Build Large Scale AI Infra Network

Clos Network

Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameters

RDMA Scale Farbic

AI Cluster Reference Design

NVIDIA DGX SuperPOD: Next Generation Scalable Infrastructure for AI Leadership

Fabric-Scheduled Ethernet as an Effective Backend Interconnect for Large AI Compute Clusters

AI Farbics Dell

HPC无损以太和AI Fabric网络技术白皮书

RDG for a Scalable, High-performance Kubernetes Cluster over NVIDIA Ethernet Fabric

Maintaining large-scale AI capacity at Meta

Scale-Out Ethernet fabric deployment for RoCE accelerated K8s cluster

Containers

RDMA Containers

Docker Macvlan Using RoCE