It is HPE’s first order for the new AMD Helios system, which features purpose-built HPE Networking scale-up switching and software. HPE and AMD will support Vultr’s effort to meet service provider and enterprise demand for model training and inferencing workloads in their AI data centers.

The new order builds on nearly three years of collaboration between Vultr and Juniper Networks that continues to grow now that Juniper Networks is a part of HPE.

“Demand for high-performance AI infrastructure continues to outpace available capacity, and our work with AMD and HPE enables us to bring Helios rack-scale compute and scale-up networking online for customers faster. Together, we’re delivering the memory capacity, interconnect performance, and energy efficiency customers need to run the largest and most demanding AI training and inference workloads,” said J.J. Kardwell, CEO of Vultr.

The AMD Helios AI Rack by HPE is part of the HPE AI Data Center Solutions portfolio, a deeply integrated, production-ready infrastructure stack with AMD that is designed using open standards and optimized for the demanding economics and scale of the AI lifecycle.

Scale AI with an open architecture

AMD designed Helios as a high-density system for trillion-parameter model training and high-volume inference.

Each rack integrates 72 AMD Instinct™ MI455X GPUs with AMD EPYC™ “Venice” CPUs, AMD Pensando™ Vulcano AI NICs networking, and AMD ROCm™ software. The unified design is optimized for power, cooling, and serviceability, while supporting open, interoperable rack-scale fabrics such as UALink™ over Ethernet (UALoE).
HPE Networking purpose-built for Helios, including six HPE Juniper Networking QFX5252 scale-up Ethernet switch trays per rack, connect all 72 GPUs with high-bandwidth, low-latency, standards-based Ethernet.
HPE Services reduces project and operational risk through expertise in end-to-end deployment and liquid-cooling for large AI clusters.
The AMD Helios by HPE order with Vultr delivers a tangible example of HPE’s integrated strategy that combines HPE Juniper Networking technologies with HPE compute, exascale liquid cooling, and rack-scale engineering to deliver an integrated, open Ethernet-based scale-up AI solution.