Issue 3, April 2009
FOCUS RESEARCH
EER: A WORK IN PROGRESS A genuine tool in the fi ght to drive greater effi ciency? The uptake of EER in its various forms is profi led here.
2008 saw the use of energy efficiency ratings (EER) as one of the fastest-growing areas of data center investment activity.
EER emerged in a manner similar to virtualisation to become a data center staple – an activity which, if not actually considered, would need some good reason for not doing so.
Possibly the strong growth of EER in 2008 represents an example of the improved packaging of techniques and methods that were not necessarily new but whose time came with the increasing emphasis on efficiency. The process
of measurement
standardisation, the confidence gained from obtaining data that can be compared broadly to that obtained from other data centers, and the links that the evolution of any efficiency or performance- oriented quantification system will develop into best practice – has meant that the development of improved and agreed metrics is one of the three industry trends of 2008 most welcomed by both end-users and vendors.
Data obtained from each region indicates that the process of adoption has been relatively swift. Around 60% of organisations in the US, more than 40% of those in Western Europe and over 50% of those surveyed in emerging markets have engaged with EER in some form, and a further 20% in each region intend to.
Regional differences do emerge – PUE is narrowly the method most adopted in the US, PUE and other methods run neck and neck in Western Europe, and DCIE has been the most favoured in emerging markets. Note that the percentages on Figure 1 refer to organisations, not to individual facilities.
There are also some differences across regions between the broad
Figure 4: The confi dence engendered by EER has been rising steadily Figure 1: Swift adoption patterns
profile of facilities that have adopted various EER methods (Figure 2). Those which have not yet adopted any tend to be far larger, slightly older and tend to run at lower maximum power demand.
Those adopting PUE are average size on most characteristics, while those that have adopted DCIE and other methods of EER are smaller in size, younger and with a higher maximum power demand (Figure 2).
Figure 2: Higher densities lead to EER
Datacenter Research Group reporting includes analyses by more specific facility and portfolio profile information provided on more than 400 facilities. One example of this is shown on Figure 3, where the averaged PUE ratings have been plotted against average and maximum power consumption. Note that facilities that have been rated at greater than a PUE of greater than 2.50, or within the category of ‘not recognised’ as suggested by the Green Grid, have been excluded. The chart indicates a generally downward PUE rating achieved by facilities that have an average power consumption of greater than 5kW/rack and a maximum power consumption of greater than 8kW/rack (Figure 3).
Figure 3: Higher densities also lead to greater discipline
The confidence engendered by EER is reflected in the higher proportions of owners and operators who consider they have the tools to collect the information they need to drive efficiency. In 2007, fewer than 25% believed this. This rose to 40% in 2008 and to almost 50% among those which conducted some form of EER (Figure 4).
The fact that there are still high proportions who do not consider they have the tools to collect the information they need indicates that this is still very much a work in progress.
www.datacenterdynamics.com
25
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56