- Pro
HPE and AMD bring Helios to commercial AI systems
Comments (0) ()When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.
(Image credit: AMD)
- HPE will ship 72-GPU racks with next-generation AMD Instinct accelerators globally
- Venice CPUs paired with GPUs target exascale-level AI performance per rack
- Helios relies on liquid cooling and double-wide chassis for thermal management
HPE has announced plans to integrate AMD’s Helios rack-scale AI architecture into its product lineup starting in 2026.
The collaboration gives Helios its first major OEM partner and positions HPE to ship full 72-GPU AI racks built around AMD’s next-generation Instinct MI455X accelerators.
These racks will pair with EPYC Venice CPUs and use an Ethernet-based scale-up fabric developed with Broadcom.
You may like-
A secret AMD Christmas tree? Nah, it's just a close up of a very bright $3 million 72-GPU MI450 Helios rack monster at OCP Summit
-
AMD embraces Meta-backed Open Rack Wide form factor with new MI450-powered Helios racks - Oracle is the first big client with a 50,000 GPU commitment, with more to come
-
Nvidia might dominate the industry, but Oracle is still betting on AMD chips for its superclusters
Rack layout and performance targets
The move creates a clear commercial route for Helios and puts the architecture in direct competition with Nvidia’s rack-scale platforms already in service.
The Helios reference design relies on Meta’s Open Rack Wide standard.
It uses a double-wide liquid-cooled chassis to house the MI450-series GPUs, Venice CPUs, and Pensando networking hardware.
AMD targets up to 2.9 exaFLOPS of FP4 compute per rack with the MI455X generation, along with 31TB of HBM4 memory.
Are you a pro? Subscribe to our newsletterContact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsorsBy submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.The system presents every GPU as part of a single pod, which allows workloads to span all accelerators without local bottlenecks.
A purpose-built HPE Juniper switch supporting Ultra Accelerator Link over Ethernet forms the high-bandwidth GPU interconnect.
It offers an alternative to Nvidia’s NVLink-centric approach.
You may like-
A secret AMD Christmas tree? Nah, it's just a close up of a very bright $3 million 72-GPU MI450 Helios rack monster at OCP Summit
-
AMD embraces Meta-backed Open Rack Wide form factor with new MI450-powered Helios racks - Oracle is the first big client with a 50,000 GPU commitment, with more to come
-
Nvidia might dominate the industry, but Oracle is still betting on AMD chips for its superclusters
The High-Performance Computing Center Stuttgart has selected HPE’s Cray GX5000 platform for its next flagship system, named Herder.
Herder will use MI430X GPUs and Venice CPUs across direct liquid-cooled blades and will replace the current Hunter system in 2027.
HPE stated that the GX5000 racks’ waste heat will warm campus buildings, which shows environmental considerations in addition to performance goals.
AMD and HPE plan to make Helios-based systems globally available next year, expanding access to rack-scale AI hardware for research institutions and enterprises.
Helios uses an Ethernet fabric to connect GPUs and CPUs, which contrasts with Nvidia’s NVLink approach.
The use of Ultra Accelerator Link over Ethernet and Ultra Ethernet Consortium-aligned hardware supports scale-out designs within an open standards framework.
Although this approach allows theoretically comparable GPU counts to other high-end AI racks, performance under sustained multi-node workloads remains untested.
However, reliance on a single Ethernet layer could introduce latency or bandwidth constraints in real applications.
That said, these specifications do not predict real-world performance, which will depend on effective cooling, network traffic handling, and software optimization.
Via Tom's Hardware
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.
TOPICS AMD
Efosa UdinmwenFreelance JournalistEfosa has been writing about technology for over 7 years, initially driven by curiosity but now fueled by a strong passion for the field. He holds both a Master's and a PhD in sciences, which provided him with a solid foundation in analytical thinking.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
Logout Read more
A secret AMD Christmas tree? Nah, it's just a close up of a very bright $3 million 72-GPU MI450 Helios rack monster at OCP Summit
AMD embraces Meta-backed Open Rack Wide form factor with new MI450-powered Helios racks - Oracle is the first big client with a 50,000 GPU commitment, with more to come
Nvidia might dominate the industry, but Oracle is still betting on AMD chips for its superclusters
AMD Megapod set to face Nvidia's Superpod with a 256-GPU rack full of Instinct MI500 chips
Huawei Atlas 950 SuperPoD vs Nvidia DGX SuperPOD vs AMD Instinct Mega POD: How do they compare?
Can't afford Nvidia's expensive AI accelerators? Then consider this 10.8Kw server cluster with 32 Intel GPUs and 768GB VRAM
Latest in Pro
AWS Graviton5 is its most powerful and efficient CPU to date - and could mean big changes for your key cloud workloads
Start 2026 strong - you can save up to 50% on Wix’s top apps right now
Customer data stolen in Freedom Mobile account management platform hack
CEOs are warning AI adoption and spending should be more strategic
Apple Final Cut Pro (2025) review
This DDoS group just smashed the previous record with a 29.7 Tbps attack
Latest in News
Roblox, FaceTime become the last targets of Russia's censorship
Marvel Rivals now has a gacha mini-game featuring a limited-time Psylocke bundle – here's how it works
YouTube to lock out under-16s in Australia as controversial social media ban looms
Forget Spotify Wrapped, get back into CDs with FiiO’s gorgeous new portable player
Sony announces partnership with Bad Robot Games to produce and publish a new four-player, co-op shooter from Left 4 Dead director
Can you guess which pro Canon camera took the best landscape photo of 2025?
LATEST ARTICLES- 1Scientists engineer erbium molecular qubits that plug directly into existing fiber networks, promising telecom-ready quantum systems
- 2CEOs are warning AI adoption and spending should be more strategic
- 3I went hands-on with Hengbot Sirius, an AI robot that left me wanting more
- 4AWS Graviton5 is its most powerful and efficient CPU to date - and could mean big changes for your key cloud workloads
- 5Roblox, FaceTime become the last targets of Russia's censorship
