Assessing the environmental impact of data centres part 1: Background, energy use and metrics

Whitehead, Beth; Andrews, Deborah; Shah, Amip; Maidment, Graeme

doi:10.1016/j.buildenv.2014.08.021

Cited by 115 publications

(46 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Opposite of this, the MNEP method simultaneously provides more precise information about the energy consumption because the working parameters of each IT component are also included in the multilayer node event processing. Second, data center energy usage creates a number of environmental problems [39,40]. A recent study done by Van Heddeghem et al [41] has found that data centers worldwide consumed 270 TWh of energy in 2014 and this consumption had a compound annual growth rate (CAGR) of 4.4% from 2007 to 2014.…”

Section: Discussionmentioning

confidence: 99%

Improved Data Center Energy Efficiency and Availability with Multilayer Node Event Processing

Matko

Brezovec²

2018

Energies

View full text Add to dashboard Cite

This article presents an overview of a new multilayer node event processing (MNEP) method for improved data center energy efficiency and event management monitoring in its physical infrastructure. Event management is essential for the provision and maintenance of the reliability and functionality of the physical infrastructure. For effective maintenance, action preventing any downtime accurate event information as well as appropriate energy support is of outmost importance. The novelty of this approach lies in the enhanced availability and reliability of the physical infrastructure, improved data center infrastructure energy efficiency, as well as lower costs when designing a new physical infrastructure system. The MNEP method presents a data model based on the node tree processing. The experimental results show that the number of all potential unexpected events is reduced significantly as a result of the accurate identification of the alarm root cause. The comparison of time parameters required to evaluate the maintenance factors shows that the time needed to detect, identify, and react to alarm events decreases and has a significant impact on maintaining a defined high level of availability and energy efficiency in a data center physical infrastructure.

show abstract

Section: Discussionmentioning

confidence: 99%

Improved Data Center Energy Efficiency and Availability with Multilayer Node Event Processing

Matko

Brezovec²

2018

Energies

View full text Add to dashboard Cite

show abstract

“…Terms like "the cloud" depicting a fluffy formation with no weight or mass, are obfuscating the fact that most of such digital solutions consume massive digital infrastructures with additional electricity demands, which are often times provided by old-fashioned and carbon-emitting energy supplies [18]. The synchronization of systems, where the end device might be located on the other side of the world than the data center, has caused a high increase of energy consumption at the sites of the central servers [19]. In an expected case scenario, the electricity consumed by communication technology will increase from about 2,000 TWh/year in 2010 to more than 8,000 TWh/year in 2030.…”

Section: Cloud Computing and Sustainable Developmentmentioning

confidence: 99%

Edge Computing: A Comprehensive Survey of Current Initiatives and a Roadmap for a Sustainable Edge Computing Development

Hamm¹,

Willner²,

Schieferdecker³

2020

WI2020 Zentrale Tracks

View full text Add to dashboard Cite

Edge Computing is a new distributed Cloud Computing paradigm in which computing and storage capabilities are pushed to the topological edge of a network. However, various standards and implementations are promoted by different initiatives. Lead by a reference architecture model for Edge Computing, current initiatives are analyzed by explorative content analysis. Providing two main contributions to the field, we present, first, how current initiatives are characterized, and second, a roadmap for sustainable Edge Computing relating three dimensions of sustainable development to four cross-concerns of Edge Computing. Findings show that most initiatives are internationally organized software development projects; important branches are currently telecom and industrial sectors; most addressed is the network virtualization layer. The roadmap reveals numerous chances and risks of Edge Computing related to sustainable development; such as the use of renewable energies, biases, new business models, increase and decrease of energy consumption, responsiveness, monitoring and traceability.

show abstract

“…By means of the recent developments in the ICT facilities such as cloud computing and big data, the energy consumption of the DC units has been drastically increasing . According to Jones, the global energy demand by DCs reached 200 TWh in 2018, whereas the total energy consumption in DCs is roundly equal the 0.5% of both the global electricity demand and the worldwide carbon dioxide equivalent . Nadjahi et al divided the DC electricity consumption into three categories which are the electricity consumption via ICT equipment with 52% share, the cooling load of DCs with 38% share, and the consumption of other equipment (eg, power distribution, Uninterruptible Power Supply (UPS) with 10% share.…”

Section: Introductionmentioning

confidence: 99%

Combined heat and power generation via hybrid data center cooling‐polymer electrolyte membrane fuel cell system

Kanbur

Duan

2020

Int J Energy Res

View full text Add to dashboard Cite

Summary Although there has been a lot of waste heat utilization studies for the air‐cooled data center (DC) systems, the waste heat utilization has not been studied for the liquid‐cooled DC systems, which have been rapidly gaining importance for the high‐performance Information and Communication Technology facilities such as cloud computing and big data storage. Compared to the air‐cooled systems, higher heat removal capacity of the liquid‐cooled DC systems provides better heat transfer performance; and therefore, the waste heat of the liquid‐cooled DC systems can be more efficiently utilized in the low‐temperature and low‐carbon energy systems such as electricity generation via polymer electrolyte membrane (PEM) fuel cells. For this purpose, the current study proposes a novel hybrid system that consists of the PEM fuel cell and the two‐phase liquid‐immersion DC cooling system. The two‐phase liquid immersion DC cooling system is one of the most recent and advanced DC cooling methods and has not been considered in the DC waste heat utilization studies before. The PEM fuel cell unit is operated with the hydrogen and compressed air flows that are pre‐heated in the DC cooling unit. Due to its original design, the hybrid system brings its own original design criteria and limitations, which are taken into account in the energetic and exergetic assessments. The power density of the PEM fuel cell reaches up to 0.99 kW/m2 with the water production rate of 0.0157 kg/s. In the electricity generation case, the highest energetic efficiency is found as 15.8% whereas the efficiency increases up to 96.16% when different multigeneration cases are considered. The hybrid design deduces that the highest exergetic efficiency and sustainability index are 43.3% and 1.76 and they are 9.4% and 6.6% higher than exergetic and sustainability performances of the stand‐alone PEM fuel cell operation, respectively.

show abstract

Assessing the environmental impact of data centres part 1: Background, energy use and metrics

Cited by 115 publications

References 9 publications

Improved Data Center Energy Efficiency and Availability with Multilayer Node Event Processing

Improved Data Center Energy Efficiency and Availability with Multilayer Node Event Processing

Edge Computing: A Comprehensive Survey of Current Initiatives and a Roadmap for a Sustainable Edge Computing Development

Combined heat and power generation via hybrid data center cooling‐polymer electrolyte membrane fuel cell system

Contact Info

Product

Resources

About