Do you require higher performance for artificial intelligence (AI) training and inference, high-performance computing (HPC) or graphics? NVIDIA® Accelerators for HPE help solve the world’s most important scientific, industrial, and business challenges with AI and HPC. Visualize complex content to create cutting-edge products, tell immersive stories, and reimagine cities of the future. Extract new insights from massive datasets. Hewlett Packard Enterprise servers with NVIDIA accelerators are designed for the age of elastic computing, providing unmatched acceleration at every scale.
Develop and Deploy AI at Any Scale
- Build new AI models with supervised or unsupervised training for generative AI, computer vision, large language models, scientific discovery, and financial market modeling with NVIDIA accelerators and HPE Cray systems.
- Get real-time inference for computer vision, natural language processing, fraud detection, predictive maintenance, and medical imaging with NVIDIA accelerators and HPE ProLiant Compute Servers.
- NVIDIA Accelerators for HPE improve computational performance, dramatically reducing the completion time for parallel tasks, offering quicker time to solutions.
NVIDIA Qualified and NVIDIA Certified HPE Servers
- NVIDIA Accelerators for HPE undergo thermal, mechanical, power, and signal integrity qualification to validate that the accelerator is fully functional in the server. NVIDIA Qualified configurations are supported for production use.
- NVIDIA Certified HPE servers are tested to validate both multi-GPU and multi-node performance for a diverse set of workloads to deliver excellent application performance, manageability, security, and scalability.
HPE Integrated Lights-Out (iLO) Management
- HPE iLO server management software enables you to securely configure, monitor, and update your NVIDIA Accelerators for HPE seamlessly, from anywhere in the world.
- Integrated Lights-Out (iLO) is an embedded technology that helps simplify server and accelerator set up, health monitoring, power, and thermal control, utilizing HPE's Silicon Root of Trust. More information: Weight & dimensions | Weight | 1020 g |  | Length | 111 mm | | Depth | 241.3 mm | | Height | 17.3 mm | Memory | Discrete graphics card memory | 20 GB | | Graphics card memory type | GDDR6 | | Memory bus | 160 bit | | Memory bandwidth (max) | 360 GB/s | Processor | Graphics processor family | NVIDIA | | Graphics processor | RTX 4000 Ada | | CUDA | Yes | | CUDA cores | 6144 | System requirements | Linux operating systems supported | Yes | | Server operating systems supported | Yes | Power | Power consumption (typical) | 130 W | Performance | DirectX version | 12.0 | | Shader model version | 6.7 | | OpenGL version | 4.6 | | OpenCL version | 3.0 | Design | Cooling type | Active | Ports & interfaces | Interface type | PCI Express x16 4.0 | | HDMI ports quantity | 0 | | DVI-D ports quantity | 0 | | DVI-I ports quantity | 0 | | DisplayPorts quantity | 4 | | DisplayPort version | 1.4a |
|