Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6403 Discussions

HDDL hardware initialization failed error with MyriadX card

swan_m_intel
Employee
1,100 Views

Hi,

I have a MyriadX card plugged into a Linux VM, and I need to run some inferencing on a container hosted on the VM.

For this I need to enable HDDL on my VM and I am following the steps mentioned here https://github.com/openvinotoolkit/docker_ci/blob/master/install_guide_vpu_hddl.md

I see an error "HDDL hardware initialization failed", seems like HDDL is not able to see the MyX card.

This is how I see the card on my VM >>

MyX-VM.PNG Error message while running 

source setupvars.sh && ./bin/hddldaemon -d

MyX-error1.PNG

Labels (3)
0 Kudos
4 Replies
Munesh_Intel
Moderator
1,088 Views

Hi Swanand,

Thanks for reaching out to us.

We have some questions for you.

1.    Which VM software are you using?

2.    Which MyriadX card are you using?

3.    Are you using IO Expander?

4.    Can you try running the verification scripts using -d MYRIAD for testing purposes?


Regards,

Munesh


0 Kudos
swan_m_intel
Employee
1,063 Views

Hi Munesh,

I ended up using Myriad plugin just to save sometime.

What are the benefits (performance benefit ?) of using HDDL plugin over Myriad ? 

Thanks,

Swanand

0 Kudos
Munesh_Intel
Moderator
1,051 Views

Hi Swanand,

The Inference Engine HDDL plugin is developed for inference of neural networks on Intel® Vision Accelerator Design with Intel® Movidius™ VPUs which is designed for use cases that require large throughput of deep learning inference. It provides multiple times of throughput as compared to MYRIAD Plugin.


The performance benchmark is given below:

https://docs.openvinotoolkit.org/2021.2/openvino_docs_performance_benchmarks.html


For example, the throughput value when using Intel® Vision Accelerator Design with Intel® Movidius™ VPUs (Uzel UI-AR8) is 39.19 FPS as compared to 4.92 FPS for Intel® Neural Compute Stick 2 for yolov3-TF public neural network.


Regards,

Munesh


0 Kudos
Munesh_Intel
Moderator
1,033 Views

Hi Swanand,

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


Regards,

Munesh


0 Kudos
Reply