HPE and AMD reveal first Helios offering for cloud service providers and neoclouds

The rack-scale AI architecture also features hardware from HPE Juniper Networking hardware and Broadcom

The empty keynote stage at the Sphere ahead of the HPE Discover 2024 keynote. A giant HPE logo appears to hang above it
(Image credit: Future/Jane McCallion)

HPE has become the latest company to offer AMD Helios infrastructure to its clients, with features that it claims will provide a single “turnkey” tool capable of trillion parameter AI training.

AMD launched Helios at its Advancing AI conference in June, with CEO Lisa Su calling it a “game changer” and is aimed at both frontier model training and large-scale AI inference.

The HPE Helios offering – announced the day before the start of the company’s European Discover conference – features technology from Juniper Networks and is one of the early showings of how HPE’s latest acquisition is being integrated into its product range.

It will include a scale-up Ethernet switch and software, developed in collaboration with Broadcom, that’s been designed specifically for Helios’ architecture. According to HPE, it’s “the first switch to deliver optimized performance for AI workloads over standard Ethernet”.

The company added that the switch “leverages HPE’s AI-native automation and assurance capabilities to simplify network operations for faster deployment and overall cost savings”.

In addition to the Juniper switch, HPE’s Helios solution also features Broadcom’s Tomahawk 6 networking chip, 72 AMD Instinct MI455X GPUs per rack, AMD Pensando networking technology and AMD ROCm software.

“With the introduction of the new AMD ‘Helios’ and our purpose-built HPE scale-up networking solution, we are providing our cloud service provider customers with faster deployments, greater flexibility, and reduced risk in how they scale AI computing in their businesses,” said HPE CEO Antonio Neri.

Hock Tan, president and CEO of Broadcom, added: “Broadcom is proud to take part in this collaboration to advance open, Ethernet -based AI infrastructure for scale-up.”

“Together with HPE and AMD, we are enabling customers to build powerful AI data centers with standard Ethernet , maximizing choice and flexibility while delivering exceptional scale,” Tan said.

HPE’s Helios product will be available in 2026.

Jane McCallion
Managing Editor

Jane McCallion is Managing Editor of ITPro and ChannelPro, specializing in data centers, enterprise IT infrastructure, and cybersecurity. Before becoming Managing Editor, she held the role of Deputy Editor and, prior to that, Features Editor, managing a pool of freelance and internal writers, while continuing to specialize in enterprise IT infrastructure, and business strategy.

Prior to joining ITPro, Jane was a freelance business journalist writing as both Jane McCallion and Jane Bordenave for titles such as European CEO, World Finance, and Business Excellence Magazine.