With a comparable carbon footprint to the airline industry, IT has gained a reputation as a major culprit in carbon emissions, writes Brian Murray, principal consultant at Morse.
A lot of this is due to the energy-intensive nature of large datacentres: the average datacentre consumes the same amount of energy as 25,000 homes, and datacentres as a whole were responsible for more than 3% of the UK's total energy consumption as far back as 2002. Thanks to this, IT is the third largest contributing factor to carbon emissions in the UK.
From April the Carbon Reduction Commitment (CRC) Energy Efficiency Scheme, the UK's mandatory climate change and energy saving "cap and trade" scheme, will come into effect. It has the best of intentions: by ranking organisations based on their peak energy use, then charging those with the highest energy use and using that money to reward those with the lowest, the CRC should encourage IT departments to make their datacentres less power-hungry.
Yet despite the impact it will have on over 5,000 public and private sector organisations, the CRC is still not doing enough to encourage more energy-efficient behaviour.
By focusing simply on peak power use, and avoiding the more challenging task of measuring actual efficiency, the CRC is in fact undermining its very purpose. For example, a datacentre can be moved offshore to a country with no similar legislation, costing the UK valuable business. As far as the CRC is concerned, that organisation will show a huge reduction of peak power use. Yet it will not have reduced its actual energy consumption and the relocation itself will have added to the amount of energy used. As a result, overall global carbon emissions will remain unchanged at best.
Similarly, the CRC on its own provides no incentive for datacentres that are powered by renewable energy sources, and so have reduced emissions. And monitoring peak energy use means that a single surge in energy use in an otherwise highly efficient data centre will give it a far worse ranking than it deserves.
The CRC is not set in stone and will hopefully evolve further in the future to address these issues. In the meantime, what should IT departments do?
Offshoring may seem tempting, but is an expensive, and most likely short-lived, solution: many other countries are already considering similar schemes.
The most effective approach is to do what should be done regardless of the CRC's effectiveness: optimise IT operations and increase efficiency. This can include retiring older equipment, deploying more energy-efficient systems and using virtualisation, all off which can help optimise the datacentre and other IT resources. An increase in energy efficiency will impact an organisation's funds far more than any fine or reward from the CRC.
Irrespective of the CRC, energy use should be monitored. Rather than simply checking for peak power use, IT departments should keep a careful track of when energy is used and by what systems, allowing measurement of what efficiencies can be made and how effective they have been. Indeed, the IT department should try and gain ownership of its own energy bills, so that it can include power savings within its own business case and demonstrate commercial governance.
With the CRC likely to expand to cover more than just the 5,000 most energy-intensive organisations in the UK, IT departments should be preparing to follow the spirit of the law in advance, if for no other reason than because it is simply sound business sense.
This was first published in March 2010