Improving Cooling Effectiveness in Air-Cooled Data Centres: Analysis and Assessment using Computational Fluid Dynamics

Abstract

Datacentre efficiency becomes increasingly important as demand for IT infrastructure grows. This thesis presents detailed Computational Fluid Dynamics analyses of flow and heat transfer within typical raised floor datacentres. Compared with previous studies, greater understanding of the unsteady flow and heat transfer in datacentres resulted from using Scale-Adaptive turbulence modelling instead of a RANS approach. This led to a novel rack-level sub-cooler design, which was shown to allow higher volumetric heat generation rates within datacentres.