Brand: HP | Category: Servers
SKU: HP-P75312B21 | Part #: P75312-B21 | MPN: P75312-B21
Contact for Pricing — Request a Quote
The HPE ProLiant Compute DL384 Gen12 from HP is enterprise-grade Servers hardware built for data centers. HPE ProLiant DL384 Gen12 is a purpose-built AI accelerated server powered by NVIDIA GH200 Grace Hopper Superchips for extreme-scale large language model training and inference. Available in brand new condition. Available from Omnixon Global — contact our team to request a quote.
| Brand | HP |
| Category | Servers |
| SKU | HP-P75312B21 |
| Part Number | P75312-B21 |
| Condition | New |
| Product Line | HPE ProLiant Compute |
| Model | DL384 Gen12 |
| Form Factor | 4U Rack Server |
| Accelerator Architecture | NVIDIA GH200 Grace Hopper Superchip |
| Number of GH200 Superchips | Up to 8 x NVIDIA GH200 Superchips |
| GPU Compute per Superchip | NVIDIA H100 Hopper GPU (SXM5-class), 80 SM / 132 SM Hopper architecture |
| GPU Memory per Superchip | 96 GB HBM3e at 3.35 TB/s bandwidth |
| CPU per Superchip | 72-core NVIDIA Grace (ARM Neoverse V2) at up to 3.1 GHz |
| CPU Memory per Superchip | Up to 480 GB LPDDR5X at 512 GB/s |
| CPU-GPU Interconnect | NVLink-C2C, 900 GB/s bidirectional per Superchip |
| GPU-to-GPU Interconnect | NVLink 4.0 / NVSwitch 3rd Gen fabric, up to 900 GB/s per GPU all-to-all |
| Total System GPU Memory (8x GH200) | Up to 768 GB HBM3e |
| Network I/O | Up to 4 x NVIDIA ConnectX-7 NDR 400GbE / InfiniBand 400Gb/s ports |
| Storage | Up to 8 x 2.5-inch NVMe SSD bays (PCIe Gen5), supports HPE SCM and U.3 drives |
| System Management | HPE iLO 7 with AI-assisted telemetry, Redfish API, SNMP, HPE Integrated Smart Update Manager (ISSUM) |
| Cooling | N+1 redundant hot-plug fans with direct liquid cooling (DLC) rear-door heat exchanger support; compliant with ASHRAE A3/A4 with DLC |
| Power Supply | Redundant 2+2 Titanium-efficiency (96%+) HPE Flex Slot PSUs, up to 3200W each |
| OS Support | Red Hat Enterprise Linux 9.x, Ubuntu 22.04/24.04 LTS, SUSE Linux Enterprise Server 15 SP5+, VMware vSphere (GPU passthrough) |
| Software Ecosystem | NVIDIA AI Enterprise 5.x, NVIDIA Base Command Manager, CUDA 12.x, cuDNN 9.x, TensorRT-LLM, HPE GreenLake for Compute Ops Management |
| Chassis Dimensions | 4U, 447mm (W) x 1752mm max depth x 175mm (H) |
| Generation | Gen12 (launched 2025 Q3) |
The HPE ProLiant Compute DL384 Gen12 is built for enterprise data-center workloads — virtualization (VMware, Proxmox, Nutanix), private cloud, database hosting, and AI/ML training. It fits standard EIA-310 server racks and supports redundant PSUs and hot-swap drives common in production environments.
Key specifications for the HPE ProLiant Compute DL384 Gen12: new condition; manufacturer HP (Hewlett Packard Enterprise); product line HPE ProLiant Compute; model DL384 Gen12; form factor 4U Rack Server; accelerator architecture NVIDIA GH200 Grace Hopper Superchip; number of gh200 superchips Up to 8 x NVIDIA GH200 Superchips. Manufacturer part number P75312-B21. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.