment and place it near the IT load, optimizing air distribution and cooling can sometimes be more nuanced. Deficiencies can lead to complications such as overheating, premature failure of IT equipment, increased energy consumption and inability to support the design load. By building a model of the data hall and analyzing it using a CFD software, the proposed design can be validated and issues such as air recirculation, bypass, mixing and hot spots can be predicted. If CFD analysis identifies potential issues, the engineer can begin to study different design solutions to alleviate them, thereby saving time and money that would be spent in the field if issues were discovered during operation. CFD can also be used to understand the effectiveness of cooling varying IT layouts and determine the feasibility of supporting proposed cabinet deployments.
Airflow distribution and cooling effectiveness become more unpredictable when considering equipment failure analysis. Using CFD, the engineer can create multiple failure scenarios to understand how active cooling units will respond and if they will be able to maintain effective cooling to the entire IT load within the data hall. By selecting worst-case failures such as adjacent cooling units, corner units or units supporting isolated loads, the engineer can gain insight into the resiliency of the data hall and whether the design should be modified to accommodate critical failures.
CFD can be just as beneficial for existing data halls. When looking to refresh IT equipment or install additional equipment in a legacy data hall, CFD can be used to determine the optimal location for the new equipment and determine whether additional cooling capacity is needed to support the installation while maintaining the required redundancy. CFD can also help with troubleshooting issues in a data center such as hot-spots and analyzing the efficacy of proposed upgrades.
CFD is a tool that allows engineers to understand the behavior of airflow and heat transfer within a space. For data centers, this means that CFD can be used to construct models of proposed or existing data hall spaces that reveal attributes related to thermal performance, air distribution and resiliency. Regardless of the type of data hall project, CFD is a powerful tool that provides valuable assistance with design validation and retro-commissioning efforts.
This article was subsequently posted by leading industry publication Engineered Systems magazine.
Reach out to Alyssa to learn more about the benefits of Indoor CFD Modelling in Data Center Applications.