OpenVINO™ Model Server – Edge AI Extension Module

[Alpa]

Hi, 

Is there a way to be able to run the Model Server on a Stratix10 FPGA? I see the OpenVINO support only for A10 FPGA listed interms of supported accelerators. 

[Munesh_Intel]

Hi Alpa,

Thanks for reaching out to us.

OpenVINO™ Model Server – Edge AI Extension Module does not provide FPGA plugin support. This information is mentioned in the following links:

Target device selection for models

OpenVINO™ Model Server – Edge AI Extension Module - in the ‘Highlights’ section.

However, in the backend, OVMS uses OpenVINO™ Toolkit Inference Engine & Models, which enables deployment of deep learning inference workloads for maximum performance across Intel® architectures.

In this scenario, Intel® Arria® 10 FPGA can be deployed, since OpenVINO™ Toolkit already provides support for Intel® Vision Accelerator Design with an Intel® Arria® 10 FPGA and Intel® Programmable Acceleration Card with Intel® Arria® 10 GX FPGA up to OpenVINO™ 2020.3.2 LTS Release.

Beginning OpenVINO™ 2021.4 LTS Release, OpenVINO™ support for FPGA has been deprecated, and OpenVINO™ Toolkit will not support FPGAs independent of a future Intel® FPGA AI Suite product.

More information regarding this is available in Release Notes for Intel® Distribution of OpenVINO™ Toolkit 2021.4 LTS, in the ‘Support Change and Deprecation Notices’ section.

Regards,

Munesh

[Munesh_Intel]

Hi Alpa,

This thread will no longer be monitored since we have provided a solution. If you need any additional information from Intel, please submit a new question.

Regards,

Munesh