Raised Floor Data Center Research Articles

Performance evaluation and modeling of active tile in raised-floor data centers: An empirical study on the single tile case

Raised-floor data centers usually suffer from the local hotspots resulted from uneven cool air delivery. These hotspots not only degrade server performance, but also threat equipment reliability. The commonly used industrial practice of increasing the Computer Room Air Conditioner (CRAC) blower speed for removing hotspots is energy inefficient and may lead to overcooling of some servers. In this paper, we explore the potential of active tiles in data center cooling management. In particular, we deploy a prototype of active tile in a production data center and conduct extensive experiments to investigate the cooling performance. It is shown that deploying the active tiles with even 10% fan speed increases the tile flow by 49%, and sealing the under-rack gap reduces the rack bottom temperature by up to 6°C. Moreover, three machine learning techniques, i.e., Gaussian Process Regression (GPR), Artificial Neural Network (ANN), and Multivariate Linear Regression (MLR) are employed to construct end-to-end data-driven thermal models for the active tile. Using field measured data as training and testing data sets, it is concluded that GPR and ANN are competent for accurate thermal modeling of active tiles. Specifically, GPR achieves the smallest prediction error which is around 0.3°C.

Optimization on jet‐induced ventilation to enhance the uniformity of airflow distribution in data center

AbstractImproving the utilization efficiency of cold airflow in data center (DC) has already attracted widespread concern. A broad consensus has been reached that cold/hot‐aisle containment technologies can reduce the mixture of cold and hot air to weaken the overheating phenomenon of the rack in a certain extent. However, local hot spots still occur in the front racks due to lower tile flow rate at the entrance of the cold aisle, which means the thermal environment of front racks can be further improved to achieve the more uniform airflow distribution horizontally and vertically in DC. In this paper, an innovative method of airflow optimization applying jet fans in the cold aisles is proposed to make up the lower tile flow rate, adjust the flow path of cold air from the perforated tiles to racks, and balance temperature heterogeneity. In addition, inductive velocity, nozzle height, horizontal position, and attachment distance of jet fans are optimized to explore the optimal parameters. Results show that the incorporation of jet‐induced ventilation can effectively improve the cooling performance and overall thermal environment in DC, while the amount of IT equipment that exceeded the ASHRAE recommended supply air temperature (SAT) is reduced by about 38%. The jet fan with optimal parameters has a broad application space in the raised floor DC.

Intelligent Rack-Level Cooling Management in Data Centers With Active Ventilation Tiles: A Deep Reinforcement Learning Approach

The raised-floor architecture is widely adopted in the current data center industry. Due to the structural design and workload imbalance, many server racks suffer from a nonuniform inlet temperature in raised-floor data centers. This compromises not only the cooling performance, but also the computing capacity and system reliability. In this article, we employ the active ventilation tiles (AVTs), i.e., ordinary ventilation tiles with attached fans, to enhance the local cold air delivery and improve the cooling performance. We propose an AVT control algorithm adapted from the recently developed deep reinforcement learning (DRL) techniques to tackle the complex data center environment and thermodynamics. Different from previous studies where the algorithm required extensive pretraining before deployment, our algorithm is fully online. In order to improve the sample efficiency and accelerate the learning speed, we integrate the Dyna architecture to take advantage of the experienced system transitions. We also leverage the idea of shared reward and fingerprint to encourage the cooperation for multi-AVT control. The performance of the proposed solution is then evaluated by deploying a prototype implementation in a production data center. Experimental results reveal that our solution significantly improves the rack inlet temperature distribution.

Numerical Investigation for the Arrangement of Perforated Tile under a Non-uniform Heat Source Condition in a Raised-floor Data Center

The research object of this paper was a typical raised-floor data center where the heat load of each rack was non-uniform. The non-uniform heat source caused that the temperature field in the room was also non-uniform and part of the rack outlet temperature was relatively high. In view of this problem, a numerical simulation was conducted to investigate the effect of the arrangement of perforated tile on the air distribution and the temperature field. The results indicate that a better arrangement of perforated tile could optimize the air distribution and make the temperature field in the room more uniform. The maximum rack outlet temperature was reduced by about 2.1°C under the proper condition of the tiles opening ratio, which means that the energy efficiency of the cooling system was accordingly improved. The adjustment of the tiles opening ratio could be a solution to the non-uniform heat source problem in a raised-floor data center.

Comparing Performance Metrics of Partial Aisle Containments in Hard Floor and Raised Floor Data Centers Using CFD

In data centers, efficient cooling systems are required to both keep the energy consumption as low as possible and to fulfill the temperature requirements. The aim of this work is to numerically investigate the effects of using partial aisle containment between the server racks for hard and raised floor configurations. The computational fluid dynamics (CFD) software ANSYS CFX was used together with the Reynolds stress turbulence model to perform the simulations. Velocity measurements in a server room were used for validation. Boundary conditions and the load of each rack were also retrieved from the experimental facility, implying an uneven load between the racks. A combination of the performance metrics Rack Cooling Index (RCI), Return Temperature Index (RTI) and Capture Index (CI) were used to evaluate the performance of the cooling systems for two supply flow rates at a 100% and 50% of operating condition. Based on the combination of performance metrics, the airflow management was improved in the raised floor configurations. With the supply flow rate set to operating conditions, the RCI was 100% for both raised floor and hard floor setups. The top- or side-cover fully prevented recirculation for the raised floor configuration, while it reduced the recirculation for the hard floor configuration. However, the RTI was low, close to 40% in the hard floor case, indicating poor energy efficiency. With the supply flow rate decreasing with 50%, the RTI increased to above 80%. Recirculation of hot air was indicated for all the containments when the supply rate was 50%, but the values of RCI still indicated an acceptable performance of the cooling system.

Impact of Tile Design on the Thermal Performance of Open and Enclosed Aisles

In raised floor data centers, tiles with high open area ratio or complex understructure are used to fulfill the demand of today's high-density computing. Using more open tiles reduces the pressure drop across the raised floor with the potential advantages of increased airflow and lower noise. However, it introduces the disadvantage of increased nonuniformity of airflow distribution. In addition, there are various tile designs available on the market with different opening shapes or understructures. Furthermore, a physical separation of cold and hot aisles (containment) has been introduced to minimize the mixing of cold and hot air. In this study, three types of floor tiles with different open area, opening geometry, and understructure are considered. Experimentally validated detail models of tiles were implemented in computational fluid dynamics (CFD) simulations to address the impact of tile design on the cooling of information technology (IT) equipment in both open and enclosed aisle configurations. Also, impacts of under-cabinet leakage on the IT equipment inlet temperature in the provisioned and under-provisioned scenarios are studied. In addition, a predictive equation for the critical under-provisioning point that can lead to a no-flow condition in IT equipment with weaker airflow systems is presented. Finally, the impact of tile design on thermal performance in a partially enclosed aisle with entrance doors is studied and discussed.

Experimentally Validated Computational Fluid Dynamics Model for Data Center With Active Tiles

Abstract This paper presents an experimentally validated room-level computational fluid dynamics (CFD) model for raised-floor data center configurations employing active tiles. Active tiles are perforated floor tiles with integrated fans, which increase the local volume flow rate by redistributing the cold air supplied by the computer room air conditioning (CRAC) unit to the under-floor plenum. The numerical model of the data center room consists of one cold aisle with 12 racks arranged on both sides and three CRAC units sited around the periphery of the room. The commercial CFD software package futurefacilities6sigmadcx is used to develop the model for three configurations: (a) an aisle populated with ten (i.e., all) passive tiles; (b) a single active tile and nine passive tiles in the cold aisle; and (c) an aisle populated with all active tiles. The predictions from the CFD model are found to be in good agreement with the experimental data, with an average discrepancy between the measured and computed values for total flow rate and rack inlet temperature less than 4% and 1.7 °C, respectively. The validated models were then used to simulate steady-state and transient scenarios following cooling failure. This physics-based and experimentally validated room-level model can be used for temperature and flow distributions prediction and identifying optimal number and locations of active tiles for hot spot mitigation in data centers.

Air Flow Measurement and Management for Improving Cooling and Energy Efficiency in Raised-Floor Data Centers: A Survey

Recently, the rapid growth in both data center power density and scale poses great challenges to the cooling system. On one hand, the data center operators try to over provision cooling resources for fear of server failures induced by accumulated heat. On the other hand, they also want to reduce the energy cost as the cooling system takes up a significant portion of overall energy consumption. Among all available cooling solutions, air cooling dominates the data center industry due to its simpleness. However, its cooling efficiency has been questioned due to the low air density and specific heat. In this paper, we provide an overview for current endeavors to improve the air cooling efficiency. We group existing researches according to the locations where they can be applied from the perspective of air flow cycle. We also discuss the thermal measurement issues. We hope this paper can help researchers and engineers to design and control their data center air cooling systems.

Studying the fan-assisted cooling using the Taguchi approach in open and closed data centers

This research examined the application of fan-assisted cooling within open and contained raised-floor data centers with the intention of assessing the effectiveness of the system’s cooling method. A computational fluid dynamics (CFD) approach that incorporated both airflow and thermal analysis was employed to examine the detailed effectiveness of fan-assisted air-cooling in a data center that exhibited the behavior of under-cabinet leakage. The CFD technique was also used to assess the impact that three distinct aspects had on the performance of the fan-assisted air-cooling system: cold aisle containment, straightened air flow and fan-to-tile distance. In addition, the mixed-level Taguchi statistical approach was applied to understand how variations in these parameters directly and interactively modified the heat transfer in the surroundings of the inlets of the server racks. The results of the research reveal that fan-assisted perforations, taking the combination of the key factors (the main and interact effects) into account, would represent a viable means through which the cooling systems employed in both open and closed aisle data centers can be enhanced. The well-constructed CFD-based Taguchi methodology might yield a good level of validity for use in substantial and timely optimization procedures to further improve the cooling effectiveness and efficiencies of data centers.

On the characteristics of airflow through the perforated tiles for raised-floor data centers

Perforated tiles are widely used in raised-floor data centers. The cooling air across the perforated tiles must be distributed properly to adequately cool the equipment. This paper aims at investigating the dependence of the pressure loss coefficient across perforated tiles with respect to the geometrical factors and flow parameters. Numerical simulations of flow distribution of perforated tiles were performed for different pore-type and flow parameters. The pressure loss coefficient upon the Reynolds number, the tile area, the porosity, the diameter of the hole, the thickness of the hole and the arrangement is studied. The fitting equation was provided to calculate the pressure loss coefficient. And experimental tests were taken to verify the validity of the model, comparison with the other empirical equations is provided. Based on the comparison, the fitting equation is more applicable to study the pressure loss coefficient for perforated tiles in data center.

Measurement of Air Flow Rate Through Perforated Floor Tiles in a Raised Floor Data Center

In a raised floor data center, cold air from a pressurized subfloor plenum reaches the data center room space through perforated floor tiles. Presently, commercial tool “Flow Hood” is used to measure the tile air flow rate. Here, we will discuss the operating principle and the shortcomings of the commercial tool and introduce two other tile air flow rate measurement tools. The first tool has an array of thermal anemometers (named as “Anemometric Tool”), and the second tool uses the principle of temperature rise across a known heat load to measure the tile air flow rate (named as “Calorimetric Tool”). The performance of the tools is discussed for different types of tiles for a wide range of tile air flow rates. It is found that the proposed tools result in lower uncertainty and work better for high porosity tiles, as compared to the commercial tool.

Numerical cooling performance evaluation of fan-assisted perforations in a raised-floor data center

Raised floors with hot-aisle/cold-aisle configurations have grown to be a critical arrangement for data center cooling. However, more sophisticated techniques are still desired to achieve proper thermal management for improving energy effectiveness and efficiency. The application of fan-assisted perforations may yield a smarter cooling solution accommodating non-uniform power densities and heat load demands to optimize the cooling performance in raised-floor data centers. The present research includes the construction and thermal analysis of a basic raised floor with compact fan assisted perforations modeled using computational fluid dynamics (CFD). The flow straightening effect of the fan-assisted tile and the fan-to-tile distance were treated as the critical design aspects under investigation. The field cooling performance affected by the parametric variation were carefully discussed and compared. Moreover, a full factorial design was employed to obtain and verify the qualitative and quantitative characteristics of the main and interaction effects of the design variation. Well-constructed fan-assisted perforations may satisfy an advanced cooling solution to better manage and optimize the heat and mass transfer in data centers.

A Brief Overview of Recent Developments in Thermal Management in Data Centers

Data centers are mission critical facilities that typically contain thousands of data processing equipment, such as servers, switches, and routers. In recent years, there has been a boom in data center usage, leading their energy consumption to grow by about 10% a year continuously. The heat generated in these data centers must be removed so as to prevent high temperatures from degrading their reliability, which would cost additional energy. Therefore, precise and reliable thermal management of the data center environment is critical. This paper focuses on recent advancements in data center modeling and energy optimization. A number of currently available and developmental thermal management technology in data centers are broadly reviewed. Computational fluid dynamics (CFD) for raised-floor data centers, experimental measurements, containment systems, economizer cooling, hybrid cooling, and device level cooling are all thoroughly reviewed. The paper concludes with a summary and presents areas of potential future research, which are based on the holistic integration of workload prediction and allocation, and thermal management using smart control systems.

Experimental Investigation of Air Flow Through a Perforated Tile in a Raised Floor Data Center

Raised floor data centers supply cold air from a pressurized plenum to the server racks through perforated floor tiles. Hence, the design of an efficient air delivery scheme requires better understanding of the flow features, through and above the perforated tiles. Different tiles with circular pores in a staggered arrangement and with the same thickness are considered. Tile sheet porosities of 23% and 40%, air flow rates of 0.56 m3/s (1177 CFM) and 0.83 m3/s (1766 CFM), and pore sizes of 3.18 mm (1/8 in.) and 6.35 mm (1/4 in.) are investigated. Tiles with 38.1 mm (1.5 in.) region blocked along the edges is compared to the base case with 12.7 mm (0.5 in.) blocked edges. Width reduced to 0.46 m (1.5 ft) from standard width of 0.61 m (2 ft) is also examined. Reduced tile width is used to simulate 0.91 m (3 ft) cold aisle instead of standard 1.22 m (4 ft) cold aisle, with potential to save floor space. A case where the rack is recessed by 76.2 mm (3 in.) from the tile edge is also included in the investigation, as there is a possibility of having racks nonadjacent to the tile edges. Particle image velocimetry (PIV) technique is used to characterize the flow field emerging from a perforated tile and entering the adjacent rack. Experiments suggest that lower tile porosity significantly increases cold air bypass from the top, possibly due to higher air jet momentum above the tile, as compared to a tile with higher porosity. For the air flow rates investigated here, the flow field was nearly identical and influence of flow rate was nondistinguishable. The influence of pore size was non-negligible, even when the porosity and flow rate for the two cases were same. Larger blockage of the tile edges resulted in higher cold air bypass from the top. Reduction in the tile width showed improved air delivery to the rack with considerably reduced cold air bypass. Recessing the rack did not affect the flow field significantly.

Rack Level Modeling of Air Flow Through Perforated Tile in a Data Center

Effective air flow distribution through perforated tiles is required to efficiently cool servers in a raised floor data center. We present detailed computational fluid dynamics (CFD) modeling of air flow through a perforated tile and its entrance to the adjacent server rack. The realistic geometrical details of the perforated tile, as well as of the rack are included in the model. Generally, models for air flow through perforated tiles specify a step pressure loss across the tile surface, or porous jump model based on the tile porosity. An improvement to this includes a momentum source specification above the tile to simulate the acceleration of the air flow through the pores, or body force model. In both of these models, geometrical details of tile such as pore locations and shapes are not included. More details increase the grid size as well as the computational time. However, the grid refinement can be controlled to achieve balance between the accuracy and computational time. We compared the results from CFD using geometrical resolution with the porous jump and body force model solution as well as with the measured flow field using particle image velocimetry (PIV) experiments. We observe that including tile geometrical details gives better results as compared to elimination of tile geometrical details and specifying physical models across and above the tile surface. A modification to the body force model is also suggested and improved results were achieved.

A Comparison of Parametric and Multivariable Optimization Techniques in a Raised-Floor Data Center

It is well known that the flow distribution in data centers can be effected by a variety of parameters such as rack and computer room air conditioning (CRAC) positions, raised-floor height, ceiling height, and percentage opening of perforated tiles. In the present paper, numerical simulations are conducted to optimize the layout of a raised-floor data center with respect to these parameters. Two different approaches have been used: parametric optimization; and a multivariable approach using response surface optimization. In the parametric optimization procedure, the data center is optimized with respect to the maximum temperature in the room. While in the multivariable approach, a cost function is constructed from all the rack inlet temperatures and is minimized. The results show that the multivariable approach is computationally economical and the optimized layout gives a better thermal performance compared to that of parametric optimization.

Modeling Strategies for Air Flow Through Perforated Tiles in a Data Center

Raised-floor data centers supply cold air to server racks through perforated tiles; hence, designing an efficient air delivery scheme requires better understanding of flow features through the tiles. Computational fluid dynamics (CFD) have been established as an important tool for examining the overall flow fields in data centers. The generally used model for air flow through the perforated tiles, the porous jump model, specifies a step pressure loss across the tile surface without affecting the velocity field across the tile. A recently proposed improvement in the porous jump model, the body force model, includes an additional momentum source above the tile surface to account for the acceleration of the flow through the pores. In both these models, geometrical details of the tile such as pore size, location, and shape are not included, which results in a significantly lower computational effort. However, to improve the solution accuracy, it is important to consider the geometrical details due to complex flow behavior through the tiles, such as flow acceleration through pores, jet–jet interactions, and downstream flow development, which simplified models may fail to capture. In this paper, we present a systematic approach to perforated tile flow modeling, with focus on consideration of geometrical details of the tile. First, different turbulence models are compared for the prediction of flow field through a sharp-edged orifice that is representative of a single pore of the tile. Comparison with the published experimental results suggests that a realizable $k{\hbox{-}}\varepsilon$ model is more appropriate for modeling flow through pores, as compared to the generally used standard $k{\hbox{-}}\varepsilon$ model. Flow characteristics through a single pore are analyzed next, and the balance between solution accuracy and grid coarsening is discussed. Subsequently, a full-scale tile model is presented, and it is observed that the elimination of plenum from the model significantly alters the flow field prediction. Body force model is compared with the full-scale tile model, and it shows promise for capturing the major features of flow-through tiles, with scope for further improvement.

A Study of Air Flow through Perforated Tile for Air Conditioning System in Data Center

The power trend of using server systems in data center is continuously increasing. Cooling system consumed 38% of total energy usage which is a significant contribution in the energy consumption. As a result, the efficient energy usage in data center is concerned.Normally a raised-floor is widely used in data center cooling system which delivers cool air through perforated tiles to a front of server racks. However it is usually found that this cool air cannot effectively remove a heat dissipation at the top of server racks. Therefore, the environmental condition in data center must be designed strictly to avoid disruption that caused by overheat.This paper gives some guidelines to help in the better design. Commercial Computational Fluid Dynamics (CFD) program was used to analyze the air flow from raised-floor air conditioning system. A tetrahedral of 1.8 million meshes with k- turbulence model were used to obtain the air flow velocity and temperature distributions in the room. The model was validated by comparing simulation results with actual measurements. As a result, dimensionless parameters in the form of supply heat index(SHI), for understanding the optimization of relative airflow distribution to the heat load of server rack was found. It shows that these parameters provide an effective tool to the improvement of energy efficiency in raised floor data center.

Optimization of Enclosed Aisle Data Centers Using Bypass Recirculation

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 is used to evaluate parametrically the total energy consumption of the data center cooling infrastructure for data centers that utilize aisle containment. The analysis highlights the importance of reducing the total power required for moving the air within the computer room air conditioners (CRACs), the plenum, and the servers, rather than focusing primarily or exclusively on reducing the refrigeration system’s power consumption. In addition, the benefits of introducing a bypass recirculation branch in enclosed aisle configurations are shown. The analysis shows a potential for as much as a 60% savings in cooling infrastructure energy consumption by utilizing an optimized enclosed aisle configuration with bypass recirculation, instead of a traditional enclosed aisle in which all the data center exhaust is forced to flow through the CRACs. Furthermore, computational fluid dynamics is used to evaluate practical arrangements for implementing bypass recirculation in raised floor data centers. A configuration where bypass tiles, with controllable low-lift fans, are placed close to the discharge of CRACs results in increased mixing and is shown to be a suitable method for providing nearly thermally uniform conditions to the inlet of the servers in an enclosed cold aisle. Other configurations of bypass implementation are also discussed and explored.

Numerical Modeling of Perforated Tile Flow Distribution in a Raised-Floor Data Center

This paper is centered on quantifying the effect of computer room and computer room air conditioning (CRAC) unit modeling on the perforated tile flow distribution in a representative raised-floor data center. Also, this study quantifies the effect of plenum pipes and perforated tile porosity on the operating points of the CRAC blowers, total CRAC air flow rate, and its distribution. It is concluded that modeling the computer room, the CRAC units, and/or the plenum pipes could make an average change of up to 17% in the tile flow rates with a maximum of up to 135% for the facility with 56% open tiles while the average and maximum changes for the facility with 25% open tiles are 6% and 60%, respectively.