A 4U two-socket, accelerator-optimized server engineered with an ultra-scalable architecture to deliver next-generation AI performance — up to eight double-wide GPUs, Intel® Xeon® 6 processors, and high-speed PCIe Gen5 I/O.
Are you looking for a server with advanced GPU scalability to accelerate AI inferencing for your largest workloads?
The HPE ProLiant Compute DL380a Gen12 is engineered with an ultra-scalable architecture to deliver next-generation AI performance for your enterprise needs — the "a" stands for accelerator optimized. This 4U 2P rack server pairs the latest Intel® Xeon® 6 processors with dense GPU acceleration, blazing-fast memory bandwidth, and PCIe Gen5 I/O to accelerate AI inferencing for large workloads with enterprise-grade reliability and industry-leading security.
| Form factor | 4U rack, two-socket (2P) |
|---|---|
| Processors | Intel® Xeon® 6, up to 144 cores (2 CPUs) |
| Memory | 32 DIMM slots, up to 8 TB DDR5 (up to 6400 MT/s at 1 DIMM per channel) |
| GPU support | Up to 8 double-wide or 16 single-wide GPUs |
| Storage | Up to 8 SFF NVMe (U.3) or 16 EDSFF E3.S drives; HPE NS204i-u v2 boot device option |
| Expansion | PCIe Gen5 — up to 6 x16 slots plus 2 OCP 3.0 slots |
| Cooling | Air-cooled with hot-plug fan assemblies |
| Management & security | HPE iLO 7, HPE Compute Ops Management, Silicon Root of Trust |
GPU model compatibility (e.g. NVIDIA L4, L40S, H100 NVL, H200, RTX PRO 6000 Blackwell) varies by configuration; contact us to confirm the accelerators certified for your build.
Download datasheet: HPE ProLiant Compute DL380a Gen12 QuickSpecs (HPE.com)
An excellent choice for AI, compute, and graphics-intensive workloads requiring maximum GPU density, core count, and I/O bandwidth. With up to 8 double-wide GPUs, up to 144 Intel Xeon 6 cores, up to 8 TB of DDR5, and PCIe Gen5 connectivity, the DL380a Gen12 is built to accelerate AI inferencing for large workloads.
The HPE Silicon Root of Trust establishes a zero-trust security framework at the silicon level to ensure firmware integrity, continuously detecting compromised servers and preventing them from booting if malicious code is found. HPE ProLiant Compute servers provide automated recovery from a security event, including restoration of validated firmware.
Simplify and secure server management from edge to cloud with HPE Compute Ops Management, an as-a-service experience that delivers global visibility, AI-informed insights, and automation for deployment, scaling, and lifecycle management. Available as a standalone purchase or as a service via HPE GreenLake.
Tell us about your AI and accelerated workloads and a ServerComputeWorks specialist will help you choose GPUs, spec the configuration, price it, and place the order.
Request a Quote*Intel and Intel Xeon are trademarks of Intel Corporation in the U.S. and other countries. NVIDIA and related GPU product names are trademarks and/or registered trademarks of NVIDIA Corporation. All other third-party trademark(s) is/are property of their respective owner(s). Specifications and availability are subject to change.