Performance Issues Running Python 3

I am trying to run a set of python scripts to read and parse data from our robot. Currently, I have been running into issues with performance. Namely, the script takes much longer to parse and plot log data on the Xavier than it does on my several year old laptop.

We have run performance scripts to try to bump up clock rates, but they didn’t seem to help any. The scripts we ran are:
$ sudo ./jetson_clocks.sh
$ sudo nvpmodel -m 0

Other things that we think may be causing the performance issues:

  • The script uses PyQt5 for the front end
  • There is a fair amount of file IO, but it doesn’t explain why it runs faster on my laptop

Setup:
OS: Ubuntu 18.04
Python 3.6.7

The first line of /etc/nv_tegra_release is:

R31 (release), REVISION: 1.0, GCID: 13194883, BOARD: t186ref, EABI: aarch64, DATE: Wed Oct 31 22:26:16 UTC 2018

Any help would be greatly appreciated!

Hi,

The order of script matters.
Please set nvpmodel first and then execute the jetson_clocks.

$ sudo nvpmodel -m 0
$ sudo ./jetson_clocks.sh

nvpmodel will adjust the max/min frequency based on the input argument.
jetson_clocks will fix the frequency into the max.

So you will need to setup the max frequency to performance mode first.
And then fix the system into max frequency.

Thanks.

Hi,

We did try changing the order of the scripts, but it didn’t seem to make much of a difference.

Is there anything else we could try?

Hi,

Could you turn-off QT to see if any difference?
Thanks.