A RoCE network for distributed AI training at scale
Engineering at Meta
AUGUST 5, 2024
Our paper, “ RDMA over Ethernet for Distributed AI Training at Meta Scale ,” provides the details on how we design, implement, and operate one of the world’s largest AI networks at scale. We opted for RDMA Over Converged Ethernet version 2 (RoCEv2) as the inter-node communication transport for the majority of our AI capacity.
132 
Let's personalize your content