Product Overview Do you require higher performance computation or graphics? Companies are being faced with greater computational and graphics requirements as large, complex computational models become more commonplace. Traditional CPU technology is no longer able to keep up with these increasing demands. NVIDIA Accelerators for HPE ProLiant Servers seamlessly integrate graphics processing unit (GPU) computing with select HPE ProLiant Server families. Designed from the ground up for power-efficient, high performance supercomputing, NVIDIA® Accelerators deliver dramatically higher application acceleration than a CPU-only approach for a range of deep learning, scientific, and commercial applications.
The thousands of NVIDIA CUDA® cores of each accelerator allow it to divide large computing or graphics tasks into thousands of smaller tasks that can be run concurrently, thus enabling much faster simulations and improved graphics fidelity for extremely demanding 3D models. Product Specification Peak single precision floating point performance 12 TFlops Number of accelerators per card 1 Cores 3840 Memory size per board (GDDR5) 24GB GDDR5 Memory bandwidth for board (ECC off) 346 GB/s Accelerator applications Deep learning Architecture features Deep learning models typically take days to weeks to train, forcing scientists to make compromises between accuracy and time to deployment.
The NVIDIA Tesla P40 GPU accelerator, based on the NVIDIA Pascal™ architecture, is designed to deliver the highest combination of single precision performance together with high memory density, as required for deep learning training.
The thousands of NVIDIA CUDA® cores of each accelerator allow it to divide large computing or graphics tasks into thousands of smaller tasks that can be run concurrently, thus enabling much faster simulations and improved graphics fidelity for extremely demanding 3D models. Product Specification Peak single precision floating point performance 12 TFlops Number of accelerators per card 1 Cores 3840 Memory size per board (GDDR5) 24GB GDDR5 Memory bandwidth for board (ECC off) 346 GB/s Accelerator applications Deep learning Architecture features Deep learning models typically take days to weeks to train, forcing scientists to make compromises between accuracy and time to deployment.
The NVIDIA Tesla P40 GPU accelerator, based on the NVIDIA Pascal™ architecture, is designed to deliver the highest combination of single precision performance together with high memory density, as required for deep learning training.
