Facebook open-sources hardware for AI model training and inference

Serving 2.7 billion people each month across a family of apps and service isn't easy -- just ask Facebook. In recent years, the Menlo Park tech giant has migrated away from general-purpose hardware in favor of specialized accelerators that promise performance, power, and efficiency boosts across its datacenters, particularly in the area of AI. And toward that end, it today announced a "next-generation" hardware platform for AI model training -- Zion -- along with custom application-specific integrated circuits (ASICs) optimized for AI inference -- Kings Canyon -- and video transcoding -- Mount Shasta. Facebook says the trio of platforms -- which it's donating to the Open Compute Project, an organization that shares designs of data center products among its members -- will dramatically accelerate AI training and inference. "AI is used across a range of services to help people in their daily interactions and provide them with unique, personalized experiences," Facebook engineers Kevin Lee, Vijay Rao, and William Christie Arnold wrote in a blog post.