India, Dec. 3 -- Hewlett Packard Enterprise (HPE) has announced it will soon offer the AMD Helios AI rack-scale architecture, an open, integrated platform designed to accelerate the deployment of AI-ready data centres. Announced on the eve of the HPE Discover 2025 in Barcelona, the solution is built for cloud service providers (CSPs), including neoclouds, and supports large-scale AI training and inference workloads.

The solution connects 72 AMD Instinct MI455X GPUs per rack, delivering up to 2.9 AI exaflops of FP4 performance and 260 terabytes per second of aggregated scale-up bandwidth. It offers 31 TB of fourth-generation high-bandwidth memory (HBM4) and 1.4 PB/s of memory bandwidth, enabling it to support trillion-parameter AI models ...