There have been several developments this year around the greening of data centres and the metrics used to measure them, so a brief update:
• After a wide review by all parties involved, the EU published its ‘2010 Best Practices for the EU Code of Conduct on Data Centres’. It provides the full list of identified best practices for data centre operators referred to in the Code of Conduct and is intended as an education and reference document.
It differentiates between aspects that can be applied to existing data centres and equipment and those that are expected with new IT equipment or software or during a retrofit of the facility. The document also identifies a subset of the best practices as being the expected minimum level of energy saving activity for Participant status.
• In June, the EPA announced the availability of the Energy Star label for stand-alone data centres and buildings that house large data centres, based on the Power Usage Effectiveness (PUE). To earn the label, data centres must be in the top 25% in energy efficiency according to EPA’s energy performance scale.
Last week storage and data management company NetApp announced that its Research Triangle Park (RTP) data centre, opened in 2009, had earned the award, the first to do so. With a score of 75% or higher needed, the RTP data centre scored an impressive 99%.
The data centre design has reduced CO2 emissions for NetApp by approximately 95,000 tons per year. As a design blueprint it has apparently been visited by representatives from more than 500 organisations.
• Back in January a group of organisations in the US got together to agree on data centre energy efficiency measurements, metrics and reporting conventions. Among others they included The Green Grid, US EPA’s Energy Star programme, US Green Building Council and the Uptime Institute.
The first fruits of their labour were released last week in a report called ‘Recommendations for Measuring and Reporting Overall Data Center Efficiency’. This is Version 1, which provides recommendations on measuring and publishing values for PUE at dedicated data centre facilities. Version 2 will cover data centres that are part of larger mixed-use facilities.
The problem they’re addressing is that while PUE is widely used as a measure of data centre efficiency, the metrics are not always applied clearly and consistently. This is an issue when data centres are compared with each other or over time. (I’ve reported on the issues in the past, including the complaints that comparisons were not being made like-for-like). This document sets out what should be measured, how and when, broken down into four different PUE categories.
The methodology also provides weighting factors for different types of energy source, but it doesn’t take renewable energy use into account (since this is a measure of efficiency). Nor does it take into account whether waste heat is re-used, for instance to heat other facilities, although ‘there are on-going industry efforts to define a metric that could be used to account for this beneficial use’.
It’s good to see all this effort going into making data centres more efficient and measuring the results – there is much to be gained. But it is only a small part of the green ICT landscape, when you consider that IT can save a lot more energy outside the data centre than in. It would be good if similar attention were paid to the savings that IT can make elsewhere. It is happening - Smart 2020 and other reports have given macro estimates and some IT companies are measuring their contribution to emissions reductions in the economy and at the individual service level. It’s much harder to use clear and comprehensive metrics on these savings but it needs to be done if ICT is not to be relegated solely to the role of polluter (as some legislative initiatives are in danger of doing).