• HPE R9H23C gold PCIe 4.0 x16 GPU with 16GB GDDR6, low-profile design
1 1

HPE

HPE R9H23C | A2 16GB Tensor Core GPU, PCIe Gen4 x8, 60W, Passive

R9H23C

Stock Status: Enquire

Request Quote
Sale Sold out
Taxes included. Shipping calculated at checkout.

Description

The HPE NVIDIA A2 16GB PCIe Accelerator delivers entry-level AI inference and edge computing performance in a compact, power-efficient design. Powered by NVIDIA Ampere architecture with 1280 CUDA cores, 40 Tensor cores, and 16GB GDDR6 memory, this low-profile, single-slot GPU accelerates machine learning inference, computer vision, and analytics workloads in space-constrained server environments. With 40-60W configurable TDP and passive cooling, the A2 enables AI deployment at scale across edge, data center, and cloud infrastructure.

Features

NVIDIA Ampere architecture with 1280 CUDA cores and 40 Tensor cores
- Third-generation Tensor Cores supporting INT4, INT8, FP16, TF32, and FP32 precision for AI workloads
- 16GB GDDR6 ECC memory with 200 GB/s bandwidth for large dataset processing
- PCIe Gen4 x8 interface delivering 32 GB/s bidirectional bandwidth
- Low-profile, single-slot, passive thermal design for space-constrained servers
- Configurable 40-60W TDP with no external power requirements
- RT Cores for ray tracing and rendering acceleration
- NVIDIA Triton Inference Server and DeepStream SDK support
- HPE iLO integration for remote configuration, health monitoring, and firmware updates
- CUDA 11.1+ and NVIDIA vGPU 14.0+ software compatibility
- Multi-instance GPU (MIG) technology not supported (single full-GPU allocation)

Warranty

All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability.

All technical data should be verified on the manufacturer data sheets.

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

GPU: NVIDIA A2 Tensor Core (Ampere GA107, 8nm)
- CUDA Cores: 1280
- Tensor Cores: 40 (3rd generation)
- Memory: 16GB GDDR6
- Memory Bandwidth: 200 GB/s
- Memory Interface: 128-bit
- Interface: PCIe Gen4 x8
- Form Factor: Low-profile, single-slot, passive cooling
- TDP: 40-60W configurable (no external power required)
- Compute Performance: 4.5 TFLOPs FP32, 18 TFLOPs FP16 Tensor, 9 TFLOPs TF32 Tensor, 36 TOPs INT8
- Dimensions: 1.37" H × 4.37" W × 10.51" L (34.80 × 111 × 266.95 mm)
- ECC: Enabled by default (can be disabled)
- Platform Support: HPE ProLiant Rack/Tower/BladeSystem/Synergy/Apollo servers
- Management: HPE iLO integration for monitoring, power and thermal control

FAQs

Q: What workloads is the HPE NVIDIA A2 optimized for?
A: The A2 is optimized for AI inference, edge computing, machine learning, computer vision, intelligent video analytics, transcoding, and data analytics workloads. It delivers entry-level inference performance with low power consumption and a compact footprint, making it ideal for deployment in space-constrained environments.

Q: Does the A2 require external power connectors?
A: No. The A2 draws 40-60W configurable TDP and is powered entirely through the PCIe slot, requiring no external power connectors. This simplifies installation in servers with limited power infrastructure.

Q: Is this GPU compatible with HPE ProLiant Gen10 Plus servers?
A: Yes, the R9H23C is designed for HPE ProLiant Rack, Tower, BladeSystem, Synergy, and Apollo platforms. Some configurations may require specific riser kits or high-performance fan/heatsink options; consult HPE compatibility documentation for your specific server model.

Q: Can the A2 be used for virtualization or VDI deployments?
A: Yes, the A2 supports NVIDIA vGPU software 14.0 or later, enabling GPU virtualization for virtual desktop infrastructure (VDI) and multi-tenant environments. It is a headless accelerator optimized for compute workloads rather than direct display output.

Q: How does the A2 compare to the NVIDIA T4 in performance?
A: The A2 offers 20-30% faster inference performance compared to the Turing-based T4 in intelligent edge use cases, with 60% better price-to-performance and 10% better power efficiency, while maintaining a smaller low-profile form factor.

Recently Viewed

  • Request a Quote

    Looking for competitive pricing? Submit a request, and our team will provide a tailored quote that fits your needs.

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Call us Now  

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Contact us