New Google carbon intelligent data center gets the most from sustainable energyApril 24, 2020
The tech giant announced it was rolling out this new renewable energy focused computing platform.
Google has announced the launch of a carbon intelligent computing platform it has implemented for its data centers.
The purpose of the new technology is to change the timing of the performance of various tasks.
By altering the timing of some kinds of computing tasks, the data center facilities are able to maximize their use of renewable energy from sources such as solar and wind. Google said that by moving its non-urgent computing task timing – such as developing new Google Photos filters, processing YouTube videos and adding new Google Translate words – it can substantially decrease the size of tis carbon footprint.
The new platform was rolled out on Earth Day, April 22, to mark the company’s support for protecting the environment. This is only one of a number of moves Google has taken to help improve its sustainability. The company already reached a state of carbon neutrality in 2007. Last year marked Google’s third consecutive year in matching its energy usage with 100 percent renewable energy purchases.
The new carbon intelligent platform is meant to help its data centers to reach the same state.
“Now, we’re working toward 24×7 carbon-free energy everywhere we have data centers, which deliver our products to billions of people around the world,” said an official Google blog post. “To achieve 24×7 carbon-free energy, our data centers need to work more closely with carbon-free energy sources like solar and wind.”
The platform works by producing two types of daily forecast. The first provides a prediction regarding the average hourly carbon intensity of the local grid and the way it will fluctuate throughout the full 24 hours of the day. The second is a prediction of the number of power resources that will be required for the data centers to be able to compete their computing tasks during that same 24 hour span.
The carbon intelligent platform’s predictions are then applied to the scheduling of non-essential computing tasks. This way, those tasks are automatically scheduled for times when there is a greater availability of low-carbon energy resources.