The Autoware Foundation
Performance analysis of a multi-container Autoware deployment #5012
oguzkaganozt
started this conversation in
Design
Replies: 1 comment
-
|
planning_control-2024-08-26-16-35-52-reaction-results.csv |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Motivation
Now that the Open AD Kit containers are being used by different parties on different demos/activities etc. Testing multi-container behaviour and conducting a performance analysis can lead the way through the best possible aproach on how to upsteam multi-container deployment to the Autoware.
Test Suite 1: Reaction analyzer tool + AWSIM
Test Suite 2: Ros Trace Tools
Possible Approach
Phase 1
Utilizing Reaction Analyzer Tool for measuring end-to-end latency of the system with 2 containers (planning & perception) by using pre-recorded rosbags and map.
Phase 2
Replicating TUM's take on modular Autoware's setup on a smaller scale by using 7 modular containers with the help of Reaction Analyzer Tool for measuring end-to-end latency of the system
Definition of done
Performance analysis report comparing bare-metal, Phase-1 and Phase-2 deployment results in a summarized format.
Beta Was this translation helpful? Give feedback.
All reactions