scholarly journals A Machine Learning Solution for Data Center Thermal Characteristics Analysis

Energies ◽  
2020 ◽  
Vol 13 (17) ◽  
pp. 4378
Author(s):  
Anastasiia Grishina ◽  
Marta Chinnici ◽  
Ah-Lian Kor ◽  
Eric Rondeau ◽  
Jean-Philippe Georges
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Air Temperature ◽  
Data Center ◽  
Waste Heat ◽  
Cooling System ◽  
Thermal Characteristics ◽  
Ambient Air ◽  
Characteristics Analysis ◽  
Temperature Ranges

The energy efficiency of Data Center (DC) operations heavily relies on a DC ambient temperature as well as its IT and cooling systems performance. A reliable and efficient cooling system is necessary to produce a persistent flow of cold air to cool servers that are subjected to constantly increasing computational load due to the advent of smart cloud-based applications. Consequently, the increased demand for computing power will inadvertently increase server waste heat creation in data centers. To improve a DC thermal profile which could undeniably influence energy efficiency and reliability of IT equipment, it is imperative to explore the thermal characteristics analysis of an IT room. This work encompasses the employment of an unsupervised machine learning technique for uncovering weaknesses of a DC cooling system based on real DC monitoring thermal data. The findings of the analysis result in the identification of areas for thermal management and cooling improvement that further feeds into DC recommendations. With the aim to identify overheated zones in a DC IT room and corresponding servers, we applied analyzed thermal characteristics of the IT room. Experimental dataset includes measurements of ambient air temperature in the hot aisle of the IT room in ENEA Portici research center hosting the CRESCO6 computing cluster. We use machine learning clustering techniques to identify overheated locations and categorize computing nodes based on surrounding air temperature ranges abstracted from the data. This work employs the principles and approaches replicable for the analysis of thermal characteristics of any DC, thereby fostering transferability. This paper demonstrates how best practices and guidelines could be applied for thermal analysis and profiling of a commercial DC based on real thermal monitoring data.

scholarly journals A Machine Learning Solution for Data Center Thermal Characteristics Analysis

2020 ◽  
Author(s):  
marta chinnici ◽  
Anastasiia GRISHIna ◽  
Ah-Lian KOR ◽  
Eric Rondeau ◽  
jean philippe georges
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Air Temperature ◽  
Data Center ◽  
Heat Dissipation ◽  
Waste Heat ◽  
Cooling System ◽  
Thermal Characteristics ◽  
Ambient Air ◽  
Characteristics Analysis

Energy efficiency of Data Center (DC) operations heavily relies on IT and cooling systems performance. A reliable and efficient cooling system is necessary to produce a persistent flow of cold air to cool servers that are subjected to constantly increasing computational load due to the advent of IoT- enabled smart systems. Consequently, increased demand for computing power will bring about increased waste heat dissipation in data centers. In order to bring about a DC energy efficiency, it is imperative to explore the thermal characteristics analysis of an IT room (due to waste heat). This work encompasses the employment of an unsupervised machine learning modelling technique for uncovering weaknesses of the DC cooling system based on real DC monitoring thermal data. The findings of the analysis result in the identification of areas for energy efficiency improvement that will feed into DC recommendations. The methodology employed for this research includes statistical analysis of IT room thermal characteristics, and the identification of individual servers that frequently occur in the hotspot zones. A critical analysis has been conducted on available big dataset of ambient air temperature in the hot aisle of ENEA Portici CRESCO6 computing cluster. Clustering techniques have been used for hotspots localization as well as categorization of nodes based on surrounding air temperature ranges. The principles and approaches covered in this work are replicable for energy efficiency evaluation of any DC and thus, foster transferability. This work showcases applicability of best practices and guidelines in the context of a real commercial DC that transcends the set of existing metrics for DC energy efficiency assessment.

Energy-Efficiency Through the Integration of Information and Communications Technology Management and Facilities Controls

2009 ◽  
Author(s):  
Michael K. Patterson ◽  
Michael Meakins ◽  
Dennis Nasont ◽  
Prasad Pusuluri ◽  
William Tschudi ◽  
...  
Keyword(s):  
Energy Efficiency ◽  
Air Temperature ◽  
Data Center ◽  
Temperature Sensor ◽  
Data Centers ◽  
Cooling System ◽  
Information And Communications Technology ◽  
Demonstration Project ◽  
Communications Technology ◽  
Control Loop

Increasing energy-efficient performance built into today’s servers has created significant opportunities for expanded Information and Communications Technology (ICT) capabilities. Unfortunately the power densities of these systems now challenge the data center cooling systems and have outpaced the ability of many data centers to support them. One of the persistent problems yet to be overcome in the data center space has been the separate worlds of the ICT and Facilities design and operations. This paper covers the implementation of a demonstration project where the integration of these two management systems can be used to gain significant energy savings while improving the operations staff’s visibility to the full data center; both ICT and facilities. The majority of servers have a host of platform information available to the ICT management network. This demonstration project takes the front panel temperature sensor data from the servers and provides that information over to the facilities management system to control the cooling system in the data center. The majority of data centers still use the cooling system return air temperature as the primary control variable to adjust supply air temperature, significantly limiting energy efficiency. Current best practices use a cold aisle temperature sensor to drive the cooling system. But even in this case the sensor is still only a proxy for what really matters; the inlet temperature to the servers. The paper presents a novel control scheme in which the control of the cooling system is split into two control loops to maximize efficiency. The first control loop is the cooling fluid which is driven by the temperature from the physically lower server to ensure the correct supply air temperature. The second control loop is the airflow in the cooling system. A variable speed drive is controlled by a differential temperature from the lower server to the server at the top of the rack. Controlling to this differential temperature will minimize the amount of air moved (and energy to do so) while ensuring no recirculation from the hot aisle. Controlling both of these facilities parameters by the server’s data will allow optimization of the energy used in the cooling system. Challenges with the integration of the ICT management data with the facilities control system are discussed. It is expected that this will be the most fruitful area in improving data center efficiency over the next several years.

scholarly journals ОХОЛОДЖЕННЯ ПОВІТРЯ НА ВХОДІ ГОЛОВНОГО СУДНОВОГО ДВИГУНА АБСОРБЦІЙНОЮ БРОМИСТОЛІТІЄВОЮ ХОЛОДИЛЬНОЮ МАШИНОЮ В ТРОПІЧНИХ УМОВАХ

2020 ◽  
pp. 18-23
Author(s):  
Роман Миколайович Радченко ◽  
Дмитро Вікторович Коновалов ◽  
Максим Андрійович Пирисунько ◽  
Чжан Цян ◽  
Луо Зевей
Keyword(s):  
Air Temperature ◽  
High Efficiency ◽  
Waste Heat ◽  
Cooling System ◽  
Ambient Air ◽  
Climatic Conditions ◽  
Air Cooling ◽  
Exhaust Gas ◽  
Low Speed ◽  
Speed Engine

The efficiency of air cooling at the inlet of the main low speed engine of a transport vessel during operation in tropical climatic conditions on the Shanghai-Karachi-Shanghai route was analyzed. The peculiarity of the tropical climate is the high relative humidity of the air at the same time its high temperatures, and hence the increased thermal load on the cooling system, which requires efficient transformation of the waste heat into the cold in the case of the use of waste heat recovery refrigeration machines. The cooling of the air at the inlet of the low speed engine by absorption lithium bromide chillers, which are characterized by high efficiency of transformation of waste heat into cold – by high coefficients of performance, is investigated. A schematic-construction solution of the air cooling system at the inlet of the ship's main engine using the heat of exhaust gases by an absorption chiller is proposed and analyzed. With this the cooling potential of the inlet air cooling from the current ambient air temperature to 15 ° C and the corresponding heat consumption for the operation of the adsorption chiller, on the one hand, was compared with the available exhaust gas heat potential, on the other hand. The effect of using the exhaust gas heat to cool the air at the inlet of the engine has been analyzed taking into account the changing climatic conditions during the voyage. Enhancement of fuel efficiency of the ship's engine by reducing the inlet air temperature were evaluated by current values of the reduction in specific and total fuel consumption. It is shown that due to the high efficiency of heat conversion in absorption chillers (high coefficients of performance 0.7…0.8), a significant amount of excessive exhaust gas heat over the heat required to cool the ambient air at the inlet of the engine to 15 ° C, which reaches almost half of the available exhaust gas heat during the Shanghai-Karachi-Shanghai route. This reveals the possibility of additional cooling a scavenge air too with almost double fuel economy due to the cooling of all cycle air of the low speed engine, including the air at the inlet.

Increasing the energy efficiency of a data center based on machine learning

2021 ◽  
Author(s):  
Zhen Yang ◽  
Jinhong Du ◽  
Yiting Lin ◽  
Zhen Du ◽  
Li Xia ◽  
...  
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Data Center

Exergy Analysis of Data Center Thermal Management Systems

2008 ◽  
Vol 130 (2) ◽  
Author(s):  
Amip J. Shah ◽  
Van P. Carey ◽  
Cullen E. Bash ◽  
Chandrakant D. Patel
Keyword(s):  
Energy Efficiency ◽  
Thermal Management ◽  
Data Center ◽  
Experimental Validation ◽  
Irreversible Process ◽  
Exergy Analysis ◽  
Cooling System ◽  
Available Energy ◽  
Thermal Management System ◽  
Data Center Cooling

The modeling of recirculation patterns in air-cooled data centers is of interest to ensure adequate thermal management of computer racks at increased heat densities. Most metrics that describe recirculation are based exclusively on temperature inside the data center, and therefore fail to provide adequate information regarding the energy efficiency of the thermal infrastructure. This paper addresses this limitation through an exergy analysis of the data center thermal management system. The approach recognizes that the mixing of hot and cold streams in the data center airspace is an irreversible process and must therefore lead to a loss of exergy. Experimental validation in a test data center confirms that such an exergy-based characterization in the cold aisle reflects the same recirculation trends as suggested by traditional temperature-based metrics. Further, by extending the exergy-based model to include irreversibilities from other components of the thermal architecture, it becomes possible to quantify the amount of available energy supplied to the cooling system that is being utilized for thermal management purposes. The energy efficiency of the entire data center cooling system can then be collapsed into the single metric of net exergy loss. When evaluated against a ground state of the external ambience, this metric enables an estimate of how much of the energy emitted into the environment could potentially be harnessed in the form of useful work. Thus, this paper successfully demonstrates that the proposed exergy-based approach can provide a foundation upon which the data center cooling system can be simultaneously evaluated for thermal manageability and energy efficiency.

scholarly journals Monitoring the efficiency of cooling air at the inlet of gas engine in integrated energy system

2020 ◽  
pp. 344-344
Author(s):  
Andrii Radchenko ◽  
Ionut-Cristian Scurtu ◽  
Mykola Radchenko ◽  
Serhiy Forduy ◽  
Anatoliy Zubarev
Keyword(s):  
Waste Heat ◽  
Cooling System ◽  
Ambient Air ◽  
Energy System ◽  
Air Conditioner ◽  
Absorption Chiller ◽  
Gas Engine ◽  
Chilled Water ◽  
Air Cooler ◽  
Cooling Air

The fuel efficiency of gas engines is effected by the temperature of intake air at the suction of turbocharger. The data on dependence of fuel consumption and engine electric power on the intake air temperature were monitored for Jenbacher gas engine JMS 420 GS-N.LC to evaluate its influence. A waste heat of engine is rejected for heating water to the temperature of about 90??. The heat received is used in absorption lithium-bromide chiller to produce a cold in the form of chilled water. A cooling capacity of absorption chiller firstly is spent for technological needs and then for feeding the central air conditioner for cooling the ambient air incoming the engine room, from where the air is sucked by the engine turbocharger. The monitoring data revealed the reserves to enhance the efficiency of traditional cooling system of intake air by absorption chiller through deeper cooling. This concept can be realized in two ways: by addition cooling a chilled water from absorption chiller to about 5-7?? for feeding engine intake air cooler or by two-stage cooling with precooling ambient air by chilled water from ACh in the first stage and subsequent deep cooling air to the temperatures 7-10?? in the second stage of intake air cooler by using a refrigerant as a coolant. In both cases the ejector chiller could be applied as the most simple in design.

The Oriented Spray Cooling System for Supplementing Cooling Lakes

2017 ◽  
Author(s):  
Charles F. Bowman
Keyword(s):  
Power Plants ◽  
Waste Heat ◽  
Cooling System ◽  
Air Flow ◽  
Ambient Air ◽  
Spray Cooling ◽  
Cooling Capacity ◽  
Ambient Temperatures ◽  
Circulating Water ◽  
Auxiliary Power

With ever-increasing ambient temperatures many electric power plants that employ cooling lakes to reject their waste heat into the environment are struggling to maintain reasonable turbine backpressures during the hot summer months when electric load demand is often the greatest. Some consider adding mechanical draft cooling towers (MDCT) to further cool the condenser circulating water (CCW) prior to entering the main condenser, but the additional auxiliary power required to drive MDCT fans often consume the additional generator output resulting from the lower backpressure. Spray ponds offer significant advantages over MDCT including superior simplicity and operability, lower power requirements, and lower capital and maintenance costs. The Oriented Spray Cooling System (OSCS) is an evolutionary spray pond design. Unlike a conventional spray pond in which spray nozzles are arranged in a flat bed and spray upward, blocking the ambient air flow to the spray region as it travels down to the pond below, the OSCS nozzles are mounted on spray trees arranged in a circle and are tilted at an angle oriented towards the center of the circle. As a result, the water droplets drag air into the spray region while the warm air concentrated in the center of the circle rises. Both of these effects work together to increase air flow through the spray region. Increased air flow reduces the local wet-bulb temperature (LWBT) of the air in the spray pattern, promoting heat transfer and more efficient cooling. During the late 1970’s the author developed a purely analytical model to predict the thermal performance of the OSCS which was successfully compared with the OSCS at the Columbia Generating Station (CGS) in the mid 1980’s. This paper describes how the OSCS may be employed to supplement the cooling capacity of an existing cooling lake to reduce the temperature of the CCW prior to entering a power plant, resulting in lower main condenser pressures and more net plant output.

Sign in / Sign up

Export Citation Format

Share Document