Improved CFD modeling of a small data center test cell

Abdelmaksoud, Waleed A.; Khalifa, H. Ezzat; Dang, Thong Q.; Schmidt, Roger; Iyengar, Madhusudan

doi:10.1109/itherm.2010.5501425

Cited by 36 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead of using fully open raisedfloor tiles as in [32], we improve the model by using perforated floor tiles. As shown by Abdelmaksoud et al [33] that a data center CFD model with perforated floor tiles can significantly improve the model accuracy. Similar to the rack model introduced in Section 4.1, we model each rack in this large-scale data center with four stacked boxes.…”

Section: Detection Probability Maximization On a Large-scale Data Cenmentioning

confidence: 98%

Intelligent Sensor Placement for Hot Server Detection in Data Centers

Wang

Xing

et al. 2013

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Abstract-Recent studies have shown that a significant portion of the total energy consumption of many data centers is caused by the inefficient operation of their cooling systems. Without effective thermal monitoring with accurate location information, the cooling systems often use unnecessarily low temperature set points to overcool the entire room, resulting in excessive energy consumption. Sensor network technology has recently been adopted for data-center thermal monitoring because of its nonintrusive nature for the already complex data center facilities and robustness to instantaneous CPU or disk activities. However, existing solutions place sensors in a simplistic way without considering the thermal dynamics in data centers, resulting in unnecessarily degraded hot server detection probability. In this paper, we first formulate the problems of sensor placement for hot server detection in a data center as constrained optimization problems in two different scenarios. We then propose a novel placement scheme based on computational fluid dynamics (CFD) to take various factors, such as cooling systems and server layout, as inputs to analyze the thermal conditions of the data center. Based on the CFD analysis in various server overheating scenarios, we apply data fusion and advanced optimization techniques to find a near-optimal sensor placement solution, such that the probability of detecting hot servers is significantly improved. Our empirical results in a real server room demonstrate the detection performance of our placement solution. Extensive simulation results in a large-scale data center with 32 racks also show that the proposed solution outperforms several commonly used placement solutions in terms of detection probability.

show abstract

Section: Detection Probability Maximization On a Large-scale Data Cenmentioning

confidence: 98%

Intelligent Sensor Placement for Hot Server Detection in Data Centers

Wang

Xing

et al. 2013

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…Abdelmaksoud et al [22] showed that the effect of buoyancy often should be taken into account in CFD modeling of data centers. Buoyancy is the driving force for natural convection.…”

Section: Additional Simulation Settingsmentioning

confidence: 99%

Computational Fluid Dynamics Modeling and Validating Experiments of Airflow in a Data Center

2018

View full text Add to dashboard Cite

Abstract:The worldwide demand on data storage continues to increase and both the number and the size of data centers are expanding rapidly. Energy efficiency is an important factor to consider in data centers since the total energy consumption is huge. The servers must be cooled and the performance of the cooling system depends on the flow field of the air. Computational Fluid Dynamics (CFD) can provide detailed information about the airflow in both existing data centers and proposed data center configurations before they are built. However, the simulations must be carried out with quality and trust. The k-ε model is the most common choice to model the turbulent airflow in data centers. The aim of this study is to examine the performance of more advanced turbulence models, not previously investigated for CFD modeling of data centers. The considered turbulence models are the k-ε model, the Reynolds Stress Model (RSM) and Detached Eddy Simulations (DES). The commercial code ANSYS CFX 16.0 is used to perform the simulations and experimental values are used for validation. It is clarified that the flow field for the different turbulence models deviate at locations that are not in the close proximity of the main components in the data center. The k-ε model fails to predict low velocity regions. RSM and DES produce very similar results and, based on the solution times, it is recommended to use RSM to model the turbulent airflow data centers.

show abstract

“…Table I The cold air enters the DC through perforated tiles with 25% open area. In order to avoid complicated meshing of detailed perforations in tiles, the simplified yet accurate momentum source method proposed by Abdelmaksoud et al [8], [9] was used for the present CFD model.…”

Section: A the Data Center Room Modelmentioning

confidence: 99%

Can a data center heat-flow model be scaled down?

Fernando

Siriwardana

Halgamuge

2012

2012 IEEE 6th International Conference on Information and Automation for Sustainability

View full text Add to dashboard Cite

Data centers require vast amounts of energy for keeping the servers cool at optimal operating temperatures. Recent research has focused on improving the cooling efficiency, and thereby lowering the energy consumption, through different rack arrangements and modifying the air-flow patterns. Thus far, this has been done using computational fluid dynamics (CFD) models as access to a real data centers is often restricted. The next step in this research is to build a physical model for testing purposes. The viability of building a scaled model of an actual data center is investigated using the scale modeling theory for airflow experiments. A full-scale prototype and a half-scale model are created using CFD software and simulated to see if similarity can be achieved in the scaled model for the temperature distribution as well as the airflow velocities. Our results show that the thermal similarity can be achieved within 5% error margin while the airflow similarity cannot be achieved with reasonable accuracy.

show abstract

Improved CFD modeling of a small data center test cell

Cited by 36 publications

References 14 publications

Intelligent Sensor Placement for Hot Server Detection in Data Centers

Intelligent Sensor Placement for Hot Server Detection in Data Centers

Computational Fluid Dynamics Modeling and Validating Experiments of Airflow in a Data Center

Can a data center heat-flow model be scaled down?

Contact Info

Product

Resources

About