[core]save the runtime memory while reading all the partitions or buckets - #8745
Merged
Conversation
weijietong
force-pushed
the
metadata_mem_p1
branch
from
July 20, 2026 11:26
0c0cbae to
1e243fe
Compare
- Change ManifestEntryCache and ObjectsCache to use the cache path for storage - Update SimpleObjectsCache with related improvements - Add FileStoreScanPartitionBucketEntryTest for testing
weijietong
force-pushed
the
metadata_mem_p1
branch
from
July 20, 2026 11:28
1e243fe to
827ec2d
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Purpose
Optimize readPartitionEntries and readBucketEntries in AbstractFileStoreScan to use per-entry converter during manifest reading, so ManifestEntry objects (with heavy DataFileMeta including stats, key ranges, embedded index, etc.) are immediately converted to lightweight PartitionEntry/BucketEntry and never accumulated. This reduces peak memory per manifest file from holding a full List to only a List — each DataFileMeta (~20 fields, several KB) is discarded after extracting the 3 needed values (rowCount, fileSize, creationTime), while PartitionEntry only retains ~6 fields (~40 bytes). For tables with millions of files the savings can be significant.
Tests
Added FileStoreScanPartitionBucketEntryTest covering single/multiple partition aggregation, multi-file merge within the same partition, partition filter pushdown, append-only table variant, and equivalent bucket entry tests — including the previously untested readBucketEntries() method which had zero coverage.