[Store] Add parallel bucket writes for SSD offload - #3226
Open
NUABO wants to merge 1 commit into
Open
Conversation
…HREADS
Add optional thread pool that dispatches per-bucket BatchOffload
calls concurrently within a single heartbeat tick. Each bucket's
D2H staging, eviction, WriteBucket (disk I/O), and master RPC run
independently — buckets write to different files so the heavy disk
writes are truly parallel.
MC_OFFLOAD_WRITE_THREADS (default 0, disabled)
• 0: original sequential per-bucket loop, zero change.
• >1: thread pool of that size; when multiple buckets are
returned by AllocateOffloadingBuckets, each bucket is
dispatched to a worker. The pool is created once in
FileStorage::Init() and reused across heartbeat ticks.
Parallel-path error handling mirrors the sequential path:
• KEYS_ULTRA_LIMIT or other non-INVALID_READ errors are
captured and propagated after all futures complete.
• INVALID_READ errors add the affected keys to failed_tasks
(mutex-protected) so the master can release refcounts.
• Keys skipped by GroupOffloadingKeysByBucket are also
reported as failed.
Signed-off-by: tan changzhi <544463199@qq.com>
|
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Add optional thread pool that dispatches per-bucket BatchOffload calls concurrently within a single heartbeat tick. Each bucket's D2H staging, eviction, WriteBucket (disk I/O), and master RPC run independently — buckets write to different files so the heavy disk writes are truly parallel.
MC_OFFLOAD_WRITE_THREADS (default 0, disabled)
• 0: original sequential per-bucket loop, zero change.
• >1: thread pool of that size; when multiple buckets are
returned by AllocateOffloadingBuckets, each bucket is
dispatched to a worker. The pool is created once in
FileStorage::Init() and reused across heartbeat ticks.
Parallel-path error handling mirrors the sequential path:
• KEYS_ULTRA_LIMIT or other non-INVALID_READ errors are
captured and propagated after all futures complete.
• INVALID_READ errors add the affected keys to failed_tasks
(mutex-protected) so the master can release refcounts.
• Keys skipped by GroupOffloadingKeysByBucket are also
reported as failed.
Module
mooncake-transfer-engine)mooncake-store)mooncake-ep)mooncake-pg)mooncake-integration)mooncake-p2p-store)mooncake-wheel)mooncake-common)mooncake-rl)Type of Change
How Has This Been Tested?
Test commands:
# Example: bash scripts/run_ci_test.shTest results:
Checklist
./scripts/code_format.shpre-commit run --all-filesand all hooks passAI Assistance Disclosure