Energy Flow in the Information Technology Stack: Introducing the Coefficient of Performance of the Ensemble

Patel, Chandrakant D.; Sharma, Ratnesh; Bash, Cullen; Beitelmal, Monem

doi:10.1115/imece2006-14830

Cited by 27 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the CRAC units provide the highest resistance, owing to the large input of high-quality electrical energy resulting from the unnecessarily high provisioning. As evidence, for the given data center configuration, the CRAC units require nearly 52% of the total cooling power within the data center -nearly twice what is typical for most data center facilities [20] [25].…”

Section: Case Studymentioning

confidence: 91%

Optimizing data center cooling infrastructures using exergothermovolumes

Shah

Bash

Patel

2010

2010 12th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems

View full text Add to dashboard Cite

Exergo-thermo-volumes provide a novel approach to simultaneously quantify the thermofluidic performance, operational cost, and environmental sustainability of a given thermal management system. In this work, we demonstrate the use of exergo-thermo-volumes to optimize the cooling infrastructure of a data center along the above criteria. The paper begins with an overview of the methodology and its use in the form of multi-component network models. Using such a network, a performance assessment model is proposed that takes into account the key vectors of thermofluidic performance, energy efficiency (operational cost), and environmental sustainability of a given system. We then apply the methodology for data center management. By applying to an example case study of a traditional data center, we show how an exergo-thermo-volume network can be used to gain insight into the design and operation of the compute as well as cooling infrastructures. Differences in infrastructure performance across various operating environments are discussed, and an optimal design that trades off key vectors of thermofluidic performance, operational cost and environmental sustainability is developed. The paper concludes by reflecting upon a generalized design and management strategy for sustainable design of future data center architectures.

show abstract

Section: Case Studymentioning

confidence: 91%

Optimizing data center cooling infrastructures using exergothermovolumes

Shah

Bash

Patel

2010

2010 12th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems

View full text Add to dashboard Cite

show abstract

“…The formulation of COP for these situations is generally more complicated than the simplistic representation described in Equation (9); for example, Patel et al [34] present a model for estimating the coefficient of performance of the ensemble of cooling devices. In the present work, we limit our focus to system-level measures of COP as described in Equation (9), but the extension to infrastructure-level models is conceptually straightforward.…”

Section: Hybrid Approach: Exergo-thermo-volumesmentioning

confidence: 99%

Designing environmentally sustainable electronic cooling systems using exergo-thermo-volumes

Shah

Patel

2009

Int. J. Energy Res.

Self Cite

View full text Add to dashboard Cite

SUMMARYThermo-volumes allow the design engineer to expediently understand the thermal resistance of a given cooling solution (an indicator of performance) along with its flow resistance (an indicator of the pumping power, or energy consumption, which will be required by the fluid handler). In the present work, we expand upon thermo-volumes by including the lifetime exergy cost (in units of Joules of availability destroyed) as a means to enable the consideration of resource consumption (and thus the environmental sustainability) of the cooling solution.To achieve these exergo-thermo-volumes, we reinterpret previous definitions of thermo-volumes in terms of the entropy generated during heat transfer and fluid flow. The Guoy-Stodola theorem is used to convert this entropy generation into an 'operational' exergy loss. Next, based on the material choice and assembly processes used in creating the product, an embedded exergy consumption that accounts for the amount of exergy destroyed during extraction, transportation and disposal of the material is attached to the operational exergy loss. Thus, the total 'cradle-to-cradle' exergy loss of the solution is devised. In this framework, the optimal solution will be that which destroys the minimal amount of exergy. Correspondingly, instead of relying upon the coefficient of performance (which is focused on operational consumption), we propose evaluation of cooling solutions in terms of the heat removal capacity per unit lifetime exergy consumption.The paper concludes by illustrating applicability of the method to the design of an enterprise server. It should be noted that although the paper is focused on electronics cooling solutions, the methodology is designed to be sufficiently general for use in any thermal management application.

show abstract

“…Finally, it is worth mentioning that an optimized lifetime-aware management of DCs could also have implications on the provisioning of air conditioning [22], as well as reducing active redundant air conditioning systems.…”

Section: Data Center Thermo-mechanical Physical Designmentioning

confidence: 99%

Lifetime-Aware Cloud Data Centers: Models and Performance Evaluation

et al. 2016

View full text Add to dashboard Cite

We present a model to evaluate the server lifetime in cloud data centers (DCs). In particular, when the server power level is decreased, the failure rate tends to be reduced as a consequence of the limited number of components powered on. However, the variation between the different power states triggers a failure rate increase. We therefore consider these two effects in a server lifetime model, subject to an energy-aware management policy. We then evaluate our model in a realistic case study. Our results show that the impact on the server lifetime is far from negligible. As a consequence, we argue that a lifetime-aware approach should be pursued to decide how and when to apply a power state change to a server.

show abstract

Energy Flow in the Information Technology Stack: Introducing the Coefficient of Performance of the Ensemble

Cited by 27 publications

References 11 publications

Optimizing data center cooling infrastructures using exergothermovolumes

Optimizing data center cooling infrastructures using exergothermovolumes

Designing environmentally sustainable electronic cooling systems using exergo-thermo-volumes

Lifetime-Aware Cloud Data Centers: Models and Performance Evaluation

Contact Info

Product

Resources

About