enhance: parallelize retrieve output field materialization - #29
Merged
Conversation
Share field-level scheduling between search and retrieve. Run field reads on the middle-priority pool with independent operation contexts, then publish results in plan order while preserving primary key handling and storage accounting. Cancel sibling reads on failure, preserve the original exception, and drain pending tasks before returning. Reserve result storage before dispatch and add coverage for concurrent reads, cancellation, failure cleanup, and schema-dependent output fields. Signed-off-by: sijie-ni-0214 <sijie.ni@zilliz.com>
sunby
merged commit Sep 10, 2026
a63fb2d
into
sunby:codex/load-1m-segments-pr-stack-rebased-qv-work
6 of 8 checks passed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The regular Query/Retrieve output-field path reads fields sequentially within each segment, including retrieval by offsets. Fetch these fields concurrently on the middle-priority pool through a helper shared with Search, while segment parent tasks remain on the search executor.
Use independent per-field operation contexts and I/O counters, then publish results in plan order while preserving PK-only retrieval, system fields, dynamic projection, and missing-field defaults. Preserve the original field error, cancel sibling reads, and drain submitted tasks before returning. Reserve result storage before dispatching tasks.
Validation:
make build-cpp-with-unittest SKIP_3RDPARTY=1passed; the C++ source tree is identical before and after the rebase.make build-gopassed after the rebase.git diff --checkpassed.