Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dora performance for big vs small bandwidth data #771

Open
RobinSchmid7 opened this issue Feb 5, 2025 · 4 comments
Open

Dora performance for big vs small bandwidth data #771

RobinSchmid7 opened this issue Feb 5, 2025 · 4 comments

Comments

@RobinSchmid7
Copy link

I see that dora outperforms ROS2 for high bandwidth topics a lot, what is the performance for small bandwidth high frequency topics? We see that ROS2 shows weak performance on these areas too.

@haixuanTao
Copy link
Collaborator

Hey Robin, so, dora-rs currently is not optimised to what it could potentially be for small bandwidth. It should be okay but would not be able to handle >500Hz like frequency low bandwidth unfortunately.

There's a lot we can improve but currently do not have the human bandwidth. 😓

@RobinSchmid7
Copy link
Author

Hey Haixuan, thanks a lot, what you have currently is amazing! What is the expected speedup for ~100Hz low bandwidth data?

@haixuanTao
Copy link
Collaborator

Thanks! So overall, I'm not super happy with our current bench which should be similar to this: #760

Basically, we have around 1ms of fixed sending<->receiving latency, which is quite big, and I wish I could cut by around half this fixed cost. ( Note that this might fluctuate depending on your CPU performance )

I would think that for 100Hz data you would typically be around the same ballpark with ROS2 everything included, meaning ROS2 protobuf serialization/deserialization included.

@RobinSchmid7
Copy link
Author

Got it, I think zmq has potentially some useful implementations for this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants