Case Study for Energy-Efficiency of Liquid-Cooled Data Centers

Author(s):  
Ricardo Rivera-Lopez ◽  
Mark Kimber

The thermal management of existing data centers is centered on forced convection using air as the transport fluid. A large portion of the energy required for typical data centers is used in maintaining reasonable operating temperatures, and many have looked to liquid cooling as a promising solution to increased energy efficiency. The current work is a case study of making this transition for a single computer board. The energy savings potential is quantified and the removal of heat via liquid cooling is characterized from the chip level to the environment. A thermal solution model is developed and validated through experimentation. The experiment consists of a rack-mounted computer board to simulate a server and cold plates attached at several key locations for cooling. Multiple measurements are made to determine the amount of heat removed and power consumed in the process. The results from this study show that liquid-cooling presents an improved thermal solution to data centers and the energy savings potential is large, which improves the power usage effectiveness since power is mostly used in data processing rather than server cooling.

2014 ◽  
Vol 548-549 ◽  
pp. 1815-1819 ◽  
Author(s):  
Xiao Chun Qin ◽  
She Gang Shao ◽  
Yi Shen

Green lighting technology has the advantages of energy efficiency, friendly environment, safety and comfort. Based on the introduction of green lighting technology, taken the Mt. Lushan West Sea tourist highway service as the case study, we analyzed light guide illumination, the optimum use of natural light and energy efficient lighting respectively from the aspects of technical characteristics and the specific highway service application. We finally made the economic analysis in the energy savings of green lights in the highway service, and the result showed that through the use of green lighting systems Mt. Lushan West Sea tourist highway service could save electricity and reduce operating costs 134,700 Yuan per year.


2011 ◽  
Vol 2011.21 (0) ◽  
pp. 248-251
Author(s):  
Ari YOSHII ◽  
Yosuke UDAGAWA ◽  
Masahide YANAGI ◽  
Shisei WARAGAI ◽  
Keigo MATSUO ◽  
...  

2022 ◽  
Vol 1216 (1) ◽  
pp. 012014
Author(s):  
R Uanov ◽  
A S Begimbetova

Abstract The article deals with the analysis of methods for assessing the energy efficiency of data centers according to the Power Usage Effectiveness method. The demand for data centers which consumes a large amount of electricity is growing with the growth of digitalization and the accumulation of big data in the network. The energy consumption of the cooling system for the machine room accounts for a significant part of the operating costs of the building. Free cooling in a refrigeration system reduces energy consumption much more than operating systems with a vapor-compression cycle. In 2006 according to The Green Grid, the assessment method of Power Usage Effectiveness has become an international standard for measuring energy efficiency and is widely used in the design and operation of data centers. In this regard, the operation principles of free-cooling chillers are considered. The calculation example of the system payback in free-cooling is also given.


Author(s):  
Tahir Cader ◽  
Levi Westra ◽  
Andres Marquez

Although semiconductor manufacturers have provided temporary relief with lower-power multi-core microprocessors, OEMs and data center operators continue to push the limits for individual rack power densities. It is not uncommon today for data center operators to deploy multiple 20 kW racks in a facility. Such rack densities are exacerbating the major issues of power and cooling in data centers. Data center operators are now forced to take a hard look at the efficiencies of their data centers. Malone and Belady (2006) have proposed three metrics, i.e., Power Usage Effectiveness (PUE), Data Center Efficiency (DCE), and the Energy-to-Acquisition Cost ratio (EAC), to help data center operators quickly quantify the efficiency of their data centers. In their paper, Malone and Belady present nominal values of PUE across a broad cross-section of data centers. PUE values are presented for data centers at four levels of optimization. One of these optimizations involves the use of Computational Fluid Dynamics (CFD). In the current paper, CFD is used to conduct an in-depth investigation of a liquid-cooled data center that would potentially be housed at the Pacific Northwest National Labs (PNNL). The boundary conditions used in the CFD model are based upon actual measurements on a rack of liquid-cooled servers housed at PNNL. The analysis shows that the liquid-cooled facility could achieve a PUE of 1.57 as compared to a PUE of 3.0 for a typical data center (the lower the PUE, the better, with values below 1.6 approaching ideal). The increase in data center efficiency is also translated into an increase in the amount of IT equipment that can be deployed. At a PUE of 1.57, the analysis shows that 91% more IT equipment can be deployed as compared to the typical data center. The paper will discuss the analysis of the PUE, and will also explore the impact of the raising data center efficiency via the use of multiple cooling technologies and CFD analysis. Complete results of the analyses will be presented in the paper.


2010 ◽  
Vol 16 (4) ◽  
pp. 567-576 ◽  
Author(s):  
Jorge S. Carlos ◽  
Helena Corvacho

A study on thermal retrofit of Portuguese elementary school buildings is presented. The type of school under analysis is one adopted by a large construction campaign that began in the 1940's. This building stock has a very poor thermal performance and their retrofit was evaluated starting with a case study of a school in the central region of Portugal, where some experimental measures were performed and a calculation method was applied for the heating energy consumption estimation. A solution for the thermal retrofit of the school building external envelope was optimized and the effect on heating energy consumption was evaluated, using ECOTECT, resulting in a reduction of 52% of heating energy needs. The national impact of the thermal retrofit of the whole building stock was characterised in terms of energy savings. Finally, the pre‐heating of the ventilation air was also tested as a complementary measure and its effect evaluated. The solution tested may provide up to 1000 kWh/year of extra heat gains by pre‐heating the ventilation air. It must be underlined though that the performance of these systems is dependent on the thermal properties of their components so higher reductions can be achieved with the improvement of these properties. Santrauka Straipsnyje pateikiami Portugalijos pradines mokyklos šiluminio atnaujinimo tyrimai. Analizuojamos mokyklos tipas yra vienas iš taikytu po 1940 metu prasidejusioje plačioje statybos kampanijoje. Šios pastatu grupes šilumines charakteristikos yra labai prastos. Ju atnaujinimo vertinimas buvo pradetas nuo centrineje Portugalijoje esančios mokyklos, kurioje buvo igyvendintos kai kurios eksperimentines priemones, ir energijos sanaudoms nustatyti pritaikytas skaičiavimo metodas. Pastato išoriniu atitvaru šiluminio atnaujinimo sprendimas buvo optimizuotas ir jo itaka šilumines energijos sanaudoms nustatyta naudojant ECOTECT. Šilumines energijos poreikis sumažejo 52 %. Iš viso pastatu fondo šiluminio atnaujinimo itaka nacionaliniu mastu vertinta sutaupytos energijos kiekiu. Pabaigoje kaip papildoma priemone buvo išbandytas pirminis vedinamo oro pašildymas, nustatytas jo naudingumas. Išbandytasis pirminis vedinamo oro pašildymas gali suteikti iki 1000 kWh/metus papildomo išsiskiriančio šilumos kiekio. Pabrežtina, kad nors šiu sistemu veikimo charakteristikos priklauso nuo ju komponentu šiluminiu savybiu, gerinant šias savybes galima daugiau sumažinti energijos sanaudu.


Author(s):  
Amip Shah ◽  
Cullen Bash ◽  
Ratnesh Sharma ◽  
Tom Christian ◽  
Brian J. Watson ◽  
...  

Numerous evaluation metrics and standards are being proposed across industry and government to measure and monitor the energy efficiency of data centers. However, the energy use of data centers is just one aspect of the environmental impact. In this paper, we explore the overall environmental footprint of data centers beyond just energy efficiency. Building upon established procedures from the environmental sciences, we create an end-to-end life-cycle model of the environmental footprint of data centers across a diverse range of impacts. We test this model in the case study of a hypothetical 2.2-MW data center. Our analysis suggests the need for evaluation metrics that go beyond just operational energy use in order to achieve sustainable data centers.


Author(s):  
Filomena Pietrapertosa ◽  
Marco Tancredi ◽  
Michele Giordano ◽  
Carmelina Cosmi ◽  
Monica Salvia

The European Union 2050 climate neutrality goal and the climate crisis require coordinated efforts to reduce energy consumption in all sectors, and mainly in buildings greatly affected by the increasing temperature, with relevant CO2 emissions due to inefficient end-use technologies. Moreover, the old building stock of most countries requires suited policies to support renovation programs aimed at improving energy performances and optimize energy uses. A toolbox was developed in the framework of the PrioritEE project to provide policy makers and technicians with a wide set of tools to support energy efficiency in Municipal Public Buildings. The toolbox, available for free, was tested in the partners’ communities, proving its effectiveness. The paper illustrates its application to the Potenza Municipality case study in which the online calculator DSTool (the core instrument of the toolbox) was utilized to select and prioritize the energy efficiency interventions in public buildings implementable in a three-year action plan in terms of costs, energy savings, CO2 emissions’ reduction and return on investments. The results highlight that improvements in the building envelopes (walls and roofs), heating and lighting and photovoltaic systems allow reducing CO2 emission approximately 644 t/year and saving about 2050 MWh/year with a total three-year investment of 1,728,823 EUR.


Author(s):  
Gregory Raffio ◽  
Ovelio Isambert ◽  
George Mertz ◽  
Charlie Schreier ◽  
Kelly Kissock

This paper describes a four-step method to analyze the utility bills and weather data from multiple residences to target buildings for specific energy conservation retrofits. The method is also useful for focusing energy assessments on the most promising opportunities. The first step of the method is to create a three-parameter changepoint regression model of energy use versus weather for each building and fuel type. The three model parameters represent weather independent energy use, the building heating or cooling coefficient and the building balance-point temperature. The second step is to drive the models using typical TMY2 weather data to determine Normalized Annual Consumption (NAC) for each fuel type. The third step is to create a sliding NAC with each set of 12 sequential months of utility data. The final step is to benchmark the NACs and coefficients of multiple buildings to identify average, best and worst energy performers, and how the performance of each building has changed over time. The method identifies billing errors, normalizes energy use for changing weather, prioritizes sites for specific energy-efficiency retrofits and tracks weather-normalized changes in energy use. The principle differences between this method and previously defined ones are that this method seeks to use inverse modeling proactively to identify energy saving opportunities rather than retroactively to measure energy savings, it tracks changes in building performance using sliding analysis, and it uses comparisons between multiple buildings to extract additional information. This paper describes the method, then demonstrates the method through a case study of about 300 low-income residences. After applying the method, targeted buildings were visited to determine the accuracy of the method at identifying energy efficiency opportunities. The case study shows that over 80% of the targeted buildings presented at least one of the expected problems from each type of retrofit.


Author(s):  
Miroslav Variny

This paper responds to the article by Pietrapertosa et al., doi:10.3390/ijerph17124434, published previously in the International Journal of Environmental Research and Public Health. Its aim is to discuss the appropriateness of the studied method, to analyze its weak sides and to propose its robustness improvement. Thus, data presented in the above study were examined and recalculated, yielding, among others, indicators of annual energy savings (in kWh per m2 of total heated area) and specific proposals investment costs (in € per m2 of total heated area). By analyzing the obtained data for all public buildings, a significantly simplified approach to this problematic has been suggested while several other features of the research method and some presented results lack proper reasoning and discussion. Individual approach to each public building has been proposed and discussed point-by-point to enhance the method’s applicability. As a result, more realistic outcomes are obtained, and suitable investment actions can be proposed.


Sign in / Sign up

Export Citation Format

Share Document