Most read articles in What's Next:

Read all articles

Reducing emissions and facilitating renewable energy in data centres
By: Fabien Petitcolas, Director for Innovation, Europe
24 June 2013

Send EMail

You may recall that a year and a half ago we talked about research on reducing datacentre emissions. The researchers have been pushing the work further: they are using electricity market analytics with the aim to reduce CO2 emissions and energy costs while increasing the contribution of renewable energy to datacentres.

Early April, Conor Kelly of Microsoft Ireland presented the research to the IEEE Green Technologies Conference 2013 in Denver. The work was done in collaboration with Antonio Ruzzelli and Eleni Mangina of University College Dublin.

The new research outlines algorithms to price datacentre resources based on electricity market conditions in order to reduce costs, reduce resulting emissions and to increase the share of renewable energy in the consumption of datacentres. The paper also provides new insight in electricity markets: the fluctuation of price and also the rapidly changing share of renewable electricity in the mix. Both are illustrated below.



Figure 1: Electricity prices (€/MWh) for 50 days on the Irish electricity market between February and June 2012.


Figure 2: Share (%) of renewable electricity on the Irish grid between January and May, 2012.

The algorithms rates different criteria (prices, emissions, current renewable percentage) on the electricity grid, to help decide when a good time to perform energy consuming computation is (see figure 3). The results could then be used to generate price incentives to move computation to these “good” times, based on the preferences of the customer or datacentre operator. Performing computation at these “good” times could greatly increase the mix of renewable energy, achieve up to a 99% reduction in CO2 emissions (see figure 4) or decrease the energy costs for the same piece of computation depending on what the customer or operator prioritises are.


Figure 2: A sample period for the ratings of different criteria on the Irish electricity market. For instance during the period around 8AM, the percentage of renewables in the mix is high and the CO¬2 emissions are low (high rating corresponds to low emissions) – this is therefore a good time to use energy in order to minimise emissions.

Energy consumption has become a major issue in terms of cost and emissions for datacentre operators. Utilising intelligence and awareness of these market and grid fluctuations allows much more favourable outcomes in terms of emissions and renewable energy contribution. Furthermore, major energy consumers like data centres can actually facilitate the addition of renewable energy sources to worldwide electricity grids by intelligently adapting their consumption to the constraints of the sources – delaying movable computation to times of renewable generation, or moving computation loads to different geographic locations. Essentially following green energy around the globe.

The penetration of renewable energy will continue to increase in electricity markets worldwide, driven by concerns about climate change and energy security. As it does, data centre operators could use the methods outlined in this research to add to the momentum of renewable energy adoption and the drive towards carbon free electricity sources.



Figure 4: This is a snapshot of the level of renewable energy versus the emissions of the grid at that time, for every 15 minutes on the Irish electricity grid in 2012. It shows that higher renewable generation corresponds to lower emissions on the grid.



Related content: