-
Notifications
You must be signed in to change notification settings - Fork 69
InferencePool: Add BBR Config #711
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
Comments
Yeah, it should act independently of the InferencePool (as it's not strictly necessary, and useful more for routing via HTTPRoute rules) @rramkumar1 I believe added some documentation for how this should/could be used in OSS. If not we absolutely should have something |
Perhaps this convo would tie into: kubernetes-sigs/gateway-api#3625 (comment) (except somewhat different, as it would happen before the HTTPRoute) |
@danehans This won't need any config in the InferencePool, it's deployed separately. Below is an example of how it would have been deployed using Istio GW (note InferencePool support there is still experimental). Other LB implementations would do something similar depending on how they allow ext_proc configuration. |
@rramkumar1 I went over this chart. overall looks good and very clear.
do have somewhere a documentation of the full flow using BBR + EPP? |
Sure I'll take an AI to add some more docs on this. /assign |
Currently,
inferencepool.spec
inlinesEndpointPickerConfig
that is defined by the following:The Body Based Routing (BBR) extension was recently added, but InferencePool was not updated. Does BBR act independently of an InferencePool? If so, how does a proxy discover and connect with this extension?
The text was updated successfully, but these errors were encountered: