Creative Compute
0

HPE ProLiant DL385 Gen11

Configurable systemConfigure to orderManufacturer documented

A mainstream enterprise server rather than a dedicated AI machine. Two AMD EPYC processors and a lot of storage, with room for some GPUs — a sensible route if AI is being added to an existing estate rather than built from scratch.

Ask Carl about this

No payment is taken online. Every request is an enquiry answered by an engineer.

Representative system image

Where does this sit?

  1. 01Local
  2. 02Workstation
  3. 03Departmental
  4. 04Enterprise
  5. 05Cluster
  6. 06Rack-scale

Team & Departmental. A server holding one to four accelerators, shared by a team or serving a single internal application. Typically bought by an sme or an internal it team. Comms room or a small rack.

Fit

Is this the right thing for you?

Who it's for

  • Enterprises adding inference to a conventional server estate
  • IT teams already standardised on ProLiant and EPYC
  • Departments with mixed database, virtualisation and AI workloads

What it runs

  • Moderate AI inference
  • Virtualisation
  • Databases and general enterprise workloads

When it is the wrong answer

  • Model training
  • High-density multi-GPU deployment
Where it goes
Comms room or datacentre rack
Complexity
moderate
Cooling
Air cooled
Specification

What the numbers mean

Every figure below is explained in plain English. Switch to the technical view for the bare specification.

System

Form factor

2U, two-socket rack server

Processors

2x AMD EPYC 9004 / 9005 series

MemoryUnverified

DDR5, up to 24 DIMM slots — maximum capacity awaiting verification

Storage baysUnverified

Up to 24 SFF, or mixed NVMe / SAS / SATA — exact count per configuration awaiting verification

GPU supportUnverified

Selected single and double-wide GPUs for inference — exact maximum awaiting verification

Networking

OCP 3.0 and embedded networking options

Power suppliesUnverified

Redundant Flexible Slot PSUs — wattage awaiting verification

Cooling

Air-cooled

Verification

Manufacturer documented — HPE ProLiant DL385 Gen11 product page and QuickSpecs a50004300enw. Reviewed 2026-08-31.

Manufacturer source
Before you commit

Practical considerations

The things that catch people out after the hardware has already been ordered.

  • This is a general-purpose server first. If AI is the primary workload, a DL380a Gen12 or an XD-series machine will serve you better.
  • The supported GPU list is narrower than the slot count suggests. We will confirm against your chosen card.
Completeness

Nothing arrives working on its own

What else will I need?

  • 2U rack space
  • Standard rack power
  • A view on which GPU actually suits the workload
AI networkingAI storageInstallationAsk Carl what I’m missing
Goes with

What normally sits alongside it

Relationships documented by the manufacturer, or by us during a deployment.