triton-inference-server / server

The Triton Inference Server provides an optimized cloud and edge inferencing solution.
https://docs.nvidia.com/deeplearning/triton-inference-server/user-guide/docs/index.html
BSD 3-Clause "New" or "Revised" License
8.39k stars 1.49k forks source link

docs: Load new model version should not reload loaded existing model version(s) #7537

Closed kthui closed 3 months ago

kthui commented 3 months ago

What does the PR do?

Document any loaded and unmodified model version(s) will not be reloaded upon changes to other model versions.

Checklist

Commit Type:

Check the conventional commit type box here and add the label to the github PR.

Related PRs:

https://github.com/triton-inference-server/server/pull/7527 https://github.com/triton-inference-server/core/pull/388

Where should the reviewer start?

Start with the core PR.

Test plan:

This is a documentation change, so any testing will be covered by related PRs linked.

Caveats:

N/A

Background

N/A

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

N/A

kthui commented 3 months ago

Discussed offline that we are ok with not having a docs for this change, because not reloading an unchanged model version is expected and this change can be mentioned in the release note, so closing this PR.