AI networking and full-stack optimization for AMD clusters
AMD and DriveNets are helping cluster builders deploy high-performance AMD-based AI environments faster and with lower risk. We are working closely together to provide a complete AI infrastructure solution that includes AMD Instinct GPU systems and DriveNets networking domains, spanning scale-up, scale-out, scale-across and frontend & storage networking. The joint solution provides a fully integrated software stack that is optimized for performance and functionality, delivering a competitive alternative to a full NVIDIA AI solution.
System-Level Optimization for AMD Instinct GPU Clusters
- Higher cluster performance: Joint AMD–DriveNets optimization across RCCL, networking, and the full cluster stack improves end-to-end workload performance.
- Faster, lower-risk deployment: A validated reference architecture, automated provisioning, and built-in testing simplify cluster bring-up and enable repeatable deployments.
- Proven results at scale:Real-workload testing demonstrates measurable performance gains against Spectrum-X, InfiniBand, and baseline AMD cluster configurations.
- Continued performance innovation:Ongoing joint R&D collaboration including Helios rack-scale infrastructure, liquid-cooled switching, and automated, adaptive, and agentic optimization to boost performance for future AI models.
AMD Instinct Fabric Reference Architecture
AMD and DriveNets released a validated reference architecture document for clusters built with AMD Instinct MI355X GPUs, AMD Pollara NICs, and DriveNets scale-out and frontend solution.
The reference architecture document provides a comprehensive end-to-end blueprint for building a high-performance, scalable AI GPU cluster, and a repeatable deployment model that reduces integration and configuration risk.
Reference Architecture
A validated reference architecture document for clusters built with AMD Instinct MI355X GPUs, AMD Pollara NICs, and DriveNets AI Fabric solution
