The datacenters consume 1% of the world’s electricity, and contribute to 0.3% of CO emissions2 in the world. With the explosion in the amount of data generated around the world and the cloud, data centers could consume up to 3% of the world’s electricity by 2030. Yet there is a simple way to reduce the footprint carbone of this equipment, highlights a team of researchers from the Technical University of Berlin: all you have to do is program the most intensive calculations in energy at the time when the production ofrenewable energies is the most available, or sometimes even in excess. Between 40 and 60% of the tasks are thus called “recurring”, that is to say that they are repeated every day and can be programmed indifferently at any time or day of the week.
The researchers simulated for four countries (France, Germany, Great Britain and California) the effects of such planning according to the energy mix from each region. In France, for example, the intensity is 22% lower on weekends compared to the week. In Germany, the best time slots are at 7 a.m. and 6 a.m. in the evening. By trusting programming calculations to artificial intelligence, it would thus be possible to save between 24.2 and 36.6% ofemissions in France, Germany and Great Britain and up to 131.2% in California, according to estimates by researchers published on the site arXiv.
Overall, we must all the same underline the enormous efforts achieved by data center operators in terms of energy efficiency. Despite a 60% increase in demand, their electricity consumption should remain stable by 2022, underlines the international energy agency.
.
The datacenters consume 1% of the world’s electricity, and contribute to 0.3% of CO emissions2 in the world. With the explosion in the amount of data generated around the world and the cloud, data centers could consume up to 3% of the world’s electricity by 2030. Yet there is a simple way to reduce the footprint carbone of this equipment, highlights a team of researchers from the Technical University of Berlin: all you have to do is program the most intensive calculations in energy at the time when the production ofrenewable energies is the most available, or sometimes even in excess. Between 40 and 60% of the tasks are thus called “recurring”, that is to say that they are repeated every day and can be programmed indifferently at any time or day of the week.
The researchers simulated for four countries (France, Germany, Great Britain and California) the effects of such planning according to the energy mix from each region. In France, for example, the intensity is 22% lower on weekends compared to the week. In Germany, the best time slots are at 7 a.m. and 6 a.m. in the evening. By trusting programming calculations to artificial intelligence, it would thus be possible to save between 24.2 and 36.6% ofemissions in France, Germany and Great Britain and up to 131.2% in California, according to estimates by researchers published on the site arXiv.
Overall, we must all the same underline the enormous efforts achieved by data center operators in terms of energy efficiency. Despite a 60% increase in demand, their electricity consumption should remain stable by 2022, underlines the international energy agency.
.