Content uploaded by Luca Castellazzi
Author content
All content in this area was uploaded by Luca Castellazzi on Jan 03, 2018
Content may be subject to copyright.
Trends in data centre energy
consumption under the European
Code of Conduct for Data Centre
Energy Efficiency
Paolo Bertoldi, Maria Avgerinou, Luca Castellazzi
2017
EUR
28874 EN
This publication is a Technical report by the Joint Research Centre (JRC), the European Commission’s science
and knowledge service. It aims to provide evidence-based scientific support to the European policymaking
process. The scientific output expressed does not imply a policy position of the European Commission. Neither
the European Commission nor any person acting on behalf of the Commission is responsible for the use that
might be made of this publication.
Contact information
Name: Paolo Bertoldi
Address: European Commission, Joint Research Centre, Via E. Fermi, 2749, I-21027 Ispra (VA), ITALY
Email: paolo.bertoldi@ec.europa.eu
Tel.: +39 0332 789299
JRC Science Hub
https://ec.europa.eu/jrc
JRC108354
EUR 28874 EN
PDF ISBN 978-92-79-76445-5 ISSN 1831-9424 doi:10.2760/358256
Luxembourg: Publications Office of the European Union, 2017
© European Union, 2017
Reuse is authorised provided the source is acknowledged. The reuse policy of European Commission documents
is regulated by Decision 2011/833/EU (OJ L 330, 14.12.2011, p. 39).
For any use or reproduction of photos or other material that is not under the EU copyright, permission must be
sought directly from the copyright holders.
How to cite this report: P. Bertoldi, M. Avgerinou, L. Castellazzi, Trends in data centre energy consumption
under the European Code of Conduct for Data Centre Energy Efficiency, EUR 28874 EN, Publications Office of
the European Union, Luxembourg, 2017, ISBN 978-92-79-76445-5, doi:10.2760/358256, JRC108354
All images © European Union 2017, except:
cover page, cybrain, image # 479232763, Source: www.shutterstock.com
i
Contents
Acknowledgements ................................................................................................ 1
Abstract ............................................................................................................... 2
1
Introduction ...................................................................................................... 3
2
European Trends ............................................................................................... 5
2.1
Data centre markets .................................................................................... 5
2.1.1
London Market .................................................................................... 6
2.1.2
Frankfurt Market .................................................................................. 6
2.1.3
Paris Market ........................................................................................ 7
2.1.4
Amsterdam Market .............................................................................. 7
2.1.5
Other Markets ..................................................................................... 8
3
European Code of Conduct for Data Centre Energy Efficiency ................................ 10
4
Methodology and Metrics .................................................................................. 13
5
Results and Analysis ........................................................................................ 14
5.1
PUE analysis ............................................................................................. 14
5.2
Sector Distribution Analysis......................................................................... 20
5.3
Geographical Distribution Analysis ............................................................... 21
5.4
Free Cooling Technologies ........................................................................... 24
5.5
Best practices............................................................................................ 26
6 Conclusions ...................................................................................................... 31
References ......................................................................................................... 32
List of abbreviations and definitions ....................................................................... 36
List of figures ...................................................................................................... 37
List of tables ....................................................................................................... 38
1
Acknowledgements
We thank Katalin Bodis, Officer at European Commission, Joint Research Centre, for the
assistance with the geographical distribution map.
Authors
Paolo Bertoldi, Maria Avgerinou, Luca Castellazzi
2
Abstract
Climate change is recognised as one of the key challenges humankind is facing. The
Information and Communication Technology (ICT) sector including data centres
generates up to 2% of the global CO
2
emissions, a number on par to the aviation sector
contribution, and data centres are estimated to have the fastest growing carbon footprint
from across the whole ICT sector, mainly due to technological advances such as the cloud
computing and the rapid growth of the use of Internet services. There are no recent
estimations of the total energy consumption of the European data centre and of their
energy efficiency. The aim of this paper is to evaluate, analyse and present the current
trends in energy consumption and efficiency in data centres in the European Union using
the data submitted by companies participating in the European Code of Conduct for Data
Centre Energy Efficiency programme, a voluntary initiative created in 2008 in response to
the increasing energy consumption in data centres and the need to reduce the related
environmental, economic and energy supply security impacts. The analysis shows that
the average Power Usage Effectiveness (PUE) of the facilities participating in the
programme is declining year after year. This confirms that voluntary approaches could be
effective in addressing climate and energy issue.
3
1 Introduction
According to Rong H et al. [5] data centres are computer warehouses that store a large
amount of data for different organisations in order to meet their daily transaction
processing needs. They contain servers for the collection of data and network
infrastructure for the utilisation and storage of the data [5]. Data centres usually run
24/7 all year round [1] and are very energy intensive with typical power densities of 538-
2153 W/m
2
and sometimes they can reach up to 10 KW/m
2
[11]. High energy
consumption can be attributed primarily to the IT demands and cooling equipment, as
well as lighting, power distribution and other requirements. The cooling system may
account for up to 40% of the energy demands of a data centre [13] [6] with the most
efficient systems using 24% of the total energy and the least efficient 61% [6]. This
power is distributed to chillers, cooling towers and water pumps. Water chillers consume
the most energy to supply chilled water to the cooling coil in order to keep the indoor
temperature low enough by removing the heat emitted by the servers [13]. Hence, the
improvement of the cooling system efficiency can lead to great energy and financial
savings. However there is no appropriate legislative framework.
There are European policy actions for buildings such as the 2010 European Energy
Performance of Building Directive (EPBD) [48] which imposes on the EU Member States
to adopt minimum efficiency requirements for buildings based on cost-optimality in their
building codes. In addition to that Directive, there is an agreement between the EU and
the US on the shared use of the Energy Star labelling programme, whose equipment is
available in Europe and promoted by public authorities and their procurement practices.
Furthermore, the Eco-design Directive introduces minimum efficiency requirements for
end-use equipment such as domestic appliances, lighting products, electronics, UPS, air-
conditioners, computers etc. Even though the Directive covers individual equipment that
is used in data centres, the overall efficiency of the facility is not guaranteed [Bertoldi].
Generally, there are no European policies that introduce specific mandatory efficiency
requirements for data centres.
Data centres have been designed to allow operational and capacity changes as well as
expansions. In addition, most of them nowadays run significant quantities of redundant
power and cooling systems to provide higher levels of reliability. Additionally, IT systems
frequently run at a low average utilisation. Over-provisioning, ensuring availability and
associated costs were previously considered as a negligible risk to business performance
because energy costs were relatively small in comparison to the IT budget and
environmental and energy responsibility was not considered to be under the control of
the IT department. However, with rising energy prices this is no longer the case, and the
issue of energy consumption at the individual data centre level is becoming increasingly
important as operational energy expenditures and environmental impact of the energy
consumed begins to play an ever important role in overall cost of ownership of data
centres [Bertoldi] [Participant Guidelines].
The energy consumption in the ICT (Information and Communication Technology) sector
has increased exponentially over the last years, mainly due to technological advances
such as the cloud computing and the rapid growth of the use of Internet services [1]. The
data centre sector in particular is estimated to account for the 1.4% of the global
electricity consumption [2] [5], and the compound annual growth rate (CAGR) of this
consumption in the period between 2007 and 2012 has been estimated as 4.4%. This
number is much higher that the projected 2.1% increase in global demand from 2012 to
2040 [2]. The ICT sector including data centres generates up to 2% of the global CO
2
emissions, a number on par to the aviation sector contribution [3][4], and data centres
are estimated to have the fastest growing carbon footprint from across the whole ICT
sector [4]. In addition, real-time video streaming, online gaming as well as mobile
devices already account for 60% of all data traffic, and it is predicted that this will rise to
80% by 2020. In general, the ICT sector nowadays consumes approximately 7 % of the
global electricity, and it is forecasted that the share will rise up to 13% by 2030 [14].
4
Climate change might not only affect the ICT sector as a whole but data centres are
particularly vulnerable to climate hazards due to the impacts on business continuity.
Acclimatise, a climate risk management consulting company, published a report
describing a number of potential risks that can be attributed to climate change [16]:
● More frequent heat waves and high temperature spikes add extra burden on
cooling equipment;
● Reduced efficiency and increased mechanical failures as a result of the increased
temperatures affecting the manufacturing parameters of the equipment;
● Power failures affecting the local grid due to increased supply demands for cooling
purposes during heatwaves;
● Restricted water supply during more frequently occurring droughts;
● Sea level rise, increased river flood or 'flash' flooding may damage the equipment
and lead to data loss due to the drainage systems blocked by heavy precipitation;
● Direct impacts on supply chains and infrastructure and loss pf business continuity;
● Employees' mobility affected by adverse weather-related events.
There are not any recent estimations of the energy consumption of European Data
Centres or of the average PUE. The aim of this report is to evaluate, analyse and present
the current trends in energy efficiency in data centres using the data submitted by
companies participating in the European Code of Conduct for Data Centre Energy
Efficiency.
5
2 European Trends
According to the Energy Efficiency Status Report published by European Commission's
Joint Research Centre in 2012, the total data centre energy consumption in 2007 was
estimated as 56 TWh (2% of the overall energy consumption in the EU annually), and it
is projected to grow to 104 TWh/ year (4%) by 2020 [8].
These figures are not too far off from Koomey's [9] estimations regarding the annual
electricity consumption in Western Europe for 2000 and 2005. He estimated the
consumption to be 18.3 TWh in 2000 and 41.3 TWh in 2005. Also, assuming an annual
12% growth rate the 2010 energy consumption was estimated as 72.5 TWh.
According to DataCentreDynamics, the total data centre energy consumption in the EU is
estimated as 48 - 55 TWh per year. Also, the same source estimates the power usage to
be 5,5-6 GW and the net data centre space as 6,0-6,5 million m2 (N. Parfitt, personal
communication, December 4, 2016).
Another study conducted by Broadgroup in 2014 provides estimations and future
projections regarding the net space of data centres, power density and usage of
European data centres:
Table 1. Data centres in Western Europe
Western Europe 2013 2014 2015 2016 2017 2018 2019 2020
Net data centre
space (thousands
of m
2
)
102,056
10221 10,105
10,055
9,875 9,555 9,365 9,155
Average power
density (kW/m2)
1.1 1.1 1.2 1.2 1.3 1.3 1.2 1.3
Total power usage
(GW)
11.3 11.2 12.1 12 12.8 12.4 11.3 10.9
Source: Broadgroup 2014
Table 1 shows that the net data centre space has been steadily declining throughout the
years and is projected to decline more in the near future. The average density and total
power usage trends show fluctuations, however, the power usage is predicted to fall.
Assuming 8,000 operating hours per year, it is estimated that energy consumption
reaches a peak in 2017 (102 TWh) and is projected to drop to 87 TWh until 2020.
It must be noted that some of the data presented in the table above contradict the
aforementioned data provided by DataCentreDynamics and the Energy Efficiency Status
Report. This is to be expected since all studies provide estimations and future projections
based on different models and data collection methodologies.
Also, the average PUE in European data centres in 2012 was reported to be 2.62 and in
2013 it declined to 2.53 [17]. However, it is rather difficult to determine a certain figure
that is widely accepted to quantify the efficiency of European data centres.
2.1 Data centre markets
There are four dominant data centre markets in Europe: UK, Germany, France and
Netherlands.
Frankfurt is well connected to Central and South East Europe, whereas the Amsterdam
market connections extend to Scandinavia and Russia. London and Frankfurt being the
largest financial business centres naturally have the strongest industries [18].
6
The latest available data cover the third quarter of 2016 and calculate the total power
supply in the four big markets as 879 MW since the start of the current year. As for the
individual supplies, they were reported in the CBRE Q3 2016 Report [19] as follows:
● London: 384MW
● Frankfurt: 199MW
● Amsterdam: 166 MW
● Paris: 129MW
2.1.1 London Market
The British data centre market, being the largest one, is well reported and analysed.
According to TechUK's Emma Fryer, it is estimated that the data centres industry's share
cover 0.6% of the total energy consumption in the UK. In general, the electricity
consumption in 2014 was the lowest since 1995, 4.3% down comparing to the previous
and it is part of a general declining trend [20].
CCA (Climate Change Agreement) for Data Centres came into force in July 2014 and is a
voluntary scheme that promotes energy efficiency strategies. The target is to reduce the
PUE by at least 15% for the base year over the life of the scheme (it runs until 2023 and
is spread over four target periods) [21].
The latest available data cover the period up to the end of 2014 and refer to colocation
data centres only. However, the sites that are not in the scheme ate relatively small and
the additional power consumption will not affect the overall number substantially. The
total energy usage of the 98 facilities participating in the scheme in 2014 was 2 TWh. It
is estimated that the whole sector accounts for 2.2-2.5 TWh a year. (E. Fryer, personal
communication, November 28, 2016).
More specifically, the total annual electricity that went through the scheme until 2014
was 1,995,810 MWh and the total IT electricity based on base year data was 1,015,360
MWh.
In addition, the average PUE of the reporting facilities between base years 2011-2014 is
1.93 [21].
2.1.2 Frankfurt Market
As for the Frankfurt market, the second largest in Europe, the increased digitization in all
sectors leads to growing energy demands. The 2015 report published by the Borderstep
Institute reported that the 2015 total energy consumption is estimated up to 12 TWh,
and the future projections for 2020 and 2025 are 14 TWh and 16.4 TWh respectively, as
demonstrated in Figure 1 [22]. Frankfurt became the first data centre market to reach
30MW of colocation take-up in less than one year, a record that no other market has
reached before [23].
7
Hintemann and Clausen (2014) estimated the number of data centres in Germany and
categorized them in relation to the size in square feet. They also determined the growth
or decline of their numbers in comparison to 2008 [24]. The table below shows their
findings:
Figure 1. Energy consumption of servers and data centres in Germany from 2010 to 2015 and
forecast to 2025
Source: Hintemann, 2015
Table 2. Number of data centres by size category and growth rates
Data centres category No of data centres
(2013)
Change in No of data
centres in 2008-
2013
Server Cabinets (3-10 m
2
) 30500 -8%
Server Rooms (11-100 m
2
) 18100 +/- 0%
Small Data centres (101-500 m
2
) 2150 +23%
Medium Data centres (501-5000 m
2
)
280 +27%
Large Data centres (over 5000 m
2
) 70 +40%
Source: Hintemann and Clausen, 2014
2.1.3 Paris Market
France has the third largest multi-tenant data centre and fourth largest hosting market in
Europe. The impacts of the economic crisis are still visible, and the multi-tenant data
centre market recovers slowly due to oversupply and slow demand in the market [25].
However, it was possible to retrieve more detailed data about the Paris data centre
market.
2.1.4 Amsterdam Market
Amsterdam ranks fourth amongst the European markets. The 2015 report published by
the Dutch Data Centres Association provides an overview of the Dutch multi-tenant sites
and focuses on data centres that rent out space in the form of housing or colocation.
Single tenant data centres that house server racks for internal use are, as well as huge
sites, were not included in the survey, therefore the final estimations may be
underestimated. The total power capacity is 350MW and the average self-reported PUE of
8
the facilities that were surveyed was 1.31 [26]. Nowadays, Amsterdam holds the 20% of
the European key data centre market. It is also projected that until 2020 there will be a
10-20% reduction of the ICT-related electricity consumption relative to 2013 [27].
Specifically, the 2013 energy consumption in the Amsterdam area only was reported as
0.46 TWh [18] whereas the national consumption for the same year is estimated as 1.4
TWh [27].
According to the 2016 report of the Dutch Data Centre Association, there are over 200
sites across the country and they cover approximately 271000 m
2
of total net data centre
surface. Another estimation, provided by Le Fevre and Leclercq in 2013 predicted the net
space growth in 2016 up to 385000 m
2
and the capacity to 500 MW taking into account
the consumption of main-process supporting data centre within companies, hence the
higher numbers [18].
2.1.5 Other Markets
Other general trends in the European sector include the Nordic market (Norway, Sweden,
Finland, Denmark and Iceland) which covers the 9.52% of all Europe. Scandinavian cold
climates turn out to be the ideal locations for data centres, offering low temperatures
that cut down the cooling energy consumption and costs, as well as low-cost, reliable
green power such as hydropower and thermal power [28]. Indicatively, Europe's largest
data centre is located in Måløy, Norway and it covers 120,000 m
2
. Seawater is used as
cooling resource as well as a renewable energy source [29].
Ireland is attracting US firms such as Apple and accounts for the 3.31% of all Europe.
Finally, there is a projected increase in outsourcing by 42.3% by 2018, comparing to
21.2% in 2013. The British companies are generally more prone to using third-party
outsourcing for their needs [30]. Ireland offers a combination of benefits that has
attracted foreign investments and particularly some of the industry giants such as
Amazon, Microsoft and Google. The cold and damp weather, suitable for reduced cooling
costs, in conjunction with the improved connectivity with the UK, rest of Europe and the
US, and last but not least, the accommodating low tax rates in comparison to the main
antagonists, UK and France, indicate a business-ready country [Jones, 2014].
Outside of the EU, US still accounts for 45% of the major cloud and internet data centre
sites globally, followed by China (8%), Japan (7%), Australia and Canada (4%). Other
non EU countries than have smaller shares include Singapore, Brazil, India and Hong
Kong [31]. In 2014, American data centres consumed an estimated 70 TWh, covering
approximately 1.8% of the total annual US electricity consumption. There was a 4%
increase in the consumption between 2010-2014, even though is significantly lower than
the 24% increase in 2005-2010, and the exponential 90% growth that was noted during
2000-2005. It is expected that by 2020 the US facilities are projected to consume 73
TWh, considering a steady 4% consumption growth rate [32].
According to Cloudscene, an online connectivity directory for colocation centres and cloud
service providers, there are over 6,122 colocation data centres around the world at the
moment [49]. 2015 global energy data centre consumption was estimated 416 TWh and
accounts roughly for 3% of the globally generated power and 2% of the global production
of greenhouse gas emissions [33].
9
Figure 2. Global Hyperscale Data Centre Operators
Source: Synergy Research Group.
10
3 European Code of Conduct for Data Centre Energy
Efficiency
In order to address complex energy efficiency issues, policy makers can either establish
mandatory regulations or voluntary agreements. However, there are sectors that are
especially difficult to regulate with many decision makers involved, therefore the
voluntary approach is generally preferred, especially between public authorities and
private enterprises [34].
The Codes of Conduct (CoC) are European voluntary programmes for the ICT sector that
were introduced since 2000. At the moment there are five CoC for ICT products in place:
External Power Supplies, Digital TV Systems, Broadband Equipment and UPS. All the
aforementioned Codes set specific efficiency requirements for specific products on a
voluntary basis, but once a company participates, they have to meet the performance
levels and report the energy consumption of their products annually. The fifth and most
recent CoC for Data Centres employs the same strategy [8]; however, it is not possible
to set a minimum efficiency requirement for data centres, given the diversity of data
centres, the different level of responsibilities (some company being only responsible for
the infrastructure, while other being responsible for the IT equipment selection and
operation). Therefore it was decided that the key criteria for the Data Centre CoC was to
ask participating companies to monitor their energy consumption and to adopt a set of
established best practices [35].
The European Code of Conduct for Data Centres programme is a voluntary initiative
managed by the Joint Research Centre (JRC), the European Commission's in-house
science service. It has been created in response to the increasing energy consumption in
data centres and the need to reduce the related environmental, economic and energy
supply impacts. The aim is to inform and stimulate operators and owners to reduce
energy consumption in a cost-effective manner and without hampering the critical
function of data centres. In addition, it provides the platform to bring together European
stakeholders to discuss and agree voluntary actions which will improve energy efficiency
following European conditions such as the climate and energy markets regulations.
The Code addresses primarily data centre owners and operators, and secondly the supply
chain and service providers. It is a "multipurpose" programme, allowing different
stakeholders to commit to improve efficiency in their own areas of competence. The
primary target of the CoC is the data centre owner/operator, who is encouraged to
commit to undertake and implement energy efficient solutions in existing or new data
centres, whilst respecting the life cycle cost effectiveness and the performance
availability of the system.
The energy saving focus of the Code of Conduct covers two main areas:
1. IT Load – this relates to the consumption of the IT equipment in the data centre.
2. Facilities Load – this relates to the mechanical and electrical systems that support the
IT electrical load [35].
For the purposes of the CoC, the term "data centres" includes all buildings, facilities and
rooms which contain enterprise servers, server communication equipment, cooling
equipment and power equipment, and provide some form of data service (e.g. large scale
mission critical facilities all the way down to small server rooms located in office
buildings) [8].
In [35] it is stated that the CoC aims to:
● Develop and promote a set of easily understood metrics to measure the current
efficiencies and improvement.
● Provide and open process and forum for discussion representing European
stakeholder requirements.
11
● Produce a common set of principles to refer to and work in coordination with other
international initiatives.
● Raise awareness among managers, owners, investors, with targeted information
and material on the opportunity to improve efficiency.
● Create and provide an enabling tool for industry to implement cost-effective
energy saving opportunities.
● Develop practical voluntary commitments which when implemented improve the
energy efficiency of data centres and in so doing minimise the total cost of
ownership.
● Determine and accelerate the application of energy efficient technologies.
● Foster the development of tools that promote energy efficient procurement
practices, including criteria for equipment based on the Energy Star programme
specifications, and other Codes of Conduct.
● Monitor and assess actions to properly determine both the progress and areas for
improvement.
● Provides reference for other participants.
The methodological core of the programme is a Registration form, signed by the
Participant, in which they commit to:
● Conduct an initial energy measurement and energy audit to identify the major
energy saving opportunities.
● Prepare and submit an action plan. Once the action plan is accepted the
Participant status will be granted
● Implement the action plan according to the agreed time table. Energy
consumption must be monitored regularly, to see overtime progresses in the
energy efficiency indicator related to the data centre.
To apply as a Participant the company must determine which of your data centres will be
included, whether it meets the criteria for a corporate level Participant and whether it has
full of partial control of the data centres.
In addition to the main CoC signing form the applicant shall complete a Reporting Form
for each data centre;
● Complete the data centre information tab
● Assess their energy metering capability to ensure that they can meet the
reporting requirements
● Enter one month of metering data on the electricity data tab
● Audit their compliance with the Best Practices
● Implement any of the minimum expected Practices that are not already in place in
each data centre
● Complete the Best Practices tab.
Energy is measured in line with the reporting requirements. At a minimum, the facility
energy is measured at the utility for a stand-alone data centre, or the data centre sub
meter. In addition to the energy consumption, optional reporting criteria are also
included in the reporting form such as meters on individual parts of the data centre such
as the chiller systems and the CRAC units. Also, meters allowing the efficiency devices to
be measured such as Power to and from the UPS system.
All Participants have the obligation to continuously monitor energy consumption and
adopt energy management in order to look for continuous improvement in energy
efficiency. One of the key objectives of the CoC is that each participant benchmark their
12
efficiency overtime, using the CoC metric in order to produce evidence of continuous
improvement in efficiency.
It is understood that not all operators are responsible for all aspects of the IT
environment defined within the best practices. This is not a barrier to Participant status
but the participant can act as an Endorser for those practices outside of their direct
control. An example of this would be a collocation provider who does not control the IT
equipment shall actively endorse the practices relating to the IT equipment to their
customers in adopting those practices. Equally, an IT operator using collocation shall
request their collocation provider to implement the practices relating to the facility [36].
13
4 Methodology and Metrics
The European CoC uses the power utilisation effectiveness metric (PUE) which is used to
help operators understand a data centre's overall efficiency and reduce energy
consumption.
PUE is defined as the ratio of total data centre input power to IT load power. Higher the
PUE value lower is the efficiency of the facility as more "overhead" energy is consumed
for powering the electrical load. The ideal case id if the value of the PUE is 1 which
indicates the maximum attainable efficiency with no overhead energy. The ideal case is
not attainable at present due to the consumption of electricity by UPS, fans, pumps,
transformers, lighting and other auxiliary equipment in addition to the consuming IT Load
[37]. Characteristically, Koomey estimated a range of global PUE values from 1.25 to
3.75 with an average on 1.92, based on the EPA's (Environmental Protection Agency)
Energy Star programme survey on 61 data centres in 2010 [38]. In addition to that,
Uptime Institute published a survey in 2013 estimating the average PUE based on self-
reporting participants to be 1.65, a much improved performance comparing to 1.89 in
2011 and 2.1 in 2007 [39].
The total data centre energy is measured from the point where it is purchased; this
includes the electricity, chilled water, gas, fuel oil and other purchases made from a
utility. Electricity generated from renewable energy sources is also included in the total.
As for the IT Load, it should be measured in the output of the Power Distribution Unit
(PDU), or at a minimum at the output of the Uninterruptible Power Supply (UPS) [40].
The data used for the analysis were taken from the registration forms submitted by the
companies applying for the Participant status. In total, there are over 120 different
businesses participating with one or multiple facilities each from all over Europe and
some from the US. All data are confidential and are only used for research purposes.
Up to December 2016, 325 data centres have applied for the Participant status, the great
majority of them located in Europe. In the present report, the analysis was applied to a
reduced sample of 289 data centres that have been approved and have submitted
complete energy data so far.
14
5 Results and Analysis
5.1 PUE analysis
There are not many studies that investigate the energy performance and efficiency of
data centres based on real data, and the existing ones have used much smaller datasets,
estimations or are solely based on Koomey's research published in 2008 [9]. Ni and Bai
presented a review of different reports investigating the air conditioning energy
performance of 100 data centres. Indicatively, one of the reports included in the review
assessed 44 data centres around the world and presented an average PUE of 2.29,
whereas a second one assessed 23 facilities in Singapore of average PUE 2.07 [6].
As shown in Table 3 with the average data of our study, the average PUE of the 289
approved facilities is 1.80. All of them have submitted their electricity and IT data,
reported either on an annual or monthly basis. In many cases, the energy entries do not
cover the required 12-month reporting period, so the annual consumption was
extrapolated based on the available data in order to calculate the PUE. Comparing to the
previous study presented by Bertoldi in 2014 [8], the total reported annual electricity
consumption has risen by 0.5 TWh reflecting the larger number of data centres
participating in the programme. The average floor area of the facilities as well as the
average Rated IT load has not changed significantly, indicating a general homogeneity of
the sites during the reporting years. Interestingly, the average annual electricity
consumption has decreased by more than 700 MWh, suggesting the adoption of better
practices and more energy efficient systems. As for the temperature and relative
humidity (RH) set points, they have not changed greatly comparing to the 2014 analysis;
the temperature set point range has become slightly wider whereas the relative humidity
is now narrower. However, it was impossible to correlate those changes with the average
PUE. It would be interesting to monitor if the data centre operators comply with the
ASHRAE thermal guidelines for higher operating temperatures within the facility.
Figure 3 illustrates the distribution of the PUE values, excluding the outliers, which are
PUE values less than 1.0 and over 3.0. A value less than 1.0 is impossible and indicates
higher IT consumption that the overall energy consumption of the facility, thus incorrect
data. On the other hand, values over 3.0 even though considered outliers to maintain
cohesion in the dataset, may indicate newly-constructed facilities or old data centres with
inadequately implemented or not yet in place energy efficient technologies and
management. Therefore, the PUE values of 268 facilities were taken into consideration.
There were 12 outliers in total; 3 incorrect entries with PUE<1.0, and 9 entries with
PUE>3.0. The remaining 9 entries were reported incorrectly, either with missing data or
in a wrong way, for instance reporting energy and IT data from different reporting
periods. All in all, most of the facilities have reported an average PUE of 1.6-1.8, followed
by the 1.8-2.0 range. In addition there are 4 sites that reported excellent efficiencies
between 1.1-1.2.
Table 4 also shows the total number of data centres that were assessed and received the
Participant status each year since the beginning of the programme. Most of the facilities
were approved in 2011 and 2013, and both years demonstrate a good PUE average.
However, not all facilities have submitted correct and reliable energy data, so the
average PUE was estimated taking into account a slightly reduced sample of sites.
15
Table 3. Average data of reporting facilities
Figure 3. PUE values ranges.
356
15
27
68
73
40
27
4
0
10
20
30
40
50
60
70
80
2.8-3.0 2.6-2.8 2.4-2.6 2.2-2.4 2.0-2.2 1.8-2.0 1.6-1.8 1.4-1.6 1.2-1.4 1.0-1.2
Number of Data Centres
PUE ranges
Total dataset 289
Total annual electricity consumption 3,735,735 MWh
Average DC floor area 2,616 m
2
Average Rated IT load 1,956 kW
Average annual electricity consumption 13,684 MWh
Average annual IT consumption 7,871 MWh
Average PUE 1.80
Average High Temp Set point 25 Degree
o
C
Average Low Temp Set point 19.5 Degree
o
C
Average High RH Set point 59 % RH
Average Low RH Set point 35 % RH
16
Table 4. Average PUE by reporting year
Figure 4 illustrates the fluctuations of the average PUE annual estimations throughout the
reporting years, from 2009 when the first data centres were approved and submitted
their energy data until the time of writing, beginning of 2017. As shown in the graph,
there is a clear declining trend ranging from 1.96 in 2010 which is the higher average
PUE value throughout the reporting years, to 1.64 in 2016 which is the lowest. The
declining PUE trend can be attributed to more efficient and advanced cooling technologies
and will be investigated in comparison to the types of economisers used widely and other
parameters such as the location of the data centres. However, all sites are self-reported
and not all operators have submitted reliable energy data; for example, in some cases
there are either missing data or data that indicate a PUE less than 1, which is impossible
by definition according to [37] and [40]. Therefore the average PUE was estimated taking
into account a slightly reduced sample of data centres. It is also noteworthy that all of
the average values thought the years are below 2.0.
Each pair of figures below (Figures 5-6, 7-8 and 9-10) demonstrates the correlation
between the average PUE and the data centre size, age of facility and IT capacity,
respectively.
It is shown that more than half of the data centres participating in the CoC contain server
room between 1000-5000 m
2
each and can be classified as medium sized. However, the
statistical analysis shows that the medium sized sites perform worse that the smaller or
the very big facilities, reaching a PUE value of 1.84, comparing to 1.77 of the other size
categories. In addition, as indicated in Figure 8, the facilities that were built between
2005 and 2010 are more energy efficient, comparing to older buildings or even those
built within the last 5-year period that show an important decline in efficiency (PUE is
1.88 on average). It is also interesting to note that most of the participating facilities fall
under this construction age category according to Figure 7. The low PUE values of the
older data centres are not statistically reliable due to the very small number of facilities
that were constructed during that period. Since the 1980s the PUE has improved due to
the energy efficiency measures.
Finally, IT Rated Load is the peak power that can be provided to the IT equipment,
measured in kW. The analysis shows that most facilities with 5000-1000 kW (5-10 MW)
IT Rated Load appear to perform better than the rest, followed by the 0.1-0.3 MW sites.
However, the intense fluctuations as illustrated in the graph, suggest that the IT capacity
has no actual impact of the efficiency of the facility.
Reporting year Number of Data Centres Average PUE
2009 33 1.86
2010 19 1.96
2011 58 1.73
2012 41 1.90
2013 54 1.78
2014 27 1.86
2015 24 1.72
2016 30 1.64
17
Figure 4. Average PUE by reporting year
Figure 5. Number of data centres by size classification
1.86
1.96
1.73
1.90
1.78
1.86
1.72
1.64
1.50
1.55
1.60
1.65
1.70
1.75
1.80
1.85
1.90
1.95
2.00
2009 2010 2011 2012 2013 2014 2015 2016
PUE
Application Year
45
44
91
70
24
14
0 10 20 30 40 50 60 70 80 90 100
1-500
500 -1000
1000-2500
2500-5000
5000-8000
8000-20000
Number of Data Centres
Size (m2)
18
Figure 6. Average PUE values for different size classifications
Figure 7. Number of data centres per year of construction
1.76
1.77
1.81
1.84
1.77 1.77
1.72
1.74
1.76
1.78
1.8
1.82
1.84
1.86
1-500 500 -1000 1000-2500 2500-5000 5000-8000 8000-20000
PUE
Size (m2)
13
23
11
42
63
81
48
0 10 20 30 40 50 60 70 80 90
1960-1980
1980-1990
1990-1995
1995-2000
2000-2005
2005-2010
2010-2015
Number of Data Centres
Construction period
19
Figure 8. Average PUE values for different construction years
Figure 9. Number of data centres per IT Rated Load (kW)
1.78
1.91
1.82
1.78 1.77
1.74
1.88
1.65
1.7
1.75
1.8
1.85
1.9
1.95
1960-1980 1980-1990 1990-1995 1995-2000 2000-2005 2005-2010 2010-2015
PUE
Construction period
17
36
26
43
91
44
17
5
0 10 20 30 40 50 60 70 80 90 100
0.1-100
100-300
300-600
600-1,000
1,000-2,500
2,500-5,000
5,000-10,000
10,000-20,000
Data Centres
20
Figure 10. Average PUE values for different IT Rated Load classifications (kW)
5.2 Sector Distribution Analysis
The Code of Conduct distinguishes between seven different types of data centres, as
described in the Reporting form for Participants [35]:
1. Traditional Enterprise: Processes data requirements for data centre owner
2. On Demand Enterprise: Processes and expands data processing capacity, when
needed, for numerous customers (this would include batch processing and
manufacturing type data processing facilities)
3. Telecom: Telecom Switching Centre
4. High Performance Computing Cluster (HPCC): Scientific and high density data
processing
5. Hosting: Sells data processing services to numerous Customers
6. Internet: Provides high capacity processing for large numbers of web clients
7. Hybrid: Combination of two or more of the above
Figures 11 and 12 present the distribution of the data centres by the types of sectors. It
is observed that more than one third of the reporting facilities belong to the Traditional
Enterprise sector, the type of facility that processes data requirements for data centre
owners. In this type of business model the ownership of the facility, IT equipment and
software systems are common [1]. The Hosting sector comes second followed by the
Hybrid. In a Hosting type of facility, the ownership and the IT equipment is common but
the software systems are run by others [1]. In addition, the majority of the data centres
(approximately 61%) are stand-alone facilities.
1.73
1.71
1.82
1.88
1.79
1.81
1.67
1.8
1.55
1.6
1.65
1.7
1.75
1.8
1.85
1.9
PUE
IT Rated Load (kW)
21
Figure 11. Percentage of each of the seven different sectors
Figure 12 . Distribution of the seven types of sectors participating in the CoC
5.3 Geographical Distribution Analysis
Weather parameters such as the ambient temperature and relative humidity can have a
significant impact on the energy consumption. A data centre located in an area with
40
1
8
2
28
1
20
1. Traditional
Enterprise
2. On-demand
Enterprise
3. Telecom
4. HPCC
5. Hosting
6. Internet
0
50
100
150
200
250
300
350
2009 2010 2011 2012 2013 2014 2015 2016
Number of Data Centres
Traditional Enterprise On-Demand Enterprise Telecom
HPCC Hosting Internet
Hybrid
22
extreme temperatures and humidity is expected to consume more energy as the cooling
system will be working harder to maintain stable operating conditions within the facility.
The table below shows the distribution of the data centres of our database by
geographical location. The results were split into five different zones: The Nordic zone,
the UK and Ireland zone, the Northern/Central European zone, the Southern
European/Mediterranean zone and finally the Non-European zone. The last one is not
assessed against the others or included further into the analysis, since the available data
are not adequate and comparable.
The Nordic countries and the countries of Continental Europe appear to be the most
efficient in terms of power effectiveness. This can be explained by the advanced
technological progress in the sector as well as by the cold climate conditions that reduce
the cooling energy demands of the facilities.
On the other hand, the data centres located in countries of the European South have an
average PUE of 2.00, a figure that can be explained by the warmer climate.
The countries in Northern and Central Europe achieve as high standards as the Nordic
countries. This is to be expected, considering that three out of the four biggest data
centre markets are located in Frankfurt, Paris and Netherlands, hence the high number of
sites under this category.
Finally, London is the biggest data centre market in Europe and Dublin is one of the
newly emerging markets, with many US companies building data centres there. This fact
explains the high number of data centres in that zone and the technological progress.
Table 5. Geographical zoning with temperature and relative humidity average data
Geographical
Zones
Countries Temperature
range (
o
C)
RH range
(%)
Average
PUE
No of
Data
Centres
Nordic countries Denmark, Finland, Norway,
Sweden
18-26 20-80 1.71 13
UK and Republic
of Ireland
England, Scotland, Wales,
Northern Ireland, Republic
of Ireland.
17-30 8-80 1.83 116
Northern/Central
Europe
Austria, Belgium, France,
Germany, Hungary,
Luxembourg, Netherlands,
Portugal, Poland,
Switzerland
14-28 16-75 1.72 122
S. Europe/
Mediterranean
Gibraltar, Greece, Italy,
Malta, Spain, Turkey,
Monaco, Romania, Bulgaria
16-26 20-80 2.00 30
Non EU Republic of Mauritius, US - - - 5
23
Figure 13. Average PUE values by geographical zone
Figure 14. Geographical Distribution of the participation data centre in the EU CoC
Figure 14 depicts the distribution of the participating European data centres. The facilities
located in the US and Mauritius have been omitted. It is observed that as per the analysis
1.55
1.6
1.65
1.7
1.75
1.8
1.85
1.9
1.95
2
2.05
Nordic countries UK and Republic of
Ireland
Northern/Continental
Europe
Mediterranean/S.
Europe
PUE
Geographical Zones
24
outlined in Table 2, the places with the greater data centre densities are the London,
Amsterdam, Frankfurt, Paris and Milan areas and the surroundings, where in fact, the
largest data centre marktes in Europe are based.
5.4 Free Cooling Technologies
Free cooling, also known as economizer cycle, takes advantage of the outside dry and
cool climate conditions to cooling cost savings through the refrigeration compressor
workload reduction required to operate the chillers [13]. When the ambient temperature
is sufficiently lower than the facility temperature, the heat naturally flows to the outside
without the use of vapour-compression refrigeration system, offering significant energy
savings [41]. Free or economised cooling designs use cool ambient conditions to meet
part or all of the facilities cooling requirements hence compressor work for cooling is
reduced or removed, which can result in significant energy reduction. Economised cooling
can be retrofitted to some facilities. The opportunities for the utilisation of free cooling
are increased in cooler and dryer climates and where increased temperature set points
are used. Where refrigeration plant can be reduced in size (or eliminated), operating and
capital costs are reduced, including that of supporting electrical infrastructure [42].
According to [43], there are four fundamental types of free cooling technologies that
operate in the facilities either individually or more often, combined:
● Direct Air (DA) free cooling: External air is used to cool the facility. Refrigeration
systems are present to deal with humidity and high external temperatures if
necessary. Exhaust air is re-circulated and mixed with intake air to control supply
air temperature and humidity.
This design tends to have the lowest temperature difference between external
temperature and IT supply air.
● Indirect Air (IA) free cooling: Re-circulated air within the facility is primarily
passed through an air to air heat exchanger against external air (may have
adiabatic cooling) to remove heat to the atmosphere.
This design tends to have a low temperature difference between external temperature
and IT supply.
● Direct Water (DW) free cooling: Chilled water cooled by the external ambient air
via a free cooling coil. This may be achieved by dry coolers or by evaporative
assistance through spray onto the dry coolers.
This design tends to have a medium temperature difference between external
temperature and IT supply air.
● Indirect Water (IW) free cooling: Chilled water is cooled by the external ambient
conditions via a heat exchanger which is used between the condenser and chilled
water circuits. This may be achieved by cooling towers or dry coolers, the dry
coolers might have evaporative assistance through spray onto the coolers.
This design tends to have a higher temperature difference between external temperature
and IT supply air restricting the economiser hours available and increasing energy
overhead.
Figure 13 provides a graphic representation of the number of sites that operate each of
the aforementioned technologies. The analysis was performed in 288 facilities that have
been granted the Participant status and submitted their best practices description.
25
Figure 15. Distribution of the sites that use a combination of free cooling technologies
Table 6. PUE average data for each free cooling technique
Generally, the air-side economizers are used in data centres located in places with cold
and dry weather conditions but are not suitable for hot and humid areas [44].
More than half of the data centres (59%) either have no free cooling system in place, or
have not reported theirs in the CoC application form.
Also, most of the data centres use a combination of free cooling technologies, instead of
an individual one.
Figure 14 and Table 5 show that the facilities that use four out of five economiser
technologies under the EU Code of Conduct, achieve an average PUE between 1.75 and
1.89. The fifth type "Adsorption and other sorption techniques" has been reported by
only one data centre, so it will not be included in the analysis considered an outlier.
The graph indicates that the most efficient cooling technology is the Indirect Water (IW)
type, since the average PUE is the lowest and the minimum PUE reported in 1.10 which is
the closest to ideal of 1.0, but the range of PUE is wider comparing to the others. On the
other hand, Direct Water (DW) technique shows a narrower range of PUE values.
32 34 39
70
14
170
0
20
40
60
80
100
120
140
160
180
DA IA DW IW A None
Number of Data Centre
Type of Economizer
Economiser Type Average PUE for individual Min PUE Max PUE
DA 1.88 1.25 2.71
IA 1.89 1.24 2.55
DW 1.83 1.27 2.33
IW 1.81 1.1 2.73
A 1.75 1.75 1.75
None 1.81 1.1 2.83
26
It is noted that even data centres that do not use a free cooling system or have achieved
very low PUE in some cases. The PUE range of this type of data centres is the widest,
ranging from 1.1 to 2.83.
Figure 16. PUE range by type of free cooling technology
It is worth mentioning that some of the facilities use Kyoto wheels as an alternative
mechanical system type apart from direct expansion (DX) and water chillers and as an
additional cooling technique apart from the economizers. The idea is simple: a wheel
which can be as big as 6 meters in diameter picks up the heat from the data centre while
rotating and releases it to the cooler outside environment as it completes the rotation.
The warm air is evacuated with the use of fans [45].
5.5 Best practices
The Best Practices document is a supplement to the Code of Conduct that provides
information and support for data centre operators to implement measures to improve
energy efficiency on their facilities.
1.88 1.89 1.83 1.74
1.35
1.81
1.25 1.24 1.27
1.1
1.35
1.1
2.71
2.55
2.33
2.73
1.35
2.83
0
0.5
1
1.5
2
2.5
3
DA IA DW IW A None
PUE
Type of Economizer Average PUE Min PUE Max PUE
27
A subgroup of practises has been established to signify the minimum expected level of
energy saving activity that is required to be granted the Participant status. These
expected practices are divided into five categories and each is signified by a different
colour:
Table 7. Minimum expected and optional Best practices categories.
Category Description
Entire Data Centre Expected to be applied to all existing IT, Mechanical and Electrical
equipment within the data centre.
New Software Expected during any new software install or upgrade
New IT Equipment Expected for new or replacement IT equipment
New build or retrofit Expected for any data centre built or undergoing a significant refit of the
M&E equipment from 2011 onwards
Optional Practices Optional (no background colour)
It is understood that not all operators are able to implement all the expected practices
due to physical, logistical and other kind of constraints. In these cases, an explanation
should be provided describing the type of constraint, and if possible, recommending
alternative practices as replacements aiming the same energy savings [42].
Tables 8 and 9 show the most and less implemented best practices in our dataset
respectively, along with the percentage of implementation per practice. Indicatively,
approximately 97% of the facilities reported that they produce periodic written reports on
energy consumption and environmental ranges, including the determination of PUE over
the reporting period. This report may be produced by an automated system.
It is noted that most of these practices require low capital expenditure or minimal
changes to business practices, thus are considered "low-hanging fruit".
On the other hand, there are practices that are implemented by less than 1% of the total
number of the data centres. In fact, most of them were introduced in the latest version
of the Best Practices Guidelines, published by the JRC in 2016 [42]. Thus, it remains to
be seen in a future study if the new data centres will adopt the new practices. In
addition, there are no practices that have been implemented by all 288 sites, even
though some of the most implemented are shown on Table 8.
It is also worth mentioning that each of the new versions of the guidelines contains a few
changes in some sections. In some cases the numbering has changed and depending on
the year of reporting the practice wording might be different. More specifically, section 5
that refers to the cooling system demonstrates significant changes in the practices
numbering between the 2014 – 2015 editions, along some crucial differences in the type
of free-cooling technologies. This fact hinders the analysis of the economisers as the
reporting changed according to the new version of guidelines.
28
Table 8. Ten top implemented Best Practices
Best Practice Brief Description Implementation (%)
9.3.1 Written report 97.2 %
5.2.3 Review of cooling before IT equipment
changes
96.5 %
9.1.1 Incoming energy consumption meter 95.8 %
3.2.1 Consider the embodied environmental
impact of installed devices
95.4 %
9.2.1 Periodic manual readings 94.8%
9.1.2 IT Energy consumption meter 94.4%
5.3.4 Review and if possible raise chilled water
temperature
94.4%
3.1.1 Group involvement 94.4%
5.1.2 Design – Contained hot or cold air 94.1%
5.2.4 Review of cooling strategy 94.1%
Table 9. Ten less implemented Best Practices
Best Practice Brief Description Implementation (%)
3.2.3 Service Charging Models 0.4 %
8.3.3 Metering of water consumption 0.4 %
9.1.9 IT Device level metering of energy
consumption
0.4 %
3.2.4 Design to maximise the part load efficiency
once provisioned
0.7 %
5.2.4 Review of cooling strategy 0.7 %
6.2.2 Power factor correction 0.7 %
5.6.4 Energy reuse metrics and reporting 2.4 %
4.2.6 Incentives to develop efficient software 2.8 %
8.2.5 Co-locate with power source 4.2 %
4.3.7 Control of system energy use 5.6 %
29
Table 10. Best Practices that have never been implemented
Best Practice Brief Description
3.2.5 Environmental Management
3.2.6 Energy Management
3.2.7 Asset Management
3.2.8 Sustainable Energy Usage
3.2.9 Powering of devices via the IT cabling
3.2.10 Impact of mobile/shifting workloads
3.2.11 Alternative power generation technologies
4.1.14 Operating temperature range-Direct liquid cooled IT equipment
5.1.14 Control of supplied air flow volumes
5.1.15 Installation of free cooling
5.4.1.6 Indirect water free cooling with condenser water cooling chilled water
5.4.1.7 Alternative cooling sources
5.4.2.9 Direct liquid cooling of IT devices
8.3.1 Capture rain water
8.3.2 Other water sources
9.4.4 Business relevant dashboard
The analysis shows that all 288 sites have implemented best practices since the
beginning of the programme, albeit different number of practices. This is expected, since
each version of the Best Practices guidelines contains a different number of mandatory or
minimum-expected guidelines comparing to the previous one. The 2016 version
incorporates 152 practices in total, 81 of which are characterised as minimum expected.
New practices are added to the expected list in each version and others are replaced or
amended.
Figure 16 illustrates how the best practices are implemented in the data centres. Each
bar represents the number of data centres that implement a certain range of practices
indicated in the horizontal axis. This is a qualitative analysis, so the number of practices
is evaluated rather than the type or category as referred in Table 7. Indicatively, the
majority of the participants implement between 26 and 50 best practices at the time of
reporting, and approximately one-third of the participants implement between 51 and 81
practices. As mentioned before, the 2016 guidelines list 81 practices as mandatory, so it
is concluded that there are only 16 businesses that have implemented the minimum
expected plus some optional practices, and 16 businesses that have only implemented
the 81 mandatory. It is important to specify that since older versions are differentiated
and contain fewer mandatory practices, it is expected that the 52-81 range is the most
appropriate to describe the average number of data centres that comply with the
practices criteria.
30
In general, the analysis shows that after 2015 the data centres that apply for the Code of
Conduct Participant status already implement more practices than those that participated
when the programme first started. This fact could be indicative of the technological
trends towards a more energy efficient way of operating a facility, with more significant
energy and financial savings.
Figure 17. Frequency of the best practices implementations
7
173
89
14
2
0
20
40
60
80
100
120
140
160
180
200
1-25 BP 26-50 BP 51-81 BP 82-100 BP 101-152 BP
Number of Data Centres
Number of Best practices
31
6 Conclusions
The study focuses on the emerging trends in energy efficiency in data centres, as
evaluated using the data submitted voluntarily by various companies under the Code of
Conduct for Data Centre Energy Efficiency. The number of participating data centres that
have claimed the Participant status is astounding having reached the 345 applications so
far. It must be noted that this is a unique study not only because of the size of the
database but also in terms of reliability, it is based on real data, instead of estimations.
The Code of Conduct is the only European independent programme that promotes data
centres energy efficiency best practices and monitors the energy consumption. It is
acknowledged as a successful example of a non-regulatory policy to improve energy
efficiency in data centres that has stimulated efficiency improvement is data centres.
Every year, the number of companies applying for the Participant Status increases.
The analysis shows that the average PUE of the facilities participating in the programme
is declining year after year, reaching the value 1.64 in 2016. The total energy
consumption so far of all the 289 approved Participants is approximately 3.7 TWh and
the average annual electricity consumption has declined since the last survey in 2014.
There is a number of facilities that already achieve PUE equal or below 1.2.
In addition, the PUE is compared against various parameters such as the facility size, the
construction year and the Rated IT Electrical Load. It is concluded that either the very
small or very large size data centres of the sample achieve the best PUE. As for the other
parameters, the facilities that were built in the period between 2005 and 2010 and have
an average of 5-10 MW capacity perform the best.
Furthermore, as expected, the data centres located in Scandinavia and Northern Europe
are more energy efficient, mainly due to the cooler ambient conditions that facilitate
economizer use. It derives that the type of economizer that contributes the most to
energy savings is the Indirect Water type, followed by the Direct Water.
Finally, regarding the best practices implementation, most data centre of the dataset
implement between 26 and 50 practices, significantly less that the mandatory number of
81, as required by the latest guidelines. However, the participants of the recent years
appear to perform better and implement more practices comparing to the past
participants, and most of those practices are the most cost-effective and with the
shortest payback period.
The future research task linked to the dataset is to track the efficiency improvements of
each data centre over time and examine the operational or infrastructure changes.
The Code of Conduct is one of the most successful policies to improve energy efficiency
in data centres in the EU, and it could be easily replicated in other countries and regions
based on the EU experience.
32
References
[1] Oró E, Depoorter V, Garcia A, Salom J. Energy efficiency and renewable energy
integration in data centres. Strategies and modelling review. Renewable and Sustainable
Energy Review 2015; 42: 429-445.
[2] Tatchell-Evans M, Kapur N, Summers J, Thompson H, Oldham D. An experimental
and theoretical investigation of the extent of bypass air within data centres employing
aisle containment, and its impact on power consumption. Applied Energy 2017; 186:
457-469.
[3] UNFCCC. ICT Helping Tackle Climate Change Could Help Cut Global Emissions 20%
by 2030. 2016. Available at: http://newsroom.unfccc.int/unfccc-newsroom/ict-sector-
helping-to-tackle-climate-change/ [Accessed: February 2017]
[4] Whitehead B, Andrews D, Shah A, Maidment G. Assessing the environmental impact
of data centres Part 1: Background, energy use and metrics. Building and Environment
2014; 82: 151-159.
[5] Rong H, Zhang H, Xiao S, Li C, Hu C. Optimizing energy consumption for data
centres. Renewable and Sustainable Energy Reviews 2016; 58: 674-691.
[6] Ni J, Bai X. A review of air conditioning energy performance in data centres.
Renewable and Sustainable Energy Reviews 2017; 67:625-640.
[7] Bertoldi P, Hirl B, Labanca N. Energy Efficiency Status Report 2012. Electricity
Consumption and Efficiency Trends in the EU-27. European Commission, Joint Research
Centre 2012. Available at:
http://publications.jrc.ec.europa.eu/repository/handle/JRC69638 [Accessed: November
2016]
[8] Bertoldi P. A Market Transformation Programme for Improving Energy Efficiency in
Data Centres. ACEEE Summer Study on Energy Efficiency in Buildings; 2014: 9.14-9.26
[9] Koomey, J. Worldwide Electricity used in Data Centres. Environmental Research
Letters 2008; 3: 8.
[10] Van Heddeghem W, Lambert S, Lannoo B, Colle D, Pickavet M, Demeester P. Trends
in worldwide ICT electricity consumption from 2007 to 2012. Computer Communications
2014; 50: 67-76.
[11] Beaty DL. Internal IT Load. Profile Variability. ASHRAE Journal 2013; 72-74.
[12] Capozzoli A, Primiceri G. Cooling Systems in data centres: state of art and emerging
technologies. Energy Procedia 2015; 83: 484-493.
[13] Khalaj A H, Scherer T, Halgamuge S K. Energy, environmental and economical
saving potential of data centres with various economizers across Australia. Applied
Energy 2016; 183: 1528-1549.
[14] Sadler R. Video demand drives up global CO
2
emissions. 2017, January 31.
Retrieved from: http://climatenewsnetwork.net/video-demand-drives-global-co2-
emissions/ [Accessed: February 2017]
[15] Directive 2010/31/EU of the European parliament and of the council of 19 May 2010
on the energy performance of buildings. Off. J. Eur. Union 2010, 3, 124–146.
[16] Acclimatise. Climate Change Risks to Data Centres; 2008.
www.acclimatise.uk.com/login/uploaded/resources/AcclimatiseDataCentres.pdf [Accessed
: February 2017].
[17] Peterson M. Is PUE Still Above 2.0 for Most Data Centres? 2015, February 19.
Message posted to: http://www.vertatique.com/no-one-can-agree-typical-pue
[Accessed: November 2017]
33
[18] Le Fevre S, Leclercq M. Data centres establishment policy. The Amsterdam region as
Green Data Port. 2013; Gemeente Amsterdam. Dienst Ruimtelijke Ordening.
[19] CBRE. CBRE Marketview. European Data Centres, Q3 2016.
[20] Fryer E. The Donald Runsfelf Guide to Data Centre Energy Consumption: The Known
Knowns, the Known Unknowns and the Unknown Unknowns. 2016, May; First published
by BroadGroup for DataCloud Europe 2016.
[21] TechUK. Climate Change Agreement (CCA) for Data Centres. Target Period One:
Report on findings. 2014; Retrieved from TechUK Website:
www.techuk.org/images/CCA_First_Target_Report_final.pdf [Accessed: November 2017]
[22] Hintemann R. Energy Consumption of Data Centres continues to increase – 2015
update. Borderstep Institute. 2015; Retrieved from Borderstep Institute Website:
https://www.borderstep.de/wp-
content/uploads/2015/01/Borderstep_Energy_Consumption_2015_Data_Centers_16_12_
2015.pdf [Accessed: November 2017]
[23] Frankfurt becomes the first data centre market to reach 30MW in a year. 2016,
November 28; retrieved from: https://digitalisationworld.com/article/50043/1 [Accessed:
November 2017]
[24] Hintemann R, Clausen J. Rechenzentren in Deutschland: Eine Studie zur Darstellung
der wirtschaftlichen Bedeutung und Wettbewerbssituation. Studie im Auftrag des
Bundesverbandes Informationswirtschaft, Telekommunikation und neue Medien e.V.
(BITKOM). Berlin. [Data centers in Germany: A study on the economic significance and
the competitive situation.] 2014; retrieved from:
https://www.bitkom.org/Bitkom/Publikationen/Rechenzentren-in-Deutschland-
Wirtschaftliche-Bedeutung-und-Wettbewerbssituation.html [Accessed: February 2017]
[25] Jones P, Duncan R. Executive Overview: Paris Multi-Tenant Datacentre and Hosting
Market. 2016; retrieved from: https://451research.com/report-
long?icid=3712&task=download&file=summary [Accessed: November 2017]
[26] Dutch Data Centre Association. Dutch Data Centre Report 2015. State of the Dutch
Data Centres. 2015; retrieved from: www.dutchdatacenters.nl/dutch-data-center-
report.html [Accessed: February 2017]
[27] Dutch Data Centre Association. Dutch Data Centre Report 2016. State of the Dutch
Data Centres. The new foundation. 2016; retrieved from:
www.dutchdatacenters.nl/dutch-data-center-report.html [Accessed: February 2017]
[28] Kontzer T. Data Centres Operators Flock to Cold Climates. 2013, September;
retrieved from: http://www.networkcomputing.com/data-centers/data-center-operators-
flock-cold-climates/1697156413 [Accessed: February 2017]
[29] Smolaks M. Nordic data centre market expected to triple power by 2017. 2015,
September 10; retrieved from: http://www.datacenterdynamics.com/content-
tracks/design-build/nordic-data-center-market-expected-to-triple-power-by-
2017/94767.fullarticle [Accessed: February 2017]
[30] Howard-Healy M. Data Centre Market Europe. 2016; retrieved from:
https://datacenter.eco.de/wp-content/blogs.dir/13/files/160119_broadgroup.pdf
[Accessed: February 2017]
[31] Synergy Research Group Hyperscale Data Centre Count Passes the 300 Milestone in
December. 2016, December 19; retrieved from:
https://www.srgresearch.com/articles/hyperscale-data-center-count-passes-300-
milestone-december [Accessed: February 2017]
[32] Shehabi A, Smith S, Sartor D, Brown R, Herrlin M, Koomey J, Masanet E, Horner N,
Azevedo I, Lintner W. United States Data Centre Energy Usage Report. 2016; retrieved
34
from: https://eta.lbl.gov/sites/all/files/publications/lbnl-1005775_v2.pdf [Accessed:
February 2017]
[33] Data Centres 'Going Green' to reduce a carbon footprint larger than the airline
industry. 2017, 27 January; retrieved from: https://data-economy.com/data-centers-
going-green-to-reduce-a-carbon-footprint-larger-than-the-airline-industry/ [Accessed:
February 2017]
[34] Rezessy S, Bertoldi P. Voluntary agreements in the field of energy efficiency and
emission reduction: Review and analysis of experiences in the European Union. Energy
Policy 2011; 39: 7121-7129.
[35] European Commission. Code of Conduct on Data Centres Energy Efficiency. Version
2.0. Participant Guidelines and Registration Form. 2009; Retrieved from:
http://iet.jrc.ec.europa.eu/energyefficiency/ict-codes-conduct/data-centres-energy-
efficiency [Accessed: February 2017]
[36] European Commission. EU Code of Conduct on Data Centre Energy Efficiency.
Introductory Guide for all applicants. Version 3.1.2. 2016; Retrieved from:
http://iet.jrc.ec.europa.eu/energyefficiency/ict-codes-conduct/data-centres-energy-
efficiency [Accessed: February 2017]
[37] Sharma M, Arunachalam K, Sharma D. Analysing the data center efficiency by using
PUE to make data centers more energy efficient by reducing the electrical consumption
and exploring new strategies. Procedia Computer Science 2015; 48: 142-148.
[38] Koomey J. Growth in Data Centre electricity use 2005-2010. A report by Analytical
Press, completed at the request of the New York Times; 2011. Available from:
http://www.analyticspress.com/datacenters.html [Accessed: February 2017]
[39] Uptime Institute. 2013 Data Centre Industry Survey. Uptime Institute, 2013.
[40] The Green Grid. Harmonizing Global Metrics for Data Centre Energy Efficiency.
2014; Available at: http://iet.jrc.ec.europa.eu/energyefficiency/ict-codes-conduct/data-
centres-energy-efficiency [Accessed: February 2017]
[41] Zhang H, Shao S, Xu H, Zou H, Tian C. Free cooling of data centres: A review.
Renewable and Sustainable Energy Reviews 2014, 35; 171-182.
[42] Newcombe, L., Acton, M., Bertoldi, P., Booth, J., Flucker, S., Rouyer, A. (2016)
2016 Best Practices Guidelines for the EU Code of Conduct on Data Centre Energy
Efficiency. Version 7.1.2. Retrieved from: http://iet.jrc.ec.europa.eu/energyefficiency/ict-
codes-conduct/data-centres-energy-efficiency [Accessed: February 2017]
[43] Newcombe L, Acton M, Booth J, Flucker S, Rabbetts A, Strutt S, Rouyer A, Tozer R.
2013 Best Practices for the EU Code of Conduct on Data Centres. Version 4.0.5. 2013.
Retrieved from: http://iet.jrc.ec.europa.eu/energyefficiency/ict-codes-conduct/data-
centres-energy-efficiency [Accessed: February 2017]
[44] Depoorter, V., Oró, E., Salom, J. (2015) The location as an energy efficiency and
renewable energy supply measure for data centres in Europe. Applied Energy, 140, pp.
338-349.
[45] McFarlane R. Look to the Kyoto wheel to maximize free data center cooling. 2010,
July. Retrieved from: http://searchdatacenter.techtarget.com/tip/Look-to-the-Kyoto-
wheel-to-maximize-free-data-center-cooling [Accessed: February 2017]
[46] Depoorter V, Oró E, Salom J. The location as an energy efficiency and renewable
energy supply measure for data centres in Europe. Applied Energy 2015; 140: 338-349.
[47] European Commission (2016). EU Code of Conduct on Data Centre Energy
Efficiency. Introductory Guide for all applicants. Version 3.1.2. Retrieved from:
http://iet.jrc.ec.europa.eu/energyefficiency/ict-codes-conduct/data-centres-energy-
efficiency
35
[48] Directive 2010/31/EU of the European parliament and of the council of 19 May 2010
on the energy performance of buildings. Off. J. Eur. Union 2010, 3, 124–146.
[49] CLOUDSCENE website: https://cloudscene.com/search/data-
centers?searchTerm=&pDc=1&pSp=1&pFb=1&pMar=1&sDc=providers&sSp=pops&sFb=
markets&sMar=facilities ]Accessed on 9 November 2017]
36
List of abbreviations and definitions
BP Best Practice
CAGR Compound Annual Growth Rate
CCA Climate Change Agreement
CoC Code of Conduct
CRAC Control Room Air Conditioning (unit)
CRAH Control Room Air Handler (unit)
DA Direct Air
DC Data Centre
DW Direct Water
IA Indirect Air
IW Indirect Water
RH Relative Humidity
HPCC High-Performance Computing Cluster
HVAC Heating, Ventilation and Air-Conditioning
ICT Information and Communication Technology
IT Information Technology
PUE Power Utilisation Effectiveness
37
List of figures
Figure 1. Energy consumption of servers and data centres in Germany from 2010 to
2015 and forecast to 2025 ...................................................................................... 7
Figure 2. Global Hyperscale Data Centre Operators .................................................. 9
Figure 3. PUE values ranges. ................................................................................15
Figure 4. Average PUE by reporting year ................................................................17
Figure 5. Number of data centres by size classification. ............................................17
Figure 6. Average PUE values for different size classifications. ..................................18
Figure 7. Number of data centres per year of construction........................................18
Figure 8. Average PUE values for different construction years. ..................................19
Figure 9. Number of data centres per IT Rated Load (kW). .......................................19
Figure 10. Average PUE values for different IT Rated Load classifications (kW). ..........20
Figure 11. Percentage of each of the seven different sectors. ...................................21
Figure 12 . Distribution of the seven types of sectors participating in the CoC. ............21
Figure 13. Average PUE values by geographical zone ...............................................23
Figure 14. Geographical Distribution of the participation data centre in the EU CoC. ....23
Figure 15. Distribution of the sites using a combination of free cooling technologies. ...25
Figure 16. PUE range by type of free cooling technology ..........................................26
Figure 17. Frequency of the best practices implementations. ....................................30
38
List of tables
Table 1. Data centres in Western Europe ................................................................. 5
Table 2. Number of data centres by size category and growth rates ............................ 7
Table 3. Average data of reporting facilities ............................................................15
Table 4. Average PUE by reporting year .................................................................16
Table 5. Geographical zoning with temperature and relative humidity average data .....22
Table 6. PUE average data for each free cooling technique .......................................25
Table 7. Minimum expected and optional Best practices categories. ...........................27
Table 8. Ten top implemented Best Practices ..........................................................28
Table 9. Ten less implemented Best Practices .........................................................28
Table 10. Best Practices that have never been implemented ....................................29
GETTING IN TOUCH WITH THE EU
In person
All over the European Union there are hundreds of Europe Direct information centres. You can find the
address of the centre nearest you at: http://europea.eu/contact
On the phone or by email
Europe Direct is a service that answers your questions about the European Union. You can contact this
service:
- by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),
- at the following standard number: +32 22999696, or
- by electronic mail via: http://europa.eu/contact
FINDING INFORMATION ABOUT THE EU
Online
Information about the European Union in all the official languages of the EU is available on the Europa
website at: http://europa.eu
EU publications
You can download or order free and priced EU publications from EU Bookshop at:
http://bookshop.europa.eu. Multiple copies of free publications may be obtained by contacting Europe
Direct or your local information centre (see http://europa.eu/contact).
KJ
-NA-28874-EN-N
doi:10.2760/358256
ISBN 978-92-79-76445-5
