Artificial Intelligence Cuts Google’s Data Center Energy Use
(3BL Media/Justmeans) – Data centers operate the equipment that processes the data consumed by Internet users. It takes large amounts of energy to keep their servers cool, so much so that some of the newer data centers are now being built in colder climates. According to some estimates, data centers are now responsible for two percent of global GHG emissions.
Google’s artificial intelligence division, DeepMind, has managed to reduce the company’s energy consumption at its data centers, using a machine-learning algorithm. DeepMind’s algorithm has created a real-time, adaptive system that cuts the cost of cooling by 40 percent and the overall energy consumption by 15 percent.
According to DeepMind co-founder Mustafa Suleyman, this will have a significant environmental impact. The system will be rolled out across all of Google's data centers by the end of the year. The company is already talking to non-Google partners about using the same algorithm because it could work equally well in large manufacturing facilities and even national energy grids.
DeepMind’s Research Engineer Rich Evans and Google’s Data Center Engineer Jim Gao said that DeepMind’s machine learning can be applied in any large scale energy-consuming environment. Given how sophisticated Google’s data centers already are, it is a phenomenal step forward.
Every improvement in data center efficiency reduces total emissions into the environment and with technology like DeepMind’s, machine learning can be used to consume less energy and help address climate change.
In its 2015 green energy report, Greenpeace had praised Apple, Google and Facebook for their commitments to the use of 100 percent renewable energy in their data centers.