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

Add routing for model adapter #183

Merged
merged 7 commits into from
Sep 19, 2024
Merged

Add routing for model adapter #183

merged 7 commits into from
Sep 19, 2024

Conversation

varungup90
Copy link
Collaborator

@varungup90 varungup90 commented Sep 16, 2024

@Jeffwan
Copy link
Collaborator

Jeffwan commented Sep 17, 2024

Do we need to create HTTPRoute for lora?

@varungup90
Copy link
Collaborator Author

varungup90 commented Sep 17, 2024

Do we need to create HTTPRoute for lora?

We can not create HTTProute for lora since HTTProute needs a backend ref which is a service. Given that, I have switched to always using targetpod for every query and pods are fetched from internal cache to save time. Going forward we do not need to create httproute, I will remove that in next PR.

Edit: added a TODO in model router controller.

@Jeffwan
Copy link
Collaborator

Jeffwan commented Sep 17, 2024

@varungup90 We do create endpoint and service for each model adapter. The purpose is to make sure all model entrypoint is service. The difference part is it use headless service in that case. Does HTTPRoute works with such setting?

@varungup90
Copy link
Collaborator Author

@varungup90 We do create endpoint and service for each model adapter. The purpose is to make sure all model entrypoint is service. The difference part is it use headless service in that case. Does HTTPRoute works with such setting?

updated to add/delete httproute based on model adapter.

@Jeffwan
Copy link
Collaborator

Jeffwan commented Sep 19, 2024

based on model adapter.

I didn't see update in model adapter controller? Did you push the commits?

@Jeffwan
Copy link
Collaborator

Jeffwan commented Sep 19, 2024

It seems HTTPRoute is still managed in model router. It's ok at this moment. The downside is user have to use some labels which could be encapsulated inside the controller by us. Let's make necessary refactor later.

Copy link
Collaborator

@Jeffwan Jeffwan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

/lgtm

@Jeffwan Jeffwan merged commit 1bc20e1 into main Sep 19, 2024
4 checks passed
@Jeffwan Jeffwan deleted the model-adapter-routing branch September 19, 2024 00:14
gangmuk pushed a commit that referenced this pull request Jan 25, 2025
* Add routing for model adapter

* nit: logging

* nit: gateway error response code refactoring

* code review comments

* add/delete httproute for model adapter

* nit

---------

Co-authored-by: varungupta <[email protected]>
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

Successfully merging this pull request may close these issues.

2 participants