DriveNets AMD System Reference Architecture

This Reference Architecture (RA) document provides an end-to-end blueprint for building a high-performance AI GPU cluster based on AMD compute infrastructure and DriveNets AI Fabric networking solution.

The document includes technical guidance, best practices, and a full-stack view for maximizing AI cluster performance and utilization, to achieve the following:

  • 15% faster Time to First Token (TTFT) at the single-node level
  • 12–16% faster TTFT at scale
  • Up to 5% higher multi-node throughput
  • Stronger performance as concurrency increases

 

Download the White Paper

TopicsAI Networking

Events

AI Infra Summit

Read more

Blog

SC25 Recap – AI and More

Heading back from St. Louis late last week, I had some time to reflect on the very busy week of the SC25 event. When it ...

Read more

CloudNets Video

CloudNets S4 E6: Data Center Interconnect (DCI)

DCI is going through something and this something is called AI. The three changes that AI made to DCI are capacity, perf ...

Read more

Continue reading

Faster LLM Inference on AMD Requires Rethinking All-Reduce

Blog

Faster LLM Inference on AMD Requires Rethinking All-Reduce

AMD Instinct GPUs offer real hardware advantages for large AI clusters, including higher HBM3 memory capacity and compet ...

Read more
How AMD Instinct Shines in Real-World LLM Inference

Blog

How AMD Instinct Shines in Real-World LLM Inference

AI workloads have moved beyond experimentation into deep production environments. Today, performance is no longer about ...

Read more

Blog

Optimizing AMD Instinct AI Clusters with DriveNets’ Lossless Ethernet Fabric

The DriveNets Network Cloud-AI networking fabric solution delivers the highest performance AI connectivity for any GPU, ...

Read more