• Green and white stylized eye logo above bold black nvidia text on HPE s0k90c NVIDIA L40 Ada Lovelace PCIe Gen4 GPU 48GB GDDR6
1 1

HPE

HPE S0K90C | NVIDIA L40 Ada Lovelace PCIe Gen4 GPU, 48GB GDDR6, 90 TFLOPS

HPE S0K90C | NVIDIA L40 Ada Lovelace PCIe Gen4 GPU, 48GB GDDR6, 90 TFLOPS

S0K90C

Stock Status: Enquire

$0.00 AUD
$0.00 AUD
$0.00 AUD
Sale Stock Status: Enquire
Taxes included. Shipping calculated at checkout.

Description

The NVIDIA L40 48GB PCIe Gen4 GPU Accelerator for HPE delivers 90 TFLOPS FP32 performance on the Ada Lovelace architecture, making it a powerhouse for generative AI, large language models (LLM), computer vision, and real-time inference workloads in HPE data centre environments. With 48GB GDDR6 memory at 864 GB/s bandwidth and 724 TFLOPS FP8 Tensor Core performance (with sparsity), it handles the most demanding AI training and inference, neural graphics, virtual workstations, and batch rendering at scale.

Designed as a full-height, full-length dual-slot passive-cooled card drawing 300W, the L40 integrates with HPE iLO for health monitoring, power, and thermal management via Silicon Root of Trust. It supports NVIDIA vGPU software for multi-user memory allocation and is validated in NVIDIA Certified HPE server configurations including ProLiant DL380 Gen11, DL385 Gen11, DL560 Gen11, ML350 Gen11, and more. Order now at iptrading.com.au — trusted by Australian IT professionals.

Features

- Ada Lovelace architecture delivers 90 TFLOPS FP32 and 724 TFLOPS FP8 Tensor Core performance (with sparsity) for AI training and inference
- 48GB GDDR6 memory with 864 GB/s bandwidth supports large model sizes and complex datasets for LLMs, computer vision, and scientific workloads
- Real-time inference support for NLP, fraud detection, predictive maintenance, and medical imaging
- Hardware-accelerated ray tracing, AI shading, and simulation for NVIDIA Omniverse, cloud gaming, and batch rendering
- PCIe Gen4 x16 interface doubles bandwidth vs Gen3 for faster data throughput in AI and data science pipelines
- NVIDIA vGPU software support for partitioned GPU memory allocation across multiple virtual machines or users
- HPE iLO integration for accelerator health monitoring, power and thermal control via Silicon Root of Trust
- Thermal, mechanical, power, and signal integrity qualified under the NVIDIA Accelerators for HPE programme
- NVIDIA Certified HPE server configurations validated for multi-GPU and multi-node performance at scale
- Passive cooling design — relies on server chassis airflow for quiet and efficient thermal management

Warranty

Hewlett Packard Enterprise (HPE) offers a standard limited hardware warranty on its products, with coverage periods varying by product line. For instance, HPE ProLiant servers typically come with a 1 to 3-year limited warranty, while other products may have different durations.
Extended Warranty Options via IP Trading:
IP Trading offers tailored extended warranty solutions and support packages for HPE products, ensuring your IT infrastructure remains protected beyond the standard warranty period. Contact our team to explore options that align with your organisation's needs.

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

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

- GPU Architecture: NVIDIA Ada Lovelace
- Memory: 48GB GDDR6
- Memory Bandwidth: 864 GB/s
- FP32 Performance: 90 TFLOPS
- FP8 Tensor Core Performance: 724 TFLOPS (with sparsity)
- System Interface: PCIe Gen4 x16
- Form Factor: Full Height, Full Length (FHFL), Dual-Slot
- Dimensions: 4.4" H x 10.5" L
- Thermal Design Power: 300W
- Cooling: Passive
- Number of GPUs per Card: 1
- Part Number: S0K90C

FAQs

Q: Which HPE servers are compatible with the NVIDIA L40 48GB PCIe Accelerator (S0K90C)?
A: The L40 is validated for use in HPE ProLiant DL320 Gen11, DL380 Gen10 Plus, DL380 Gen11, DL380a Gen11, DL385 Gen11, DL560 Gen11, ML350 Gen11, XL675d Gen10 Plus, as well as HPE Superdome 3200, Superdome Flex, and Superdome Flex 280. Always verify your specific server's PCIe slot availability and power budget before ordering.

Q: What AI and compute workloads is the NVIDIA L40 best suited for?
A: The L40 excels at generative AI, large language model (LLM) training and inference, computer vision, natural language processing, fraud detection, medical imaging, scientific discovery, financial modelling, and neural graphics workloads such as hardware-accelerated ray tracing and NVIDIA Omniverse.

Q: Does the NVIDIA L40 support GPU virtualisation?
A: Yes. The L40 supports NVIDIA vGPU software, enabling the 48GB GDDR6 memory pool to be partitioned and allocated across multiple virtual machines or users — ideal for virtual workstation deployments and large-scale multi-tenant AI inference environments.

Q: Is the HPE NVIDIA L40 passively cooled, and what are the power requirements?
A: Yes, the L40 is passively cooled and relies on server chassis airflow for thermal management. It has a Thermal Design Power (TDP) of 300W, so ensure your target HPE server has sufficient power headroom and adequate chassis cooling before deploying.

Q: Can I purchase the HPE NVIDIA L40 48GB PCIe Accelerator (S0K90C) in Australia?
A: Yes. IP Trading (iptrading.com.au) stocks and supplies the HPE NVIDIA L40 48GB PCIe Accelerator (S0K90C) to Australian businesses. Our team can assist with compatibility checks, multi-unit orders, and HPE server configurations. Contact us or order directly online.

Q: How does PCIe Gen4 benefit AI workloads compared to PCIe Gen3?
A: PCIe Gen4 doubles the bandwidth of PCIe Gen3 (up to 64 GB/s vs 32 GB/s on a x16 slot), significantly reducing data transfer bottlenecks between the CPU and GPU. This is particularly beneficial for AI training, data science pipelines, and inference workloads that are sensitive to memory and I/O throughput.

Recently Viewed

  • Request for 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