As part of an ongoing effort to reduce the environmental impact of cloud computing, Google announced it has developed a platform that maximizes the use of alternative energy sources at its sprawling network of datacenters.

In a blog post, Google officials described the system as a “carbon-intelligent computing platform.” It is designed to rapidly analyze and prioritize the vast number of tasks that could potentially be performed and shift the less urgent ones to periods when energy from sources such as wind or solar is more likely to be available.

The latest announcement comes as tech giants are embracing renewable energy in an effort to reduce their environmental footprint while meeting the growing demand for online services. Companies like Apple, Microsoft, Amazon, Google, and Facebook have in recent years announced programs aimed at reaching carbon neutrality.

While there is no silver bullet, evidence suggests collective attempts to bring greater efficiency to datacenters are in fact slowing the rate of growth in energy use.

Where does your organization stand on the AI curve? Find out with this survey!

Authors of a recent article in the journal Science noted that datacenters consumed 1% of all energy in 2018. While that represents a 6% increase since 2010, the authors calculated that total datacenter computing power had increased 550% during the same time period.

Still, with that usage pattern, increasing the mix of renewables remains an essential goal in the fight against climate change.

Google said it designed its carbon platform to ensure there would be no impact on Search and Maps. Instead, it identified categories of “non-urgent” chores, such as “creating new filter features on Google Photos, YouTube video processing, or adding new words to Google Translate” that could be shifted to other times.

The platform does this by examining the patterns of energy use on the local power grid and matching that against an internal diagnostic of the power a datacenter needs throughout the day. The result is an hourly guideline that can realign the timing of various computing tasks. 

 

Google reported that its pilot study of the technology showed a clear increase in renewable energy use.

Going forward, the company hopes to extend this system by moving tasks from one datacenter to another where renewable energy sources might be more readily available.