Q0V80A - Refurbished - HPE NVIDIA Tesla P40 24GB Computational Accelerator

$2,349.85

Selling fast

Condition: Refurbished

Requires 8P Keyed Cable Kit (871829-B21)

HPE NVIDIA Tesla P40 24GB Computational Accelerator (Part Number: Q0V80A) is a high-performance GPU designed for data center applications such as deep learning inference, machine learning, and high-performance computing (HPC). It leverages NVIDIA's Pascal architecture and is optimized for inference workloads.

Specifications

  • GPU Architecture: NVIDIA Pascal (GP102)

  • CUDA Cores: 3,840

  • Base Clock: 1,300 MHz

  • Boost Clock: 1,530 MHz

  • Memory: 24 GB GDDR5

  • Memory Interface: 384-bit

  • Memory Bandwidth: 346 GB/s

  • Interface: PCI Express 3.0 x16

  • Form Factor: Dual-slot

  • Cooling: Passive (requires adequate system airflow)

  • Power Consumption: 250W (1x 8-pin PCIe power connector)

  • ECC Support: Yes

  • Display Outputs: None

Performance Highlights

  • FP32 (Single Precision): 11.7 TFLOPS

  • FP64 (Double Precision): 367 GFLOPS

  • INT8 Inference Performance: 47 TOPS

Compatibility & Deployment

The Tesla P40 is compatible with systems that have a PCIe 3.0 x16 slot and sufficient airflow for passive cooling. It has been used in various server models, including:

  • HPE Apollo pc40

  • HPE ProLiant DL380 Gen9

  • HPE ProLiant XL190r Gen9

  • HPE Synergy 480 Gen10

Delivery

Share some information about your product

Share some information about your product