-
Notifications
You must be signed in to change notification settings - Fork 737
feat: vllm scheduler (dis-630) #3061
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Draft
ryanolson
wants to merge
9
commits into
main
Choose a base branch
from
ryan/dis-630-vllm-scheduler
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Draft
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
…utput Signed-off-by: Ryan Olson <[email protected]>
Signed-off-by: Ryan Olson <[email protected]>
…es it; no actual scheduling logic Signed-off-by: Ryan Olson <[email protected]>
Signed-off-by: Ryan Olson <[email protected]>
…Pool to InactivePool - Add TinyLFU frequency sketch with Count-Min Sketch implementation - Create FrequencyTracker trait with TinyLFUTracker and NoOpTracker implementations - Add optional frequency tracking to BlockRegistry with touch() and count() methods - Implement transfer_registration() for internal pool transfers without tracking - Rename RegisteredPool to InactivePool to better reflect dormant block semantics - Update all references across manager, registry, and test utilities All 41 tests passing. Phase 1-3 complete for block manager v2 refactoring.
…ager Replace constructor-based approach with fluent builder API: - Add BlockManagerConfigBuilder with backend selection enum - Support HashMap, LRU, and MultiLRU backend configurations - Automatic LRU capacity sizing (equals block_count to prevent eviction) - Custom MultiLRU frequency thresholds with validation - Frequency tracker size validation (2^18 to 2^24, power of 2) - Comprehensive test suite for all builder configurations Provides clean API: BlockManager::builder().block_count(100).with_lru_backend().build() Resolves ergonomic issues with previous new() and new_with_multi_lru() constructors. Signed-off-by: Ryan Olson <[email protected]>
- Replace Mutex with parking_lot::RwLock in InactivePool and BlockRegistry for improved concurrent read performance - Add #[inline] annotations to hot path methods including all drop implementations and registry lookups - Pre-allocate vectors in bulk operations (match_blocks, allocate_blocks) - Add comprehensive Criterion benchmarks for registration, lookup, drop, and allocation operations - Fix clippy warnings: remove unused imports, add Default implementations, use modern Rust patterns (is_multiple_of, Result matching) These optimizations address high variance in benchmarks and improve contention in multi-threaded scenarios by allowing concurrent reads while maintaining write safety.
|
This PR is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days. |
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.
No description provided.