0
Research Papers

Energy Optimization of Air-Cooled Data Centers

[+] Author and Article Information
H. E. Khalifa1

Department of Mechanical and Aerospace Engineering, Syracuse University, Syracuse, NY 13244hekhalif@syr.edu

D. W. Demetriou

Department of Mechanical and Aerospace Engineering, Syracuse University, Syracuse, NY 13244

1

Corresponding author.

J. Thermal Sci. Eng. Appl 2(4), 041005 (Feb 08, 2011) (13 pages) doi:10.1115/1.4003427 History: Received August 24, 2010; Revised January 06, 2011; Published February 08, 2011; Online February 08, 2011

Abstract

The work presented in this paper describes a simplified thermodynamic model that can be used for exploring optimization possibilities in air-cooled data centers. The model has been used to identify optimal, energy-efficient designs, operating scenarios, and operating parameters such as flow rates and air supply temperatures. The results of this analysis highlight the important features that need to be considered when optimizing the operation of air-cooled data centers, especially the trade-off between low air supply temperature and increased air flow rate. The model was shown to be especially valuable in defining the optimal operating strategies for enclosed aisle configurations with fixed and variable server flows, and to elucidate the deleterious effect of temperature nonuniformity at the inlet of the racks on the data center cooling infrastructure power consumption. The analysis shows a potential for as much as an $∼58%$ savings in cooling infrastructure energy consumption by utilizing an optimized enclosed aisle configuration with bypass recirculation, instead of a traditional enclosed aisle, where all the data center exhaust is forced to flow through the computer room air conditioners. The analysis of open-aisle data centers shows that as the temperature at the inlet of the racks becomes more nonuniform, optimal operation tends toward lower recirculation and higher power consumption; again, stressing the importance of providing as uniform a temperature to the racks as possible. It is also revealed that servers with a modest temperature rise $(∼10°C)$ have a wider latitude for cooling infrastructure optimization than those with a high temperature rise $(≥20°C)$, which tend to consume less cooling power when the aisles are enclosed.

<>

Figures

Figure 1

Rack face coordinates

Figure 2

Simplified schematic of an air-cooled datacenter

Figure 3

Optimization results for constant server flow example

Figure 4

(a) Component-by-component energy breakdown for θ=6.0, (b) component-by-component energy breakdown for θ=0.0, and (c) refrigeration system COP for θ=0.0 and 6.0

Figure 5

Effect of server temperature rise on optimized cooling power

Figure 6

(a) Effect of server temperature rise on optimum ψ¯ and (b) effect of server temperature rise on optimum Ta

Figure 7

Typical server flow control algorithm

Figure 8

Simplified φ distribution over the racks’ inlet face

Figure 9

Optimization results for variable server flow example

Figure 10

Relationship between Ta and ψ¯o for variable server flow example

Figure 11

Minimum power consumption for variable flow server example

Figure 12

Optimum ψ¯o and its conjugate Ta for variable server flow example

Figure 13

Relationship between ψ¯ and ψ¯o

Figure 14

Conjugate values of Ta and ψ¯o for enclosed aisles cases (θ=0)

Figure 15

Optimum conditions for enclosed aisles (θ=0) with variable sever flow

Figure 16

Optimum values for enclosed aisle (θ=0) case

Errata

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections