Enable parallel instance loading backend attribute #284
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Tabrizian
approved these changes
Aug 1, 2023
krishung5
approved these changes
Aug 1, 2023
|
Going to merge for the sake of getting more pipelines running before code freeze to catch any unforeseen issues as early as possible. Tests may have some minor changes, as discussed above, I don't anticipate any python backend specific changes other than enabling the attribute. If we run into issues before the release, I can disable the attribute. I'll update some docs separately where applicable. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As far as I can tell, the Python Backend is completely thread-safe on calls to TRITONBACKEND_ModelInstanceInitialize
All use of
model_state(the problem area with instance initialization in current backends) appears to be read-only.It is passed to constructor for ModelInstanceState very simply.
L0_backend_python has been hanging on BLS model load tests, but this is now happening on
mainbranches too, so I don't believe this is due to any parallel instance changes.Corresponding tests: triton-inference-server/server#6126