HPE Partner · 2U NVIDIA Grace HopperGen12 — Aktuell

HPE ProLiant DL384 Gen12

NVIDIA Grace CPU + Hopper GPU · bis zu 144 Kerne · 1.2 TB Unified · iLO 6

HPE ProLiant DL384 Gen12
Polarise

Europe's leading sovereign AI infrastructure platform trusts

clouditiv
Arista Networks
HPE
Authorized Partner

Rooted in critical infrastructure

Vodafone
Telefónica
M-net
Commerzbank
Arvato

clouditiv builds on 30+ years of SETUP Protokolltester GmbH's experience in telecommunications, finance, and critical infrastructure.

Configure your server

Configure HPE ProLiant DL384 Gen12

Build your exact configuration — choose chassis, CPU, RAM, storage and more.

Chassis / Base Model

Select your server form factor

Required

configurator.catGraceHopperDesc

High-performance NVMe SSDs for the fastest storage

0 of 4 slots used

Select a RAID controller for data protection

OCP 3.0 mezzanine network adapters

PCIe network adapter cards

0 of 2 slots used

Choose redundant power supply units

Select an OS boot device (SD, USB, or M.2)

Rack mounting rails for your server

Power cables for your PSU connection

iLO Advanced management license and cables

Prices are indicative and subject to change. Final pricing is confirmed in your personalised quote.

Your Configuration

HPE ProLiant DL384 Gen12

DL384 Gen12 1× Grace Hopper Module · 4× SFF NVMe

HPE NS204i-u Gen12 x4 NVMe 2-port Boot Controller

P51788-B21

Mellanox MCX623106AN ConnectX-6 Dx 100Gb 2-port QSFP56 OCP3

P60509-B21

HPE 2200W Flex Slot Platinum Hot Plug Low Halogen PSU

P51094-B21

HPE 64GB MicroSD RAID-1 USB Boot Flash Media Kit

P55776-B21

TotalOn request

* Indicative price excl. VAT. Binding price in the quote.

Why choose HPE ProLiant DL384

Engineered for the most demanding workloads in its class.

GPU-Optimised Architecture

Designed to maximise GPU performance with high-bandwidth PCIe interconnects and optimised airflow.

Extreme AI Performance

Support for multiple enterprise GPUs delivers the compute density needed for LLM training and inference.

Open Infrastructure

Standards-based platform runs any AI/ML framework — PyTorch, TensorFlow, CUDA — without vendor lock-in.

Use cases

What the HPE ProLiant DL384 is designed for.

Generative AI & LLM

Purpose-built for fine-tuning and running large language models with maximum GPU density.

Gen AILLMFine-Tuning

Real-Time AI Inference

RAG pipelines and real-time inference workloads benefit from the DL384's extreme GPU bandwidth.

RAGInferenceReal-Time AI

HPC & Scientific Computing

High-performance computing, simulation and scientific workloads at research-lab scale.

HPCSimulationScientific
Perfect for:Generative AILLM Training & InferenceHPC / SimulationMLPerf Leader
HPE iLO 6

Intelligent remote management

HPE iLO gives you full out-of-band management — power cycling, console access, firmware updates and health monitoring, all without physical access.

Integrated Lights-Out 6

Redfish API · HTML5 KVM · Silicon Root of Trust

Remote Console

Full HTML5 KVM access to the system console from anywhere.

Power & Lifecycle

Remote power on/off, reset and firmware update without physical access.

Redfish API

Industry-standard API for automation and integration with infrastructure tools.

Health Monitoring

Real-time hardware health data — temperature, fans, PSU and more.

HPE Security

Enterprise security, built in

HPE servers are designed from the silicon up with security as a core requirement — not an afterthought.

Silicon Root of Trust

Hardware-anchored boot chain verification prevents firmware tampering at the lowest level.

Secure Encryption

Hardware-accelerated encryption for data at rest and in transit.

BIOS/UEFI Protection

Signed firmware and secure boot ensure only trusted code runs on your hardware.

How we deliver your hardware

From your order to a running system — handled end to end by us.

Step 01
< 30 min

Configure & Order

Choose your chassis, CPU, RAM and storage using our configurator. We generate a binding quote within 24 hours.

Step 02
2–5 business days

Integration & Testing

We assemble, configure and burn-in your system in our integration center. Full factory acceptance test included.

Step 03
Delivery incl.

Rack-Ready Delivery

Your server arrives configured, tested and ready to rack. Optional on-site installation available.