Skip to content

[core] Snapshot expiration deletes data and changelog files in batches #9409

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.

Sign up for GitHub

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

Merged
yuzelin merged 10 commits into from
Aug 26, 2026
Merged

[core] Snapshot expiration deletes data and changelog files in batches #9409

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
b8fd672
[core] Snapshot expiration deletes data and changelog files in batches
Aug 26, 2026
cc334f2
fix
Aug 26, 2026
3ca61cd
fix
Aug 26, 2026
cf2648d
fix
Aug 26, 2026
b6fe8e5
fix
Aug 26, 2026
ee436d7
fix
Aug 26, 2026
0615b14
fix
Aug 26, 2026
4f1555b
fix
Aug 26, 2026
5605233
fix
Aug 26, 2026
ba1db80
fix
Aug 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 0 additions & 5 deletions paimon-core/src/main/java/org/apache/paimon/operation/FileDeletionBase.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,7 +82,6 @@ public abstract class FileDeletionBase<T extends Snapshot> {
protected final Map<BinaryRow, Set<Integer>> deletionBuckets;

private final Executor fileExecutor;
private final int fileOperationParallelism;
@Nullable private final Integer manifestReadParallelism;

protected boolean changelogDecoupled;
Expand Down Expand Up @@ -112,10 +111,6 @@ public FileDeletionBase(
this.cleanEmptyDirectories = cleanEmptyDirectories;
this.deletionBuckets = new ConcurrentHashMap<>();
this.fileExecutor = FileOperationThreadPool.getExecutorService(fileOperationThreadNum);
this.fileOperationParallelism =
fileOperationThreadNum > 0
? fileOperationThreadNum
: Runtime.getRuntime().availableProcessors();
this.manifestReadParallelism = manifestReadParallelism;
}

Expand Down
3 changes: 2 additions & 1 deletion paimon-core/src/main/java/org/apache/paimon/table/AbstractFileStoreTable.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -450,7 +450,8 @@ public ExpireSnapshots newExpireSnapshots() {
snapshotManager(),
changelogManager(),
store().newSnapshotDeletion(),
store().newTagManager());
store().newTagManager(),
store().options().scanManifestParallelism());
}

@Override
Expand Down
41 changes: 32 additions & 9 deletions paimon-core/src/main/java/org/apache/paimon/table/ExpireSnapshotsImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,8 @@
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import javax.annotation.Nullable;

import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.UncheckedIOException;
Expand Down Expand Up @@ -68,6 +70,7 @@ public class ExpireSnapshotsImpl implements ExpireSnapshots {
private final SnapshotDeletion snapshotDeletion;
private final Executor fileExecutor;
private final TagManager tagManager;
private final int snapshotExpireBatchSize;

private ExpireConfig expireConfig;
private Supplier<Long> currentTimeMillis = System::currentTimeMillis;
Expand All @@ -76,7 +79,8 @@ public ExpireSnapshotsImpl(
SnapshotManager snapshotManager,
ChangelogManager changelogManager,
SnapshotDeletion snapshotDeletion,
TagManager tagManager) {
TagManager tagManager,
@Nullable Integer scanManifestParallelism) {
this.snapshotManager = snapshotManager;
this.changelogManager = changelogManager;
this.consumerManager =
Expand All @@ -88,6 +92,10 @@ public ExpireSnapshotsImpl(
this.tagManager = tagManager;
this.expireConfig = ExpireConfig.builder().build();
this.fileExecutor = snapshotDeletion.fileExecutor();
this.snapshotExpireBatchSize =
scanManifestParallelism == null || scanManifestParallelism <= 0
? Runtime.getRuntime().availableProcessors()
: scanManifestParallelism;
}

@VisibleForTesting
Expand Down Expand Up @@ -207,12 +215,11 @@ private int innerExpireUntil(long earliestId, long endExclusiveId)
// delete merge tree files
// deleted merge tree files in a snapshot are not used by the next snapshot, so the range of
// id should be (beginInclusiveId, endExclusiveId]
snapshotDeletion.cleanDataFiles(
collectDataFilesToDelete(snapshotsIncludingEnd, taggedSnapshots, beginInclusiveId));
cleanDataFiles(snapshotsIncludingEnd, taggedSnapshots, beginInclusiveId);

// delete changelog files
if (!expireConfig.isChangelogDecoupled()) {
snapshotDeletion.cleanDataFiles(collectChangelogFilesToDelete(snapshotsExcludingEnd));
cleanChangelogFiles(snapshotsExcludingEnd);
}

// data files and changelog files in bucket directories has been deleted
Expand Down Expand Up @@ -262,7 +269,7 @@ private int innerExpireUntil(long earliestId, long endExclusiveId)
return snapshotsExcludingEnd.size();
}

private Collection<Path> collectDataFilesToDelete(
private void cleanDataFiles(
List<Snapshot> snapshotsIncludingEnd,
List<Snapshot> taggedSnapshots,
long beginInclusiveId)
Expand All @@ -288,6 +295,7 @@ private Collection<Path> collectDataFilesToDelete(
collectTagSkippers(tags.values());
Predicate<ExpireFileEntry> deleteAll = entry -> false;
List<CompletableFuture<List<Path>>> futures = new ArrayList<>();
int plannedSnapshots = 0;
for (Snapshot snapshot : snapshotsIncludingEnd) {
long id = snapshot.id();
if (id == beginInclusiveId) {
Expand All @@ -308,15 +316,18 @@ private Collection<Path> collectDataFilesToDelete(
id);
continue;
}

futures.add(
CompletableFuture.supplyAsync(
() ->
snapshotDeletion.planDeletedInDeltaManifest(
snapshot, skipper.get()),
fileExecutor));
if (++plannedSnapshots >= snapshotExpireBatchSize) {
cleanBatch(futures);
plannedSnapshots = 0;
}
}
return flatten(getAll(futures));
cleanBatch(futures);
}

private Map<Long, Optional<Predicate<ExpireFileEntry>>> collectTagSkippers(
Expand Down Expand Up @@ -350,9 +361,10 @@ private Map<Long, Optional<Predicate<ExpireFileEntry>>> collectTagSkippers(
return skippers;
}

private Collection<Path> collectChangelogFilesToDelete(List<Snapshot> snapshots)
private void cleanChangelogFiles(List<Snapshot> snapshots)
throws ExecutionException, InterruptedException {
List<CompletableFuture<List<Path>>> futures = new ArrayList<>();
int plannedSnapshots = 0;
for (Snapshot snapshot : snapshots) {
if (LOG.isDebugEnabled()) {
LOG.debug("Ready to delete changelog files from snapshot #{}", snapshot.id());
Expand All @@ -362,9 +374,20 @@ private Collection<Path> collectChangelogFilesToDelete(List<Snapshot> snapshots)
CompletableFuture.supplyAsync(
() -> snapshotDeletion.planAddedInChangelogManifest(snapshot),
fileExecutor));
if (++plannedSnapshots >= snapshotExpireBatchSize) {
cleanBatch(futures);
plannedSnapshots = 0;
}
}
}
return flatten(getAll(futures));
cleanBatch(futures);
}

private void cleanBatch(List<CompletableFuture<List<Path>>> futures)
throws ExecutionException, InterruptedException {
Collection<Path> paths = flatten(getAll(futures));
futures.clear();
snapshotDeletion.cleanDataFiles(paths);
}

private Collection<Runnable> collectManifestDeletionTasks(
Expand Down
6 changes: 4 additions & 2 deletions paimon-core/src/test/java/org/apache/paimon/TestFileStore.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,7 +175,8 @@ public ExpireSnapshots newExpire(int numRetainedMin, int numRetainedMax, long mi
snapshotManager(),
changelogManager(),
newSnapshotDeletion(),
new TagManager(fileIO, options.path()))
new TagManager(fileIO, options.path()),
null)
.config(
ExpireConfig.builder()
.snapshotRetainMax(numRetainedMax)
Expand All @@ -189,7 +190,8 @@ public ExpireSnapshots newExpire(ExpireConfig expireConfig) {
snapshotManager(),
changelogManager(),
newSnapshotDeletion(),
new TagManager(fileIO, options.path()))
new TagManager(fileIO, options.path()),
null)
.config(expireConfig);
}

Expand Down
12 changes: 9 additions & 3 deletions paimon-core/src/test/java/org/apache/paimon/operation/ExpireSnapshotsTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -635,7 +635,8 @@ public void testExpireCollectsSnapshotsConcurrently() throws Exception {
blockingSnapshotManager,
changelogManager,
store.newSnapshotDeletion(),
store.newTagManager());
store.newTagManager(),
store.options().scanManifestParallelism());

expire.expireUntil(1, latestSnapshotId);

Expand Down Expand Up @@ -961,7 +962,8 @@ public void testExpireWithTimeDoesNotReadProtectedRange() throws Exception {
failingSnapshotManager,
changelogManager,
store.newSnapshotDeletion(),
store.newTagManager());
store.newTagManager(),
store.options().scanManifestParallelism());
expire.config(config);
expire.setCurrentTimeMillis(() -> 6000L);

Expand Down Expand Up @@ -1217,7 +1219,11 @@ private ExpireSnapshotsImpl newExpireWithSnapshotDeletion(
SnapshotManager snapshotManager,
SnapshotDeletion snapshotDeletion) {
return new ExpireSnapshotsImpl(
snapshotManager, store.changelogManager(), snapshotDeletion, store.newTagManager());
snapshotManager,
store.changelogManager(),
snapshotDeletion,
store.newTagManager(),
store.options().scanManifestParallelism());
}

private void rewriteSnapshotTime(long snapshotId, long newTimeMillis) throws IOException {
Expand Down
18 changes: 15 additions & 3 deletions paimon-core/src/test/java/org/apache/paimon/operation/FileDeletionTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -682,7 +682,11 @@ public void testExpireWithDeletingTags() throws Exception {
// result: exist A & B (because of tag2)
ExpireSnapshots expireSnapshots =
new ExpireSnapshotsImpl(
snapshotManager, changelogManager, store.newSnapshotDeletion(), tagManager);
snapshotManager,
changelogManager,
store.newSnapshotDeletion(),
tagManager,
store.options().scanManifestParallelism());
expireSnapshots
.config(
ExpireConfig.builder()
Expand Down Expand Up @@ -755,7 +759,11 @@ public void testDataFileSkippingSetException() throws Exception {

ExpireSnapshots expireSnapshots =
new ExpireSnapshotsImpl(
snapshotManager, changelogManager, snapshotDeletion, tagManager);
snapshotManager,
changelogManager,
snapshotDeletion,
tagManager,
store.options().scanManifestParallelism());
snapshotDeletion.readMergedDataFilesThrowException = true;
expireSnapshots
.config(
Expand Down Expand Up @@ -820,7 +828,11 @@ public void testManifestFileSkippingSetException() throws Exception {
store.options().scanManifestParallelism());
ExpireSnapshots expireSnapshots =
new ExpireSnapshotsImpl(
snapshotManager, changelogManager, snapshotDeletion, tagManager);
snapshotManager,
changelogManager,
snapshotDeletion,
tagManager,
store.options().scanManifestParallelism());
snapshotDeletion.manifestSkippingSetThrowException = true;
expireSnapshots
.config(
Expand Down
Loading