Thermal management and performance enhancement of data centers architectures using aligned/staggered in-row cooling arrangements

ABSTRACT This research investigates the effects of using aligned/staggered arrangements of cooling units and top aisles containments on the performance of in-rows cooling architectures of data centers. Four different configurations of cooling architectures are numerically set up inside identical data centers of aligned/staggered cooling-units arrangements and with/without top aisles containments. ANSYS IcePak CFD software is used to build models for the four configurations. A scaled physical module of the data center is experimentally constructed and tested for model validation. Air streamlines, velocity vectors and temperature distributions in addition to two-thermal perfromance indices (SHI and IOM) and two-energy efficiency metrics (β and ηr) are used to evaluate and compare thermal environmental and energy efficiency performances of the different configurations. Results show that (i) aligned arrangement has better thermal and energy efficiency performance, (ii) thermal perfromance and energy efficiency indices of intermediate racks of the racks array are better than those of end racks, especially in the staggered arrangements, (iii) thermal environment problem of end racks can be solved by starting and ending racks row by CRAC units, and (iv) using top aisles containment improves the thermal and energy efficiency perfromance indices of data centers racks, especially for intermediate racks.