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

Performance scaling with multiple sticks

idata
Employee
‎09-21-2017 07:49 AM
835 Views

How does performance scale with multiple sticks? Suppose I have two sticks.

 

     

  • Can I use -s24 during compilation, assuming two sticks at deployment? (0.5x latency at 2x framerate, distribute inference over 24 SHAVEs)

    •  

  • Or is this just a matter of sending half the images on each stick? (1x latency at 2x framerate, distribute inference calls to 2 VPUs)

     

    I'd guess it is the latter, but I would like confirmation. Thanks.

    •  

0 Kudos

Link Copied

1 Reply
idata
Employee
‎09-22-2017 05:44 PM
452 Views

@pashmina.bendale Your second guess is correct. You can send images to each NCS to distribute the workload. You may find more information by referring to this thread: https://ncsforum.movidius.com/discussion/112/chaining-movidius-sticks#latest

0 Kudos
Copy link
Reply

Community support is provided during standard business hours (Monday to Friday 7AM - 5PM PST). Other contact methods are available here.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

For more complete information about compiler optimizations, see our Optimization Notice.