Facebook revealed new compute and storage servers along with custom training server platform as part of a server “refresh.”

Compared to Big Sur, the new server enables Facebook to train machine learning models that are 30 percent larger, thanks to two factors: an increase in memory from 12 GB to 16 GB, as well as the availability of greater arithmetic throughput.

Facebook plans to make the server design open to the public in the near future. That is standard at the company, which participates and helped create the Open Compute Project for sharing and collaborating on data center software and hardware. So anyone — even server design specialists in competing companies — will soon be able to download the Big Basin schematics once they’re posted online.

facebook storage server
storage server

Facebook claims that the 1.86 billion users on its flagship social network currently watch 100 million hours of video per day. Not to be left behind, the photo-sharing app Instagram sees more than 95 million photos and videos uploaded daily. Messenger is piling the load on Facebook’s data centers via its video and voice chat services, which are now used 400 million people every month.

The new Bryce Canyon server also reflects the growing requirement for high-density storage of photos and video. The combination of the more processing power and increased memory capacity along with a 20 percent boost in hard drive density yielded a four-fold boost in computing power over the Honey Badger storage server introduced in 2015.

Meanwhile, a pair of new compute servers stress power efficiency in hyper-scale datacenters along with increased IO bandwidth. This new version of the Yosemite compute server supports ‘hot service,’ meaning that it continues to operate when pulled from a rack for the maintenance. Meanwhile, Tioga Pass incorporates dual-socket motherboards and increased bandwidth to the GPUs, flash memory and network cards.