Deep learning and Artificial Intelligence are the most computationally intensive applications facing technology today. The combination of large data sets, high-performance computational capabilities and evolving and improving algorithms have long presented technical challenges that have been difficult to overcome...

until now.

NVXL is developing a revolutionary, massively scalable FPGA based software-defined acceleration platform demo, along with NVXL Acceleration Layer Software features and applications currently supported on this ground-breaking platform.

With unprecedented flexibility and scale, NVXL is transforming the way compute acceleration is deployed, consumed, and delivered. Running on standard hardware, NVXL’s platform handles low-level implementation details so developers and data scientists can be more productive. The platform can scale to thousands of FPGA and ASIC compute acceleration devices connected via a redundant, persistent fabric and can virtualize the compute power required for a myriad of cloud-scale applications. 

 

500.jpg
 

KEY BENEFITS

  • +

    Massively cloud-scalable software-defined compute acceleration

  • +

    Elastic scaling

    • Across 1000s of FPGAs and ASICs
    • Accelerators that accommodate rapidly evolving workloads in the data center
    • Efficient Performance & Model Scaling
  • +

    Dramatic reduction in Total Cost of Ownership

  • +

    Extreme performance on-demand