A new white paper from Future Facilities explores how Kao Data used computational fluid dynamic (CFD) simulations to validate and make informed decisions around the design, implementation, and operation of their indirect evaporative cooling (IEC) system.
Kao Data is a wholesale co-location data centre located in the UK. The first of the company’s four planned data centres is designed to sustainably support high performance computing and intensive artificial intelligence. According to a new white paper from Future Facilities, Kao Data was determined to design their data centre to be as energy efficient as possible.
One way in which the company is achieving this goal is through the use of an indirect evaporative cooling (IEC) system, which was “adopted to create a data centre that was greener, as well as more efficient and cost-effective to run.”
The company turned to computational fluid dynamic (CFD) simulations to ensure that the IEC system would meet the needs of a data centre that caters to “the increasingly high-density deployments characterised by machine learning and deep learning workloads and offering the ultimate in Open Compute Ready (OCP) infrastructure.” The company used CFD simulations to test various scenarios for both the external and internal environments around the data centre campus.
One simulation was around wind speed and direction changes and how they might impact air flow around the building. “The aim was to understand which wind directions caused the worst recirculation of air and how best to optimise set-up to avoid moisture getting trapped around the building and drawn back into the IECs.” Kao Data also used CFD modelling to configure their whitespace and validate thermal performance.
The paper outlines the data centre’s specifications and provides an overview of the unique features of IEC systems, which are designed for data centre environments. The paper then explores the reasons why CFD simulation was critical to the success of the project and how they’re currently using simulations.