A person with a laptop reviews technical diagrams at a desk. Photo by ThisisEngineering on Unsplash. Source: https://unsplash.com/photos/woman-in-white-long-sleeve-shirt-using-black-laptop-computer-1dwyU46p7eE (Unsplash License).

HPE booked its first order for the AMD Helios AI rack, and the buyer is Vultr. The headline number is $1.2 billion. The line that matters is the network inside the rack.

  • Each rack holds 72 AMD Instinct MI455X GPUs, AMD EPYC Venice CPUs, AMD Pensando Vulcano NICs and the ROCm software stack.
  • Six HPE Juniper QFX5252 switch trays connect all 72 GPUs over Ethernet, using the UALink over Ethernet standard.
  • That puts HPE switches inside the scale-up domain, the GPU-to-GPU fabric Nvidia holds with NVLink.
  • An open Ethernet fabric lets buyers split compute, fabric and cooling into separate line items.
  • HPE raised its data center networking outlook the same day and lifted its Juniper cost-savings target to $800 million.

Read the full analysis. Vultr’s $1.2 billion bet is Ethernet inside the GPU rack.

By Ivan Tarin

Ivan Tarin is a Principal Product Marketing Manager at SUSE, where he owns go-to-market strategy and positioning for a seven-product cloud-native portfolio spanning Kubernetes, virtualization, storage, security, and observability. A former full-stack developer who shipped production code for enterprise and public-sector clients including U.S. national laboratories, Ivan translates complex infrastructure and AI technology into messaging that lands with developers, platform teams, and enterprise buyers. He has presented at KubeCon, SUSECON, and AWS Developer Week, and is currently pursuing an MS in Artificial Intelligence at the University of Colorado Boulder.

Leave a Reply

Your email address will not be published. Required fields are marked *

Get the next one before it is old news

Independent analysis of cloud-native infrastructure, Kubernetes and data center economics. No vendor spin.