HPE ProLiant DL325 Gen11 1U Rack Server with AMD EPYC 9004 Processors, Compact Design & Dual Low-Profile GPU Support

The HPE ProLiant DL325 Gen11 is a compact 1U single-socket rack server built on AMD EPYC 9004 processors, optimized for cost-sensitive edge AI inference, video stream analytics, retail smart monitoring, lightweight virtualization, web caching and branch office high-concurrency workloads. This space-saving 1U chassis delivers outstanding multi-threaded performance, flexible low-profile GPU expansion, high-density NVMe flash storage, energy-efficient cooling and enterprise-grade remote management via HPE iLO 6. Compared with dual-socket and Intel single-socket alternatives, the AMD EPYC platform provides more cores per dollar, making the DL325 Gen11 the ideal cost-balanced compute node for mass-deployed edge AI clusters, multi-tenant lightweight services and distributed video analysis systems without excessive capital expenditure.

Powered by single-socket AMD EPYC 9004 processors with up to 96 cores, the DL325 Gen11 delivers massive multi-threading capacity for parallel video decoding, concurrent AI inference requests and web cache workloads. More cores at lower procurement cost than equivalent single-socket Intel servers reduce total cluster investment for large-scale edge deployments.

The slim 1U form factor maximizes rack node density, allowing dozens of inference servers to be deployed within a single standard cabinet. Tuned low-noise cooling fans support deployment in branch office server rooms and edge computer rooms without dedicated heavy cooling infrastructure.

Up to two low-profile double-width GPUs fit inside the chassis, perfectly optimized for lightweight large language model serving, multi-camera video analytics, retail object recognition and VDI visualization. It avoids the high cost of full-height multi-GPU training servers for edge-only inference tasks.

16 memory slots scale to a maximum of 4TB ECC DDR5 memory, enabling multiple AI models and large feature datasets to be fully loaded into memory to cut repeated disk I/O latency. Hardware memory error correction prevents service crashes during unattended 7×24 edge operation.

The 10-bay 2.5-inch U.3 NVMe configuration delivers ultra-low flash latency for caching model weights, video footage and inference media files. Universal AnyBay trays support mixed deployment of high-speed NVMe SSDs and high-capacity SATA HDDs to balance performance and storage costs for edge sites.

80 PLUS Platinum high-efficiency power supplies cut long-term electricity consumption for mass edge clusters. Optional N+1 redundant power configuration removes single power failure points, ensuring stable online AI services for smart city, transportation and retail edge nodes.

The independent iLO 6 management chip enables centralized remote monitoring and maintenance for hundreds of geographically distributed edge servers. IT teams can diagnose, reboot and upgrade firmware remotely without traveling to scattered branch and edge locations, drastically reducing field maintenance labor costs.

• 1U Single-Socket AMD Edge Server: Compact single EPYC chassis focused on cost-effective AI inference and distributed edge workloads
• iLO 6 (Integrated Lights-Out): HPE exclusive out-of-band remote management system independent of server OS
• Smart Array RAID: HPE embedded storage controller offering drive mirroring and striping for edge data protection
• 4th Gen AMD EPYC SP5: High-core-count enterprise processor platform optimized for parallel multi-threaded workloads
• 80 PLUS Platinum PSU: Ultra-high efficiency modular power supply to lower edge cluster power expenditure
• OCP 3.0 Mezzanine Slot: Modular expansion port for high-speed 10GbE/25GbE cluster networking upgrades

1. White background full front chassis render (Homepage top banner main cover image)
2. Front 2.5-inch U.3 AnyBay drive bay close-up (Storage & edge dataset caching section)
3. Internal CPU, memory & dual low-profile GPU layout diagram (Edge AI acceleration selling point block)

1. Top Banner Area: Place the white-background full chassis render paired with full product title — HPE ProLiant DL325 Gen11 1U Single-Socket AMD Edge AI Rack Server
2. Storage Specification Table: Embed drive bay close-up thumbnail beside the parameter table
3. Core Edge AI Inference Performance Section: Display internal hardware layout diagram below feature text content

Price Overview

The HPE ProLiant DL325 Gen11 single-socket 1U AMD edge rack server provides cost-effective customizable configurations for distributed AI inference and branch workloads. Entry barebone units equipped with entry-level AMD EPYC CPU, 32GB DDR5 memory and single Platinum PSU start at USD $1,520 / CNY ¥10,300. Standard mainstream edge inference nodes with mid-tier EPYC processor, 256GB ECC memory, 4 × U.3 NVMe SSD and one low-profile GPU range from USD $3,280–$6,150. Fully expanded high-concurrency edge servers with top-tier 96-core EPYC CPU, maximum 4TB memory, full 10-bay NVMe flash array, dual GPUs and N+1 redundant power can exceed USD $8,700. Final pricing varies by CPU core tier, total memory capacity, drive bay configuration and storage media, GPU model and quantity, power redundancy mode, high-speed OCP network upgrades, warranty service tiers and bulk edge cluster order discounts.