Ecowarriors save the planet in the data centre

29 June 2012

Robert Doe

Robert Doe

Founder and Consultant

We know there's a huge problem - but we have done so little since
2008

Just as climate change has moved progressively up the corporate and public agenda in recent years, so has the depth of the current once-in-a-lifetime recession.  As an increasing number of technologies offer ever quicker paybacks, partly as a result of rising utility prices, cash has got ever tighter and corporate payback thresholds moved typically from around 3 years to often inside 12 months. Sadly, at a time when most people now recognise the need to arrest the ever-rising CO2 levels, financial resources have been scarce, so several years have slipped by with a fraction of what could have been achieved in most western economies.

In the UK alone we are told that there needs to be over £100bn of investment by 2020 to modernise our power supply infrastructure, using renewables and cleaner versions of traditional power generation.  So capital rationing and wise investment are also crucial on a wider scale to make this affordable: if we do the big power supply projects with the longest payback periods first then greening the UK economy will bring an almost certain further drop in the standard of living.  But there is a way forwards with the right mindset.  Marks & Spencer, highly regarded for its marketing, has introduced 'Plan A' as it's sustainable (and profit-enhancing) agenda.  Thereason?  Simple: 'there is no Plan B - because there is no Planet B'.

Supply side and demand side solutions - one is economically much
better

The renewables agenda must be embraced, but let's not forget that any 'investor opportunity' in this area is supported today with substantial feed-in-tariffs and the like: the original domestic FIT (a production subsidy) was 41.3p/kWh when the wholesale value of that power for sale back to the grid was only 3p - that's nearly a 1,400% subsidy.  Someone else is paying - the public at large - to make this necessary investment attractive in the short-term to stimulate the industry - and effectively tax carbon to level the playing field vis a vis fossil fuels that pollute.  That's the supply end of the equation.

At the demand end, opportunities abound - without any need for subsidy: these projects add real value to those organisations implementing the changes - and to society at large.  And while shareholders may be happy to wait for great savings, the planet can't. 

Best practice by Sustainability champions

There are now 3 well-rehearsed golden rules, in order of priority (easily illustrated with lighting):

  1. Turn it off (demand)
  2. Turn it down (demand)
  3. Make it greener (supply)

Turn it off: a mixture of cultural awareness (turn the lights off when you leave the room) and sensors (that both turn them off when you leave).

Turn it down: you might be able to turn the lights down if it's a sunny day, maybe with the aid of sensors again.  Or, more importantly, you might be able to use a more efficient (light) source - an opportunity that applies across the whole energy use spectrum.

Make it greener: as we move towards a time when new buildings will have to be carbon neutral, those dimmed lights will need to be powered by photo-voltaics, wind power and biomass.

Capital rationing: identifying the best projects (it's the data centre,
stupid)

So, in terms of capital rationing and making the sustainable transition affordable, the very best opportunities are where good value technology can be sweated 24/7.  As regards energy efficiency, every IT function is sat on a goldmine, indeed the goldmine.  All data centres operate 24/7; many are lit 24/7; all are chilled 24/7.  Every watt saved using energy-efficient lighting saves around another 3 watts in chilling: chilling is not an efficient process.  And for energy-efficient lighting read LED.  So while some users install LED in ambient conditions for the direct energy savings alone, data centres can gain that benefit fourfold!  What is more, there are other power management technologies that can make the rest of the data centre operation more energy-efficient too - and with payback periods that should have the Finance Director salivating.

As a rule of thumb, every watt costs about £1 per annum when run 24/7: each watt that runs 24/7/365 = 8,760 watt hours = 8.76kWh.  If your power costs about 11.3p / kWh then that multiplied by 8.76kWh = £1.00.  And the 58W 5' fluorescent tube also needs a ballast to power it which adds about 20% - around 70W / £70 per annum for every 5' tube (plus maintenance / replacement every 10,000 hours or so).  A 5' LED tube will run at about 25W, saving 45W / £45.00; now multiply that £45 by 4 to save £180 per tube with chilling and it gets interesting.  Even if your power price is nearer 7.5p/kWh then you'll still save £120 in energy (+ maintenance).  This gives a power-only payback of around 6 months on a decent tube costing £60 installed that will last over 5 years!

Yes, it's the data centre - but there are loads of similar opportunities
as well

This same argument applies more broadly to a range of sectors.  Hotels light and air-condition countless corridors where the same economics apply; hospitals are similar.  Modern retail chains invariably operate air conditioning in stores for extended hours; food retailers also light their chiller and freezer cabinets, while upstream some FMCG (chilled food producers) need to light their shop floors but also keep the temperature down.

LED-specific advice: setting expectations and how and where to buy

When it comes to selecting LED solutions, the buyer should understand that the production process for LED chips is inherently variable by nature - and that the industry uses a binning system to separate the best from the rest.  Also, power efficiency (lumens per watt) tends to go hand in hand with life expectancy and light quality and reliability.  Manufacturers will then assemble the best components with the best chips and vice versa, so total underlying costs vary - and so does performance.  For domestic use, a cheap LED that will only be used a few hours a day is the best proposition to make savings right now; in a corporate environment, with long operating hours, high maintenance costs and aesthetics and corporate image to worry about, you want the best.

As regards the lighting supply chain, the oligarchy of traditional lighting - Philips, GE, Osram, Sylvania - risk being left behind by this disruptive technology; they seem to be slow to adapt, happy to sweat their existing investment and charge a hefty brand premium on their LED range.  What's more, the traditional lighting wholesalers and retailers that work with this oligarchy have been accustomed to applying a series of healthy mark-ups on a low value consumable: their value offering is the physical supply chain that replenishes constantly.  LED isn't a consumable but a multi-year life asset investment and should be treated as such - and put on the balance sheet - LEDs used in an office environment operating 40 hours a week could last 20 years.  And the way to find best value is to look beyond the traditional lighting supply chain to a totally new and shorter one with fewer if any middlemen that has been established for LED.

And how to justify energy-efficiency projects

IT professionals often face a challenge in justifying IT spend: you know the common sense is staring you in the face and an owner-manager would act accordingly, but how easy is it to prove incremental sales or headcount reductions in a corporate environment - savings are often marginal and difficult to isolate from other factors.  The financial justification for LED can be readily proven with the simple aid of a multi-meter and a light meter (+/- a power bill that drops).  As regards the power bill, for those of you that haven't already been immersed in sub-metering infrastructure projects, be aware that your power provider stores half-hourly data on your consumption for each site against an 'MPAN' number.  And they are obliged to share this HH data with you, so you will see the savings the day (the hour actually) that you install, even if you don't have your own metering.

A rallying cry to the ecowarrior in you: no delays, no more
procrastination. Just do it!

A rallying cry to the ecowarrior in you: start making the case to save the planet in your own data centre!  The savings from a project with a 6 month payback can then be reinvested in the next area of low-hanging fruit - and the march towards a greener, cleaner planet in your organisation is underway with full vigour.  Find the funding to invest £1,000 in a project enjoying a 6-month payback today and you can afford to invest twice as much, for free, 12 months later!  Very quickly, the best projects will achieve a head of steam that will help us all 'eat this sustainability elephant' one mouthful at a time - so long as sustainability is not just seen as another way to squeeze costs but starts to be managed as a profit centre in its own right and ploughs back the profits for greater future financial and environmental success.

It just needs someone to step forwards to start the process.  So what are you waiting for?

--------------------------------------------------------------------------------------------------------------------

Robert is a chartered accountant by training, but a very operationally minded one.  He has more than once instituted major changes to double operating profits and sees sustainability as the untapped goldmine in most organisations. He works with a number of cleantech companies and those offering global supply chain solutions that can also reduce CO2.

Disclaimer: The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of Bluefin Solutions Ltd.

View comments

Comments

Blog post currently doesn't have any comments.

About the author

Robert Doe

Founder and Consultant

Bluefin and SAP S/4HANA - welcome to the one horse race

We use cookies to provide you with the best browsing experience. By continuing to use this site you agree to our use of cookies.