refactor: implement unified AI backend control plane - #94
Merged
Conversation
This commit implements a full overhaul of the AI backend control plane and related task scheduling system: - Replace legacy remote backend synchronization with the new `BackendAssignmentController` - Restructure task APIs to remove hardcoded single-node binding, add `desired_replicas` and spread scheduling strategy - Add new protobuf definitions for AI backend records, placements, statuses, and task assignments - Set up web console with Tailwind CSS, shared UI components, and unified theme mode - Update documentation, configuration files, and test suites across the codebase - Clean up deprecated legacy modules including model_deployments, managed_backends, and remote_backend_policy - Update local Docker deployment to use RocksDB feature by default for persistent storage
Add new backend support matrix documentation covering current adapter capabilities and UI create flows. Update all Web Admin overview, usage guides, and project READMEs to match the latest control plane and UI implementation. Remove outdated, unused documentation files including coverage setup guides, old code cleanup summaries, legacy registry API design docs, and redundant test fix records.
replace global dynamic backend and AI engine singletons with per-test instances refactor execution status test to use matches! macro for clearer checks fix makefile to properly detect npm, adjust test args, and update web UI commands
Conditionalize the import of global_dynamic_backends in the router module to only be included when not building tests, avoiding test compilation errors. Add a helper function to spin up a local gRPC SMS server for tests, and update the stale update event test to use a real gRPC connection and properly clean up the server handle after completion.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
This commit implements a full overhaul of the AI backend control plane and related task scheduling system:
BackendAssignmentControllerdesired_replicasand spread scheduling strategy