In this whitepaper, “Deep Learning GPU Cluster,” our friends over at Lambda walk you through the Lambda Echelon multi-node cluster reference design: a node design, a rack design, and an entire cluster level architecture. This document is for technical decision-makers and engineers. You’ll learn about the Echelon’s compute, storage, networking,  power distribution, and thermal design. This is not a cluster administration handbook, this is a high level technical overview of one possible system architecture.