HPE DL380A Gen12 (4U Dual-Socket AI Server with Intel® Xeon® 6 CPUs and Support for 10 Double-Width GPUs))

Description

This server is built for organizations that need serious GPU power. It supports up to 10 double-width GPUs, each up to 600W, and can connect four GPUs together using NVIDIA’s 4-way NVLink Bridge for faster AI inference. The system runs on dual Intel® Xeon® 6 processors, offering up to 144 cores total, and supports up to 4TB of DDR5 memory at blazing speeds of 6400 MT/s. The DL380a Gen12 is also designed with three separate power domains, including six dedicated power supplies for GPUs, ensuring stable and efficient performance even under heavy workloads.

Key Specifications

This server is built for the AI era. Whether you’re running deep learning models, processing massive datasets, or deploying inference workloads at scale, the DL380a Gen12 gives you the flexibility, speed, and reliability you need. It’s also designed to work seamlessly in hybrid cloud environments, with built-in tools for automation, monitoring, and lifecycle management. Security is baked into every layer—from silicon to firmware—so your data and applications stay protected. And with HPE’s management software and GreenLake services, you can scale and operate your infrastructure with ease.