Ed Walsh

From Chip to Cooling Tower Data Center Modeling: Influence of Server Inlet Temperature and Temperature Rise Across Cabinet

To achieve reductions in the power consumption of the data center cooling infrastructure, the current strategy in data center design is to increase the inlet temperature to the rack, while the current strategy for energy-efficient system thermal design is to allow increased temperature rise across the rack. Either strategy, or a combination of both, intuitively provides enhancements in the coefficient of performance of the data center in terms of computing energy usage relative to cooling energy consumption. However, this strategy is currently more of an empirically based approach from practical experience, rather than a result of a good understanding of how the impact of varying temperatures and flow rates at rack level influences each component in the chain from the chip level to the cooling tower. The aim of this paper is to provide a model to represent the physics of this strategy by developing a modeling tool that represents the heat flow from the rack level to the cooling tower for an air cooled data center with chillers. This model presents the performance of a complete data center cooling system infrastructure. After detailing the model, two parametric studies are presented that illustrate the influence of increasing rack inlet air temperature, and temperature rise across the rack, on different components in the data center cooling architecture. By considering the total data center, and each components influence on the greater infrastructure, it is possible to identify the components that contribute most to the resulting inefficiencies in the heat flow from chip to cooling tower and thereby identify the components in need of possible redesign. For the data center model considered here it is shown that the strategy of increasing temperature rise across the rack may be a better strategy than increasing inlet temperature to the rack.

From chip to cooling tower data center modeling: Part I Influence of server inlet temperature and temperature rise across cabinet

To achieve reductions in the power consumption of the data center cooling infrastructure, the current strategy in data center design is to increase the inlet temperature to the rack, while the current strategy for energy-efficient system thermal design is to allow increased temperature rise across the rack. Either strategy, or a combination of both, intuitively provides enhancements in the coefficient of performance (COP) of the data center in terms of computing energy usage relative to cooling energy consumption. However, this strategy is currently more of an empirically based approach from practical experience, rather than a result of a good understanding of how the impact of varying temperatures and flow rates at rack level influences each component in the chain from the chip level to the cooling tower. The aim of this paper is to provide a model to represent the physics of this strategy by developing a modeling tool that represents the heat flow from the rack level to the cooling tower for an air cooled data center with chillers. This model presents the performance of a complete data center cooling system infrastructure. After detailing the model, two parametric studies are presented that illustrate the influence of increasing rack inlet air temperature, and temperature rise across the rack, on different components in the data center cooling architecture. By considering the total data center, and each components influence on the greater infrastructure, it is possible to identify the components that contribute most to the resulting inefficiencies in the heat flow from chip to cooling tower and thereby identify the components in need of possible redesign. For the data center model considered here it is shown that the strategy of increasing temperature rise across the rack may be a better strategy than increasing inlet temperature to the rack. Part II of this work expands on this paper with further parametric studies to evaluate the robustness of this data center cooling strategy, with conditions for optimal strategy deployment.

An Experimental Study on the Design of Miniature Heat Sinks for Forced Convection Air Cooling

An experimental study is performed on one of the smallest commercially available miniature fans, suitable for cooling portable electronic devices, used in conjunction with both finned and finless heat sinks of equal exterior dimensions. The maximum overall footprint area of the cooling solution is 534 mm 2 with a profile height of 5 mm. Previous analysis has shown that due to fan exit angle, flow does not enter the heat sinks parallel to the fins or bounding walls. This results in a nonuniform flow rate within the channels of the finned and finless heat sinks along with impingement of the flow at the entrance giving rise to large entrance pressure losses. In this paper straightening diffusers were attached at the exit of the fan, which resulted in aligning the flow entering the heat sinks with the fins and channel walls. Detailed velocity measurements were obtained using particle image velocimetry, which provided a further insight into the physics of the flow in such miniature geometries and in designing the straightening diffusers. The thermal analysis results indicate that the cooling power of the solution is increased by up to 20% through the introduction of a diffuser, hence demonstrating the need for integrated fan and heat sink design of low profile applications.

Thermal Management of Low Profile Electronic Equipment Using Radial Fans and Heat Sinks

There is an increasing need for low profile thermal management solutions for applications in the range of 5-10 W, targeted at portable electronic devices. This need is emerging due to enhanced power dissipation levels in portable electronics, such as mobile phones, portable gaming machines, and ultraportable personal computers. This work focuses on the optimization of such a solution within the constraints of the profile and footprint area. A number of fan geometries have been investigated where both the inlet and exit rotor angles are varied relative to the heat conducting fins on a heat sink. The ratio of the fan diameter to the heat sink fin length was also varied. The objective was to determine the optimal solution from a thermal management perspective within the defined constraints. The results show a good thermal performance and highlight the need to develop the heat sink and fan as an integrated thermal solution rather than in isolation as is the traditional methodology. An interesting finding is that the heat transfer scales are in line with turbulent rather than laminar correlations despite the low Reynolds number. It is also found that while increasing the pumping power generally improves the thermal performance, only small gains are achieved for relatively large pumping power increases. This is important in optimizing portable systems where reduced power consumption is a competitive advantage in the marketplace.

Characterizing convective heat transfer using infrared thermography and the heated-thin-foil technique

Convective heat transfer, due to axial flow fans impinging air onto a heated flat plate, is investigated with infrared thermography to assess the heated-thin-foil technique commonly used to quantify two-dimensional heat transfer performance. Flow conditions generating complex thermal profiles have been considered in the analysis to account for dominant sources of error in the technique. Uncertainties were obtained in the measured variables and the influences on the resultant heat transfer data are outlined. Correction methods to accurately account for secondary heat transfer mechanisms were developed and results show that as convective heat transfer coefficients and length scales decrease, the importance of accounting for errors increases. Combined with flow patterns that produce large temperature gradients, the influence of heat flow within the foil on the resultant heat transfer becomes significant. Substantial errors in the heat transfer coefficient are apparent by neglecting corrections to the measured data for the cases examined. Methods to account for these errors are presented here, and demonstrated to result in an accurate measurement of the local heat transfer map on the surface.

Effect of geometric scaling on aerodynamic performance

Miniaturization of modern electronics and simultaneous elevation in heat dissipation means that future compact electronic systems are likely to be too hot to be held in the users hand. As a result, novel compact cooling technologies are required. In systems such as mobile phones and palmtop computers, macroscale fans cannot be used to overcome this problem because they are too large. As a solution, the implementation of microfan technology is proposed. Aerodynamic scaling issues in microaxial flow fans are addressed. Analysis shows how reduction of fan dimensions to the microscale causes increased local loss. Numerical simulations were performed to investigate the validity of the scaling theory, the results of which give confidence in the scaling analysis. Measurements were carried out on two geometrically similar fans to validate the theory under experimental conditions. Results of these measurements were in good agreement with the analysis. The fundamental finding is that a reduction in scale is accompanied by a reduction in efficiency and, thus, fan performance. It is concluded that geometric scaling alone of macroscale designs is not sufficient to produce microscale cooling fans: Modifications must be made to the geometry that account for changes in flow physics.

From chip to cooling tower data center modeling: Part II Influence of chip temperature control philosophy

The chiller cooled data center environment consists of many interlinked elements that are usually treated as individual components. This chain of components and their influences on each other must be considered in determining the benefits of any data center design and operational strategies seeking to improve efficiency, such as temperature control fan algorithms. Using the models developed in part I of this work, this paper extends the analysis to include the electronics within the rack through considering the processor heat sink temperature. This has allowed determination of the influence of various cooling strategies on the data center coefficient of performance. The strategy of increasing inlet aisle temperature is examined in some detail and found not to be a robust methodology for improving the overall energy performance of the data center, while tight temperature controls at the chip level consistently provides better performance, yielding more computing per watt of cooling power. These findings are of strong practical relevance for the design of fan control algorithms at the rack level and general operational strategies in data centers. Finally, the impact of heat sink thermal resistance is considered and the potential data center efficiency gains from improved heat sink designs are discussed.

A Novel Approach to Low Profile Heat Sink Design

This paper discusses the importance of developing cooling solutions for low profile devices. This is addressed with an experimental and theoretical study on forced convection cooling solution designs that could be implemented into such devices. Conventional finned and corresponding finless designs of equal exterior dimensions are considered for three different heat sink profiles ranging from 1 mm to 4 mm in combination with a commercially available radial blower. The results show that forced convection heat transfer rates can be enhanced by up to 55% using finless designs at low profiles with relatively small footprint areas. Overall, this paper provides optimization and geometry selection criteria, which are relevant to designers of low profile cooling solutions. DOI: 10.1115/1.4001626

The performance of active cooling in a mobile phone

Power dissipation levels in mobile electronics devices are heading towards five watts and above. With this power dissipation level, products such as mobile phones will require active cooling to ensure that the devices operate within an acceptable temperature envelop from both user comfort and reliability perspectives. To the authors knowledge no studies to date have been carried out to determine the potential performance of fans within mobile phone architectures. In this paper a centrifugal fan is implemented into a Nokia mobile phone. Its performance is compared in terms of aerodynamic characteristics, maximum phone surface temperature, and allowable phone heat dissipation, for various levels of blockage in the phone, which are simulated using perforated plates with varying porosity. The results show that for the best case scenario, with minimal blockage increased power dissipation levels of order 75% can be achieved but with realistic blockages this value is more likely to be in the region of 50%.

From Chip to Cooling Tower Data Center Modeling: Chip Leakage Power and Its Impact on Cooling Infrastructure Energy Efficiency

The power consumption of the chip package is known to vary with operating temperature, independently of the workload processing power. This variation is commonly known as chip leakage power, typically accounting for ~10% of total chip power consumption. The influence of operating temperature on leakage power consumption is a major concern for the information technology (IT) industry for design optimization where IT system power densities are steadily increasing and leakage power expected to account for up to ~50% of chip power in the near future associated with the reducing package size. Much attention has been placed on developing models of the chip leakage power as a function of package temperature, ranging from simple linear models to complex super-linear models. This knowledge is crucial for IT system designers to improve chip level energy efficiency and minimize heat dissipation. However, this work has been focused on the component level with little thought given to the impact of chip leakage power on entire data center efficiency. Studies on data center power consumption quote IT system heat dissipation as a constant value without accounting for the variance of chip power with operating temperature due to leakage power. Previous modeling techniques have also omitted this temperature dependent relationship. In this paper, we discuss the need for chip leakage power to be included in the analysis of holistic data center performance. A chip leakage power model is defined and its implementation into an existing multiscale data center energy model is discussed. Parametric studies are conducted over a range of system and environment operating conditions to evaluate the impact of varying degrees of chip leakage power. Possible strategies for mitigating the impact of leakage power are also illustrated in this study. This work illustrates that when including chip leakage power in the data center model, a compromise exists between increasing operating temperatures to improve cooling infrastructure efficiency and the increase in heat load at higher operating temperatures due to leakage power.