Cold Aisle Research Articles

A robust data-driven model predictive thermal control for rack-based data center

Optimizing and controlling of air-cooled data centers cooling systems are essential for reducing energy consumption. However, traditional cooling system control strategies have been proven inadequate due to the thermodynamic processess complexity and the challenge for accuratly modeling the thermal behavior of data center (DC). Therefore, a data-driven control method whithout explicit modeling process is urgently needed. In this paper, a robust data-driven model predictive control (MPC) method for air-cooled rack-based DCs, based on Willems fundamental lemma, is proposed. The methodology is executed in three phases: Firstly, we show that the system is approximately linear time-invariant (LTI) by zonal model method and experiment. Secondly, we propose a data-driven MPC framework for air-cooled rack-based DC. Thirdly, considering the uncertainties caused by disturbances and the nonlinear dynamics, the data-driven MPC approach is tailored as a robust one. Finally, the proposed approach is verified by computational fluid dynamics simulation experiments. This approach ensures the rack-based DC cold aisle temperature to the reference temperature as well as within acceptable limits. The experiment shows that the steady-state error is guaranteed to be less than 2%, with a control accuracy increased by 2.4% compared with the regular MPC. Thus, this method provides an alternative solution for maintaining the reference temperature in air-cooled rack-based DCs by utilizing only input–output data.

Investigating thermal performance and energy efficiency in under-floor air distribution data centers: A case study

This study conducted on-site measurements of air temperature and airflow velocity at the cold aisle outlet, combined with infrared imaging, to derive optimized layouts for rack servers. It was found that installing blind plates effectively improves the uniformity of cold air distribution within the racks, as indicated by the temperature and airflow velocity at the server inlet and outlet under both blind and non-blind plate conditions. Finally, thermal performance indicators (i.e., SHI, β, RHI, RCI) and Reynolds number were used to assess the thermal performance of the data center, while energy efficiency indicators (i.e., PUE, WUE, CUE) were employed to evaluate its energy efficiency. The results revealed that the cold aisle has less air distribution at both ends and is more concentrated in the middle. It is advisable to position high-power servers in the middle racks and place racks with relatively lower power at the front and rear ends to ensure effective heat dissipation of the servers. The installation of blind plates can effectively enhance the uniformity of airflow within the racks and improve server cooling. Among the tested racks, 16 out of 17 exhibited low temperatures (RCILO) below 90%, indicating heat loss and energy waste. The PUE value of the data center is 1.19, and the CUE value is 0.77.

Effects and optimization of airflow on the thermal environment in a data center

In this research, the escalating energy consumption challenges in data centers are addressed by optimizing airflow organization designs. Through the use of computational fluid dynamics (CFD) simulations, three different airflow strategies were evaluated and improved: underfloor precision air conditioning, inter-column air conditioning, and backplane air conditioning. These cooling systems, which are usually considered in isolation, were compared in a comprehensive manner to get a full picture of their efficiency and effectiveness. The findings reveal that the implementation of cold aisle containment (CAC) or hot aisle containment (HAC) significantly improves air supply efficiency (ASE) and reduces the supply heat index (SHI), leading to a more uniform temperature distribution and enhanced cooling performance. Specifically, the ASE increased from 65.69% to 85.57% and 90.25% for underfloor precision air conditioning and from 71.29% to 92.16% and 92.17% for inter-column air conditioning, with corresponding reductions in SHI. The backplane cooling system offered consistent ambient temperatures throughout the room, eliminating thermal hotspots without the need for aisle containment. This study offers a comparative analysis of different airflow organization schemes, highlighting the benefits of aisle containment in precision and inter-column air conditioning and the suitability of backplane air conditioning for high-density cooling without the need for traditional aisle separation. The results are crucial for informing energy-efficient cooling strategies in data center design and operation.

A Comparative Numerical Study of Effectiveness of Cold Aisle Containment in Data Centers by Varying Rack Porosity Using Computational Fluid Dynamics

Depending on specific needs and workloads, several racks with varied component densities may be used in a data center. As server density increases, porosity decreases, and the opposite is also true. A frequently used method called cold aisle containment separates hot air and cold air flows in data center settings to improve cooling effectiveness. In this paper, the performance of a data center is investigated using computational fluid dynamics, and the influence of porosity on cold aisle containment is evaluated using well-established non-dimensional performance parameters. The value of RTI without containment increased with an increase in porosity and a maximum RTI of 217 was found with a porosity of 0.75. Regardless of the rack number and porosity, containment provides the optimal RTI values. The results indicate that the SHI and RHI values for rack 1 for all porosities, without confinement, are outside of the permissible range and at higher porosities containment has no significant effects on SHI and RHI. RCILO values for racks 2, 3, and 4, with or without containment, fall within the 80-85% range, indicating temperatures below 13°C. RCIHI value is 1 for all cases considered indicating no rack is out of the recommended temperature of 25°C.

Utilizing waste heat from data centers with adsorptive heat transformation – Heat exchanger design and choice of adsorbent

The electricity and water consumption of data centers is growing on a global scale. A shift towards liquid cooled racks in combination with thermally driven cooling can help to reduce the electricity and water demand associated with the necessary heat rejection. To quantify the potential of adsorptive heat transformation devices in reducing the electricity and water demand, the prediction of thermal efficiency, heat flow rates and energy efficiency ratio is required. To this end, a numerical model is newly developed using basic adsorption heat exchanger theory. This model can predict the necessary performance indicators with respect to temperatures and volume flow rates, heat exchanger design and adsorbent. The full performance map of a market-available adsorption chiller (71 points) and own measurements are used for calibration and rigorous validation of the model. An average deviation (experiment vs. calculation) of 8.3 % in terms of thermal efficiency and 7.2 % in terms of heat flow rates is achieved, indicating a very good agreement for a wide range of temperatures. At a moderate liquid cooled rack outlet temperature of 50 °C, a heat rejection temperature of 26 °C and a cold aisle inlet temperature of 18 °C the cooling power of the silica gel reference chiller of 5.3 kW can be increased by 59 % to 8.5 kW at a partial energy efficiency ratio (pumps and control, no fans) of > 20 by assuming MIL-100(Fe) as adsorbent on a flat-tube lamella heat exchanger. The model can be used in subsequent annual system simulations to quantify the savings in electrical power and water consumption, which strongly depend on the ambient conditions.

Research on the distribution characteristics of static pressure in data centers from the perspective of airflow transmission path

A new observation perspective was proposed to manage the airflow distribution in data centers from the floor plenum and the cold aisle as a unified entity, which is different from the current single space only with geometric parameter adjusting in the floor plenum or adding resistance measures in the cold aisle to improve the airflow uniformity only in one-dimensional direction. Based on it, the transfer rule of the static pressure between two continuous space was explored to analyze the transfer rules of the static pressure difference in the two spaces, and find the regulation mechanism of the airflow to achieve uniform airflow distribution of the cold aisle in two-dimensional direction. The results show that decreasting the perforation rates can increase the airflow uniformity effectively at the inlet of the rack along the horizontal direction, but reduce the airflow uniformity at the inlet of the rack along the vertical direction in data center. Increasting the flow rate appropriately as well as larger porosity of the rack perforated plate can improve the airflow uniformity effectively at the inlet of the rack both horizontally and vertically in data center. The conclusions will help guide the improvement and regulation of airflow organisation to reduce the occurrence of local hotspots and ensure safe operation of the data centers.

Experimental evaluation of direct-to-chip cold plate liquid cooling for high-heat-density data centers

Owing to the dramatic increase in IT power density and energy consumption, the data center (DC) sector has started adopting thermally- and energy-efficient liquid cooling methods. This study examines a single-phase direct-to-chip liquid cooling approach for three high-heat-density racks, utilizing two liquid-to-air (L2A) cooled coolant distribution units (CDUs) and a combined total heat load of 128 kW. An experimental setup was developed to test different types of CDUs, cooling loops, and thermal testing vehicles (TTVs) for different operating conditions. IR images and the collected data were used to investigate the effect of air recirculation between cold and hot aisle containments on the CDU’s performance and stability of supply air temperature (SAT). Three different types of cooling loops (X, Y, and Z) were characterized thermally and hydraulically. Results show that Type Y has the lowest cold plate thermal resistance and pressure drop, among others. In a later test that included a single rack at a heat load of 53 kW and a single CDU, the heat capture ratio for fluid was found to be 94%. Experiments show that using blanking panels on the back of the racks limits hot air recirculation and maintains a steady SAT in the cold aisle. Finally, the CDU performance was evaluated at a high heat load for the three racks at 128 kW, and the average cooling capacity of the units is 58.6 kW, and the effectiveness values for CDU 1 and CDU 2 are 0.83 and 0.82, respectively.

Super-resolution-assisted rapid high-fidelity CFD modeling of data centers

Data center thermal management requires a good understanding of critical cooling airflow path. While CFD modeling excels at portraying airflow and temperature fields, it is often computationally intensive. Recent advancement in deep learning offers a rapid prediction of key parameters at several points of interest based on operation conditions of data centers. However, these models fail to deliver the comprehensive flow physics of an entire domain that CFD provides. This paper presents an innovative super-resolution approach, originating from computer vision, to model the air flow and temperature field in the cold aisle of a realistic data center. The proposed model reconstructs a high-fidelity CFD-generated flow field that typically takes several hours by using a low-fidelity CFD-generated flow field which only takes several minutes. The drastic reduction in computational time makes real-time prediction feasible, while providing detailed information for data center engineers to understand the flow field. Two variations of the framework are proposed and compared with the ground truth generated by the high-fidelity CFD model. A Mean Absolute Error less than 0.5 °C is achieved for air temperature prediction across all test scenarios, and less than 0.1 m/s for air velocity prediction. A sensitivity study is also conducted to determine the importance of input features.

BIM and Computational Fluid Dynamics Analysis for Thermal Management Improvement in Data Centres

One of the most energy-intensive facilities requiring a comprehensive and well-optimised cooling system is the data centre. Air containment across the data centre is a key thermal management and energy-saving strategy that enhances the performance of data centres. The majority of modern energy-efficient data centres use some type of air containment. The primary advantage of aisle separation and containment is the decrease in the air temperature at the server inlet by reducing the mixing of hot air with cold air. In order to ascertain the volume of literature relating to corridor insulation, we conducted a literature review. Currently, there have been numerous articles regarding the application of computational fluid dynamics (CFD) analysis, however, publications delineating the integration of building information modelling (BIM) principles for corridor separation are still limited. Research specifically targeting data centre corridor insulation is somewhat limited. As a result of this analysis, the most common methods used to isolate hot or cold aisles within a data centre were identified. To determine the most effective type of corridor insulation, the BIM family was created in Autodesk Revit. The model includes 15 telecom cabinets containing information technology (IT) equipment, eight inter-row air conditioners, and one UPS. The model was used for the CFD analysis of the air temperature in different zones of the room. Visualisation of the results using gradient temperature distributions at different levels provides a complete picture of the microclimate formation in the room and allowed the advantage of the hot aisle isolation scheme to be demonstrated.

Artificial Intelligence-Based Temperature Twinning and Pre-Control for Data Center Airflow Organization

Green and low-carbon has become the main theme of global energy development. Data centers are the core of the digital age, carrying huge arithmetic demand. Data centers must implement green low-carbon energy efficiency management to improve energy efficiency, reduce energy waste and carbon emissions, and achieve sustainable development. As a result, an intelligent management strategy for dynamic energy efficiency of data center networks with Artificial Intelligence (AI) fitting control is proposed. Firstly, a Long Short-Term Memory (LSTM) network is used for long sequence trend prediction to predict the temperature of the data center in the next sequence using the temperature of the past 15 sequences and the power consumption of the equipment as parameters. Then, based on the prediction results, the intelligent air conditioning controller based on Deep Q-Network (DQN) is designed to update the parameters by using the gradient of double-Q network and error backpropagation, and the optimal control action is selected by using the ε-greedy strategy to ensure that the prediction of the hotspot does not occur. Experiments show that the average absolute errors of temperature prediction for supply air, return air, cold aisle as well as hot aisle are 0.32 °C, 0.21 °C, 0.36 °C and 0.19 °C, respectively. The Power Usage Effectiveness (PUE) and Water Usage Effectiveness (WUE) decreased by an average of 2.6% and 2.5%, respectively. The method achieves the purpose of predicting future temperatures and intelligently controlling the output so that the data center can satisfy the premise of normal operation and thus achieve more efficient energy use.

Numerical Investigation of Thermal Performance with Adaptive Terminal Devices for Cold Aisle Containment in Data Centers

The energy consumption of data center cooling systems accounts for a large proportion of total energy consumption. The optimization of airflow organization is one of the most important methods to improve the energy efficiency of cooling systems. The adjustment scale of many current air flow organization methods, however, is too large and does not support the data center’s refined operation. In this paper, a new type of air supply terminal device is proposed, and it could adaptively adjust according to the power of servers in the rack for cold air redistribution. In addition, the corresponding regulation strategy is proposed. A CFD model is established according to field investigation of a real data center in Shanghai to investigate the adjustment range and the energy saving potential of the device. The simulation results indicate that the device can suppress the local hot spots caused by excessive server power to some extent and greatly improve the uniformity of servers exhaust temperature. The case study shows that the device can save energy consumption by 20.1% and 4.2% in mitigating local hot spots compared with reducing supply air temperature and increasing supply air flowrate.

Research on Airflow distribution in Internet Data Center

With the rapid growth of IDC (Internet Data Center) construction, the energy consumption is also increasing gradually. It is extremely important to design a reasonable air distribution to reduce the energy consumption of IDC. The research object is an IDC Plant in Beijing. Airpak software is used to conduct DeST ( Building energy consumption analysis software) simulation of the air flow organization in IDC room. Under the same cooling load conditions, through the simulation of the three types of diffuse air supply, closed cold aisle and closed hot aisle in the IDC room, it is concluded that the air flow organization after the closed cold and hot aisle is excellent. The energy consumption analysis of the enclosure structure of the three air flow organizations of IDC room shows that the energy consumption of the closed cold aisle is lower than that of the closed hot aisle air-conditioning system.

ANALYSIS OF COOLING SYSTEM EFFICIENCY

Introduction: The main task of cooling systems in the machine room of a data center is to maintain optimal air parameters at the inlet to IT equipment in accordance with the applicable regulatory documents. If the value of air temperature at the inlet to IT equipment exceeds the optimal value, then so-called hot spots occur — the operation of IT equipment in such places can lead to its overheating and failure. The main reason for the occurrence of hot spots is the violation of air circulation between the hot and cold aisles. Purpose of the study: We aimed to analyze the influence of different methods of hot aisle isolation on the efficiency of IT equipment cooling. Methods: We performed modeling of different methods of hot aisle isolation at different capacity values of IT equipment in the STAR-CCM+ program. Result: As a result of studying temperature and air conditions in a data center, we found the most rational way of isolating the hot aisle — top horizontal containments with vertical blind panels in the free rack space.

Temperature, Humidity, and Power Outage Monitoring System of Pamapersada Nusantara’s Server Racks

Data center is a vital location that must be managed. The room contains central devices that are essential for operational continuity of the information system. Like the PT Pamapersada Nusantara data center, this room needs to be maintained according to standards. One thing that needs to be maintained is temperature, humidity, and electricity because they have a direct impact on IT equipment and can be potentially fatal if they are not in standard conditions. Meanwhile, the data center being researched uses manual checks for temperature, humidity, and electrical power supply. The problem is the inconsistency of the administrator in documenting these parameters and the lack of responsiveness of repairs if an anomaly occurs in the parameter value, because the data is not real time. This research aims to build a monitoring system for monitoring temperature, humidity, and power outage at the PT Pamapersada Nusantara data center which provides direct notifications to administrators. Current similar studies generally do not include the sensor location factor which is essential. The sensor installation is not randomly placed in the room, nor in the hot aisle or in the cold aisle, but in the server rack so that the temperature and humidity truly represent the surroundings of the device as ASHRAE recommend. Referring to ASHRAE, the standard temperature is 18° C - 27° C while the humidity is 20% - 60%. This research starts from a needs analysis through a survey, followed by equipment design, equipment testing by comparing with manual measuring tools. ESP8266 and DHT22 sensors are used, telegram as system output in the form of notifications. Notifications are sent when the temperature and humidity values are not in standard conditions, they are sent every day at 07.00 and 15.00, and they are sent if there is 5% increasing or decreasing values. The results show that the system successfully sends a notification on telegram when the conditions are met. Tests using comparisons with other measuring instruments also show a deviation of values below 2%. Measurements on the server rack show varying value depending on the occupancy of the rack and its position against the air conditioner

Development of an independent modular air containment system for high-density data centers: Experimental investigation of row-based cooling performance and PUE

This study presents a prototype of an independent modular air containment (MAC) system that overcomes the limitations of the existing room-based cooling and applies a row-based cooling system for a high-density data center that can meet the requirements of energy efficiency. Furthermore, to improve the cooling efficiency and safety, we developed a novel in-row cooling package with multiple heat-transfer media (a sequential water-to-refrigerant-to-air heat exchange system). While complying with the standard test method, the objective cooling performance and PUE are derived through in-situ measurement in connection with the actual operation of a reference data center, and the energy saving contribution is analyzed. As observed from the result of the experimental measurement for the MAC with the novel in-row cooling package, the thermal balance is maintained between the water to refrigerant primary cycle and the refrigerant to air secondary cycle, and the air temperature in the cold aisle converges to 23 °C, satisfying the ASHRAE thermal guidelines. The energy influence of the in-row cooling package is extremely low with average 0.019 pPUE. For a reference data center, the analyzed energy efficiency can be improved from baseline PUE 1.563 to PUE 1.361 through 43.5% reduction in the cooling system.

Experimental analysis of airflow uniformity and energy consumption in data centers

Temperature and airflow uniformity are experimentally investigated in this research study along with energy consumption analysis on the laboratory container size small data center. A raised floor plenum (RFP) data center system of size 6 (L) × 3.1 (W) × 2.3 (H) m with one Computer Room Air Handler (CRAH), two rows and each row contains five racks, five 8U simulator cabinets in each rack was constructed to test partial and closed cold aisle containment (CAC). Seven different cases were investigated with variables of perforated tiles (32%, 50 % & 70%), CRAH fan speed (60 Hz, 42 Hz), chiller return water temperature (CRWT) (15 °C & 17 °C) and heating load (full 50 kW, 75% load 37.5 kW). These cases were further analyzed using Rack Cooling Index (RCI) and Supply Heat Index (SHI) along with power consumption scenarios in terms of Coefficient of Performance (COP) and Power Usage Effectiveness (PUE). The uniqueness of this study is experimental investigations of system COP and PUE analysis for a wide range of important variables for partial and closed CAC. It was found that 50% porous tile provided the best airflow and temperature uniformity for the case of partial containment with full and 25% reduced heat load along with full and 30% reduced CRAH speed respectively. Slight recirculation was observed during the reduced CRAH fan speed case at the corner edge of data racks. However, due to additional supply airflow usage, these recirculation’s were within the acceptable SHI. Reduction of CRWT reduces the total Heating Ventilation and Air Conditioning (HVAC) power consumption and improves the overall system COP.

Multi-objective optimization of supply air jet enhancing airflow uniformity in data center with Taguchi-based grey relational analysis

In most data centers (DCs), hot spots reduce the reliability, durability, and efficiency of server cooling. The hot spots are mainly attributed to the uneven airflow distribution and the resulting mixing of cold and hot air. To evenly distribute the airflow, the relay jet fan, an active local adjustment technology, is introduced into the under-floor air distribution in DCs. The jet fans are set in front of the closed cold aisle where the hot spots most often appear. The nozzle height, nozzle angle, horizontal position and attachment distance greatly influence the performance of the new system. To comprehensively evaluate the multiple levels of these affecting parameters, a large number of cases that need testing make it impossible to conduct research. An approach based on the Taguchi method with grey relational analysis optimizes the system and the L16 (44) orthogonal array is employed to design the experiments. The Taguchi method combined with computational fluid dynamics determines the optimal combinations of the Supply Heat Index (SHI), Return Temperature Index (RTI), β, and Index of Mixing (IOM). The significance of the affecting parameters is revealed and ranked, and through the main effect analysis and the analysis of variance, the nozzle angle is chosen as the most important among the four affecting parameters. After that, the Taguchi-based grey relational analysis transforms the multi-objective into a single-objective to determine the optimal combination of design variables. Finally, a confirmatory test was conducted, and the optimal combination reduces the server temperature by up to 7.2 °C.

Numerical and experimental investigations on thermal management for data center with cold aisle containment configuration

This study proposes the container data center with the featured cold aisle containment (CAC) as effective thermal control strategy. In design, the overhead downward flow system is implemented with a heat exchanger arranged right above the data center on the air side and an evaporative water chiller on the water side to form the cooling approach. The cold airflows and hot exhausts of racks are separately transported by the contained cold and hot aisles to alleviate the problem of cold and hot air mixing. The measurements of air temperature and velocity of racks are used to validate the prediction accuracy of the computational fluid dynamics (CFD) model. The performance metrics in terms of the rack cooling index (RCI), return temperature index (RTI), supply heat index (SHI) are used to examine the design effectiveness of the proposed test data center. The simulations are then extended to assess the air distribution and thermal management at varied supply air temperatures and velocities for a large-scale data center to be built in the green energy technology demonstration site of the Shalun smart green energy science city. Overall, the calculated average PUE of 1.38 for the large-scale data center is notably less than the average PUE of 1.59 from the results of 2020 data center industry survey, indicating the potential savings of cooling energy and cost. This paper demonstrates a generalized approach as an easily adaptable, cost-effective solution for data centers to be deployed in tropical and subtropical areas.

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.

Shifting to energy efficient hybrid cooled data centers using novel embedded floor tiles heat exchangers

With the remarkable surge in IT power densities and energy consumption, the data center industry is starting to adopt energy and thermally-efficient single-phase cooling technologies. For many legacy data centers worldwide, the lack of a primary chilled water source is a substantial barrier to adopting single-phase technologies. In this study, the necessity of having a facility chilled water source for deploying liquid cooling solutions is eliminated by introducing a solution that does not require bringing in any modification of the data center layout. This simple solution proposes integrating liquid to air heat exchangers with the floor tiles to eject heat from the liquid-cooled IT equipment into the room air. An experimentally validated CFD modeling approach is used to investigate the performance of these heat exchangers. Results show that by installing two heat exchangers in different cold aisles, a significant amount of cooling load is handled at the design point conditions. When they are tested under different operational conditions, the cooling capacity of these heat exchangers is shown to be sensitive to the supply air temperature. To overcome the operational constraints stemming from the supply air temperature, an alternative configuration with just one heat exchanger installed in the hot aisle is considered and discussed. Results indicate that the second configuration shows superior performance to the first one in terms of sensitivity to supply air temperature and cooling capacity. Lastly, with this configuration, the data center utilizes energy extremely efficiently and achieves a power usage effectiveness of 1.06.