HPE ProLiant DL380 Gen11 NVIDIA H100 NVL Configure-to-Order Server

The HPE ProLiant DL380 Gen11 NVIDIA H100 NVL Configure-to-Order Server (P60633-B21) is a purpose-built, rack-optimized 2U platform designed to accelerate enterprise AI, deep learning, and high-performance computing workloads at scale. Built on the Intel Xeon Scalable 4th Generation (Sapphire Rapids) processor architecture and integrated with NVIDIA H100 NVL GPU technology based on the Hopper architecture, this server delivers exceptional compute density and memory bandwidth for the most demanding datacenter environments. The H100 NVL GPU features 188GB of HBM3 memory across a dual-GPU NVLink configuration, enabling transformer model training and large-scale inference at unprecedented throughput.

The DL380 Gen11 chassis supports the HPE Configure-to-Order (CTO) model, allowing enterprises to tailor memory, storage, networking, and GPU configurations to precise workload requirements without over-provisioning. The platform accommodates up to 8TB of DDR5 system memory across 32 DIMM slots with support for high-speed PCIe Gen5 interconnects, ensuring that data pipelines between CPUs, GPUs, and NVMe storage subsystems remain free of bottlenecks. HPE's silicon root of trust and integrated iLO 6 management controller provide hardware-level security and remote lifecycle management across distributed datacenter deployments.

Designed for enterprise IT teams operating in datacenter, cloud, and AI infrastructure roles, the ProLiant DL380 Gen11 H100 NVL CTO server is validated and supported through HPE's full ecosystem including HPE GreenLake cloud services integration, Pointnext operational support, and compatibility with major AI frameworks including PyTorch, TensorFlow, and NVIDIA AI Enterprise software. The system targets organizations in financial services, healthcare, research, and large-scale enterprise AI buildouts requiring reliable, serviceable, and scalable GPU compute infrastructure.

Ideal for

• Large language model (LLM) training and fine-tuning requiring high HBM3 memory capacity and NVLink GPU-to-GPU bandwidth
• Enterprise-scale AI inference serving for real-time natural language processing, recommendation engines, and computer vision pipelines
• High-performance scientific computing and simulation workloads in research institutions and national laboratories
• Genomics, drug discovery, and medical imaging AI in regulated healthcare and life sciences environments
• Financial services risk modeling, quantitative analytics, and fraud detection requiring deterministic high-throughput compute
• Datacenter AI infrastructure consolidation replacing multiple smaller GPU nodes with a dense, manageable 2U platform

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.