Evaluation of air distribution system's airflow performance for cooling energy savings in high-density data centers

Experimental investigation on the thermal performance of water cooled multi-split heat pipe system (MSHPS) for space cooling in modular data centers

“…, a in cw sys e T T R Q − = (6) Where Q e is the cooling capacity of evaporator section, T a,in is the inlet air temperature, and T cw is the average cooling water temperature defined as:…”

“…Among the overall energy consumption in data centers, 30%~50% is used for device and space cooling [4]. In addition, the average power density in most high density data center is 1.5 kW/m 2 or 5~8 kW/rack [5][6][7]. In order to quantify the cooling efficiency, data centers commonly employ an index called Power Usage Effectiveness (PUE), which is defined as the ratio of the total energy consumption of a data center to the energy consumed by the IT equipment inside.…”

Energy Efficient Data Centers

“…For Stride (1), flows sourcing from the first five servers of the edge switch go to servers in the same subnet, and flows from the sixth server travel to the server in the next subnet or in the next pod. In contrast, all the flows in Stride (6) go to the neighboring subnet, and all the traffic in Stride (36) and Stride(216) is inter-pod traffic. For Staggered(n), it has fixed values for p 1 and p 2 as the probabilities of flow going to the same subnet and other subnets of the same pod, respectively.…”

Effect of plenum chamber obstructions on data center performance

“…Combined experimental and computational studies were carried out to quantify the importance of geometry and it was found that for proper prediction of the mixing process, the detailed tile opening patterns must be taken into consideration. To evaluate air distribution system's air flow performance of high density data-centers, performance metrics were computed with CFD modeling of 46 different designs by Cho et al [10] and derived energy saving factors. The effect of perforated tile was captured in CFD by many studies, while taking into account additional effects like buoyancy and employed improved thermal boundary conditions that yielded a better agreement to actual data center values [11e15].…”