Add BackendAttribute for parallel model instance loading #235
Conversation
This file contains 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
…ing API to opt-in
…lly or concurrently
…lly or concurrently
rmccorm4
commented
Jul 21, 2023
GuanLuo
reviewed
Jul 24, 2023
…operator to avoid interweaving when multi-threaded
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Outdated suggestions cannot be applied.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.
Current default value of the attribute is false, so no parallel loading will be done unless a backend implements and calls the API to enable it. This is just the scaffolding for such capabilities.
Identity Backend opt-in for sanity check in existing pipelines: triton-inference-server/identity_backend#26
Seeing ballpark 2-3x speedup on my machine with Identity Backend loading 100 instances. I think the speedup will be more meaningful for model instances that take longer to initialize in more complicated backends. Will do more thorough performance analysis after enabling support in other backends.