Add new API ListenForPolicyViolations to replace Policy #54
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
nikkon-dev
reviewed
Jan 8, 2024
dran-dev
force-pushed
the
main
branch
2 times, most recently
from
8587c10 to
84f6454
Compare
This adds a new api ListenForPolicyViolations for setting policy for all the gpus instead of individual gpus. The primary modifications include the deprecation of the existing Policy API and the introduction of the ListenForPolicyViolations API. Policy API is deprecated due to usability constraints. Moreover listening to policy violations for one gpu at a time is not very useful. The new API enables users to set policies for all GPUs collectively, eliminating the need to configure individual GPUs separately. Additionally, ListenForPolicyViolations API allows users to register and monitor policy violations across all GPUs concurrently. Policy callbacks are required to be registered only once during the lifetime of the program. Signed-off-by: dran <[email protected]>
Signed-off-by: dran <[email protected]>
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Outdated suggestions cannot be applied.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.
Summary of Changes
This pull request introduces enhancements to the Policy API in the go-dcgm bindings for NVIDIA Data Center GPU Manager (DCGM) library. The primary modifications include the deprecation of the existing Policy API and the introduction of the ListenForPolicyViolations API. The new API enables users to set policies for all GPUs collectively, eliminating the need to configure individual GPUs separately. Additionally, the ListenForPolicyViolations API allows users to register and monitor policy violations across all GPUs concurrently, addressing usability constraints and making the API more efficient.
Context and Rationale
The decision to deprecate the Policy API and introduce ListenForPolicyViolations stems from usability constraints and the recognition that monitoring policy violations for individual GPUs at a time may not be useful in most scenarios. The changes aim to improve the overall usability and efficiency of policy callback registration with the DCGM library.
Deprecation Notice
Developers are advised to migrate from the deprecated Policy API to the new ListenForPolicyViolations API for improved functionality and to ensure compatibility with future releases.