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

[feature request] Support different serving servers #14

Open
gaocegege opened this issue Apr 11, 2019 · 3 comments

Comments

Projects
None yet
2 participants
@gaocegege
Copy link
Member

commented Apr 11, 2019

Now there are some inference servers such as TensorRT Inference Server, GraphPipe, TensorFlow Serving, and so on. Different may want to use different servers. Thus I think we should support different servers.

BTW, some servers support serving multiples models and multiple frameworks. For example, Graphpipe supports TensorFlow, PyTorch and Caffe. Maybe we also should investigate how to support co-serving multiple models in one serving CRD.

@issue-label-bot

This comment has been minimized.

Copy link

commented Apr 11, 2019

Issue-Label Bot is automatically applying the label feature_request to this issue, with a confidence of 0.94. Please mark this comment with 👍 or 👎 to give our bot feedback!

Links: dashboard, app homepage and code for this bot.

@gaocegege

This comment has been minimized.

Copy link
Member Author

commented Apr 11, 2019

/kind feature

@ellis-bigelow

This comment has been minimized.

Copy link
Collaborator

commented Apr 11, 2019

Thanks for the feature! Given our data model (high level "Tensorflow" spec), do you think it would make most sense to swap different Tensorflow technologies with annotations?

Regarding:
"Maybe we also should investigate how to support co-serving multiple models in one serving CRD.".

I think this is an anti-pattern in KFServing. A KFServing Service is a "unit" of model serving. The idea of hosting multiple models in a single model server is extremely interesting and something I've been working on in my share time. I've been thinking we should control this via implementation, not interface, and allow users to specify an annotation like "enable-multitenancy".

I'll make a separate issue for this since I think it's a huge topic.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.