Detailed Specifications
HPE R9W59A NVIDIA A10 24 GB GPU Accelerator
Overview
The HPE R9W59A is a single-slot server accelerator card based on the NVIDIA A10 GPU architecture, designed for enterprise and data-centre servers requiring high-throughput graphics, AI inference, or virtualised GPU workloads. Equipped with 24 GB of GDDR6 memory and integrated into HPE ProLiant or Apollo servers, this accelerator supports demanding workloads such as VDI, AI inferencing, and graphics rendering while fitting within standard airflow and power budgets of rack servers. With PCIe (Gen 4-capable) interface and enterprise-grade manageability, it enables customers to scale GPU resources smoothly without requiring bulky dual-slot cards. It is suited to designs where consolidation of GPU compute in a conventional server platform is required×be it for engineering graphics, virtual desktops, AI-driven inference pipelines or blended compute/graphics workloads.
Key Features
24 GB GDDR6 memory supporting substantial dataset workloads
Single-slot form-factor for efficient server installation
NVIDIA A10 GPU architecture with multi-workload support (graphics, AI, VDI)
Compatible with HPE server accelerators and frameworks for enterprise deployment
Technical Specifications
| Specification | Value |
|---|
| Brand | HPE |
| Part Number | R9W59A |
| Product Type | NVIDIA A10 24 GB GPU Accelerator |
| GPU Architecture | NVIDIA A10 |
| Memory Capacity | 24 GB GDDR6 |
| Memory Bandwidth | ~600 GB/s |
| Host Interface | PCI Express Gen4 (x16) |
| Form Factor | Single-slot server accelerator |
| Typical Dimensions (L × W × H) | 10.51 in × 4.37 in × 0.68 in |
| Weight | ~1.21 lb |
| Power Consumption (Typical) | ~150 W |
| Use Case | AI inference, VDI, graphics, HPC |
| Compatible Platforms | HPE ProLiant Gen10/Gen10+ servers |
| Virtualisation Support | GPU-virtualisation (vGPU), multi-workload |
| Cooling/Slot Requirement | Single-slot air-cooled server enclosure |
