GREEN NETWORKING
The Exploitation of Power Usage Effectiveness My Network is Greener Than Yours By Jan Wiersma, Technical Manager, EvoSwitch
Data centres are deploying a huge range of new energy-effi ciency measures, from innovative cooling units to the latest climate-controlling techniques. With organisations such as The Green Grid seeking to create universal metrics for energy effi ciency, industry professionals are having to not only claim how ‘green’ their data centre is, but to actually prove it. A by-product of this movement
The way today’s data centre landscape is developing is, quite
unavoidably, dictated by the need to be energy- effi cient. Jan Wiersma looks at the bigger picture for green networking.
is the concept of Power Usage Effectiveness (PUE). Developed four years ago to measure how effi ciently a data centre uses its power, PUE is calculated by dividing the total amount of power entering the data centre by the amount delivered to IT equipment. The resulting fi gure can be used as an indicator of how energy effi cient, and therefore how environmentally friendly, that data centre is. As a consequence, the data centre industry is becoming increasingly populated by the ‘my PUE is better than yours’ debate and, unfortunately, a general misuse of the metric. This misuse is not only generated, but also exacerbated by data centre professionals themselves.
PUE as a USP For those in the data centre colocation or wholesale business in particular, marketing and sales representatives are constantly trying to look for their next unique selling point. PUE
represents a simple, single fi gure which provides exactly that. With the on-going trend towards energy effi cient data centres, it’s highly desirable to have a single fi gure telling the customer that your data centre is not only more effi cient than the competition, but that this is accompanied by a lower price! PUE is happily adapted to promote
green IT with those claiming to have built ‘the greenest data centre yet.’ With data centres becoming increasingly more diffi cult to sell as a ‘commodity product,’ then, it’s easy to see the appeal of PUE to the marketing and sales division of the industry. Indeed, from the enterprise & cloud data centre perspective, the angle differs only slightly. Larger data centre owners in
particular are pushed from a marketing perspective thanks to NGOs like Greenpeace. These organisations demand full disclosure of data centre energy use and sustainability on behalf of the world’s population, further encouraging and indeed demanding the use of PUE. Many state and local government
organisations are looking at data centres from an energy-use perspective. This focus is thanks to reports from research and governmental institutes such as the Environmental Protection Agency (EPA) and organisations like Greenpeace. While governments can freely use the ‘carrot or stick’ (subsidies or penalties) approach, this nonetheless requires some type of regulation and, more specifi cally, a form of measurement. Some of these governmental organisations are also using PUE for their own colocation tenders, and they are not the only ones. More and more requests for
proposals on the data centre market state a specifi c PUE number or target with additional credits for the lowest number. It’s hard to blame the customer; with the data centre a complex topic for anyone outside of the industry, it’s easy to see the appeal of a measured, standardised fi gure.
PUE was developed as a way of benchmarking a single data centre internally. 36 NETCOMMS europe Volume II Issue 3 2012
In Context The main problem for this is that PUE is not ready to be audited. While
www.netcommseurope.com
general guidelines for PUE calculation are available, these do not provide full details and leave opportunities for the fi gure to be manipulated. They also fail to address the holistic data centre approach as well. PUE was developed by The Green Grid as a way of benchmarking a single data centre internally, and measuring its progress and development in terms of energy effi ciency. For this reason, it is not a reliable method of comparing different data centres. However, the customer cannot solely be blamed. Those within the industry are just as happy to use the fi gure to suit their purposes, and do so freely. These days PUE fi gures are everywhere; on the equipment specifi cation sheets, on the vendor website and in the sales meeting. In the end, though, it can only be concluded that as a stand-alone fi gure, PUE is meaningless with out context. Of course, this is not to say that the
metric is worthless. If used correctly, in the way indicated by the EU Data Centre Code of Conduct for instance, the fi gure can be used as a benchmarking tool in order to show progress. If used to compare like against like,
PUE cannot be faulted. It is only when misused and exploited in the ways detailed above that it becomes a liability. The industry as a whole, data centre owners, consumers and industry groups, must be held accountable for the misuse of PUE. If we ignore the guidelines, it’s unfortunately likely that this misuse will continue as it’s simply a way of trying to put a number on what can be a very complex stack of functions that make up the data centre.
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