A Comparative Numerical Study of Effectiveness of Cold Aisle Containment in Data Centers by Varying Rack Porosity Using Computational Fluid Dynamics

Abstract

Depending on specific needs and workloads, several racks with varied component densities may be used in a data center. As server density increases, porosity decreases, and the opposite is also true. A frequently used method called cold aisle containment separates hot air and cold air flows in data center settings to improve cooling effectiveness. In this paper, the performance of a data center is investigated using computational fluid dynamics, and the influence of porosity on cold aisle containment is evaluated using well-established non-dimensional performance parameters. The value of RTI without containment increased with an increase in porosity and a maximum RTI of 217 was found with a porosity of 0.75. Regardless of the rack number and porosity, containment provides the optimal RTI values. The results indicate that the SHI and RHI values for rack 1 for all porosities, without confinement, are outside of the permissible range and at higher porosities containment has no significant effects on SHI and RHI. RCILO values for racks 2, 3, and 4, with or without containment, fall within the 80-85% range, indicating temperatures below 13°C. RCIHI value is 1 for all cases considered indicating no rack is out of the recommended temperature of 25°C.