HPE ProLiant DL560 Gen12

The HPE ProLiant DL560 Gen12 is a 4U four-socket rack server built around Intel Xeon 6 processors with High Bandwidth Memory (HBM), delivering an unprecedented combination of compute density and memory bandwidth in a data-center-optimized chassis. By supporting HBM-equipped Xeon 6 SKUs, the DL560 Gen12 enables system-level memory bandwidth that far exceeds what conventional DDR5 DIMM configurations can achieve, making it purpose-built for workloads where data throughput between CPU and memory is the primary performance constraint. The platform supports multi-socket NUMA topologies across four processor sockets, allowing enterprises to scale both core count and memory capacity simultaneously without fragmenting workloads across separate nodes.

The DL560 Gen12 targets mission-critical in-memory database platforms such as SAP HANA, Oracle Database In-Memory, and Microsoft SQL Server, where terabytes of working data must be accessed at near-cache latencies. The server's architecture also addresses the growing demand for large-scale AI inference consolidation, allowing multiple AI models to reside simultaneously in a unified high-bandwidth memory pool rather than requiring GPU offload for memory-resident inference tasks. HPE's Gen12 platform refresh introduces updated PCIe Gen 5 I/O fabric, enhanced RAS (Reliability, Availability, and Serviceability) features, and tighter integration with HPE iLO 7 for intelligent telemetry-driven management.

From a management and lifecycle perspective, the DL560 Gen12 integrates with HPE GreenLake cloud services, HPE Integrated Smart Update Tools, and the broader HPE OneView infrastructure management ecosystem, enabling automated firmware lifecycle management, predictive health analytics, and policy-driven workload orchestration. The server is designed for enterprise data centers requiring maximum uptime, with redundant power supplies, hot-plug NVMe storage, and advanced memory fault tolerance mechanisms ensuring continuous availability for Tier-1 business applications.

Ideal for

• SAP HANA scale-up deployments requiring terabyte-scale in-memory OLAP and OLTP processing on a single system image
• Large-scale AI inference serving where multiple LLM or recommendation models must reside concurrently in high-bandwidth memory for low-latency response
• Oracle Database In-Memory and Microsoft SQL Server in-memory columnstore workloads demanding sustained multi-TB/s memory bandwidth
• High-frequency financial analytics and real-time risk calculation engines processing massive in-memory datasets across four NUMA domains
• Genomics and life sciences bioinformatics pipelines requiring rapid random-access traversal of large reference datasets held entirely in memory
• Enterprise data warehousing consolidation where multiple analytical workloads are co-located on a single high-memory-bandwidth four-socket platform

Technical specifications

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