HPE ProLiant DL325 Gen12

The HPE ProLiant DL325 Gen12 is a purpose-built 1U single-socket server powered by the AMD EPYC 9005 series processors (codenamed Turin), delivering up to 192 cores within a compact rack footprint. Built on TSMC's 3nm process node and leveraging AMD's Zen 5 microarchitecture, the EPYC 9005 processors integrated into this platform provide unprecedented compute density for workloads requiring massive parallelism, high core counts, and exceptional memory bandwidth. The platform supports up to 12 channels of DDR5 memory per socket, enabling aggregate memory bandwidths exceeding 500 GB/s, which is critical for in-memory databases, real-time analytics, and AI inference pipelines running at the edge or core datacenter.

The DL325 Gen12 is specifically optimized for cloud-native and scale-out enterprise architectures where per-server software licensing costs make single-socket efficiency a strategic advantage. Its 1U form factor maximizes rack density while the PCIe 5.0 fabric delivers high-throughput connectivity for NVMe storage and high-speed networking accelerators. HPE's iLO 7 baseboard management controller provides advanced security features including silicon root of trust, firmware integrity validation, and integration with HPE Compute Ops Manager for cloud-based fleet lifecycle management across hybrid environments.

From a storage and I/O perspective, the DL325 Gen12 supports a flexible SFF/LFF drive bay configuration accommodating up to 10 SFF NVMe or SAS/SATA drives, complemented by an OCP 3.0 mezzanine slot and multiple PCIe 5.0 expansion slots for network adapters, accelerators, or additional NVMe controllers. The server ships with HPE's Embedded 10GbE connectivity and is validated for major hypervisors, Kubernetes distributions, and cloud-native stacks including Red Hat OpenShift and VMware vSphere, making it a reliable foundation for enterprise modernization initiatives.

Ideal for

• Scale-out cloud-native Kubernetes worker node clusters where high core count per 1U minimizes rack footprint and software licensing overhead
• Virtualized infrastructure consolidation using VMware vSphere or Red Hat OpenShift Virtualization, leveraging up to 192 cores for maximum VM density
• In-memory database workloads such as SAP HANA scale-up instances and Redis Enterprise, benefiting from DDR5 multi-channel bandwidth exceeding 500 GB/s
• AI/ML inference serving and edge AI deployments requiring CPU-based inferencing with high core parallelism for batch inference pipelines
• High-throughput software-defined networking and network function virtualization (NFV) platforms demanding core density and PCIe 5.0 NIC throughput
• Media transcoding, rendering farms, and HPC simulation workloads that scale linearly with available compute threads across dense rack deployments

Technical specifications

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