Skip to content

[lake/paimon] Avoid repeated manifest scans for historical tiering #4731

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

Open
beryllw wants to merge 4 commits into
base: main
Choose a base branch
from
Open

[lake/paimon] Avoid repeated manifest scans for historical tiering #4731

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
4b83e9e
[lake/paimon] Optimize historical split initialization (#4656)
beryllw Oct 19, 2026
752df79
[test] Split historical Paimon tiering tests
beryllw Oct 19, 2026
a5ce5df
[test] Verify historical tiering scan counts without mocks
beryllw Oct 19, 2026
1b16131
[docs] Clarify Paimon version for historical tiering optimization
beryllw Oct 19, 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
11 changes: 11 additions & 0 deletions ...fluss-lake-paimon/src/main/java/org/apache/fluss/lake/paimon/utils/PaimonConversions.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,8 @@
import org.apache.fluss.annotation.VisibleForTesting;
import org.apache.fluss.config.AutoPartitionTimeUnit;
import org.apache.fluss.config.ConfigOptions;
import org.apache.fluss.config.Configuration;
import org.apache.fluss.config.TableConfig;
import org.apache.fluss.exception.InvalidConfigException;
import org.apache.fluss.exception.InvalidTableException;
import org.apache.fluss.lake.paimon.source.FlussRowAsPaimonRow;
Expand Down Expand Up @@ -397,6 +399,15 @@ public static Schema toPaimonSchema(TableDescriptor tableDescriptor) {
options.set(
CoreOptions.CHANGELOG_PRODUCER.key(),
CoreOptions.ChangelogProducer.INPUT.toString());
TableConfig tableConfig =
new TableConfig(Configuration.fromMap(tableDescriptor.getProperties()));
if (tableConfig.isHistoricalPartitionEnabled()
&& !tableConfig.isDataLakeAutoCompaction()) {
// Use the snapshot's maximum sequence number to avoid repeated manifest scans
// when historical splits write to multiple partitions and buckets.
// Use a literal key because Paimon 1.4 does not define this option.
options.set("write.sequence-number-init-mode", "snapshot");
}
}

// set partition keys
Expand Down
75 changes: 63 additions & 12 deletions ...e/fluss-lake-paimon/src/test/java/org/apache/fluss/lake/paimon/PaimonLakeCatalogTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,19 +44,26 @@
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;
import org.junit.jupiter.api.io.TempDir;
import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.CsvSource;

import javax.annotation.Nullable;

import java.io.File;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.Optional;

import static org.apache.fluss.config.ConfigOptions.TABLE_DATALAKE_AUTO_COMPACTION;
import static org.apache.fluss.config.ConfigOptions.TABLE_DATALAKE_ENABLED;
import static org.apache.fluss.config.ConfigOptions.TABLE_DATALAKE_FORMAT;
import static org.apache.fluss.config.ConfigOptions.TABLE_DATALAKE_HISTORICAL_PARTITION_ENABLED;
import static org.apache.fluss.lake.paimon.utils.PaimonConversions.LAKESTREAM_ENABLED_OPTION_KEY;
import static org.apache.fluss.lake.paimon.utils.PaimonConversions.PARTITION_GENERATE_LEGACY_NAME_OPTION_KEY;
import static org.apache.fluss.lake.paimon.utils.PaimonConversions.toPaimon;
import static org.apache.fluss.lake.paimon.utils.PaimonTableValidation.isPaimonSchemaCompatible;
import static org.apache.paimon.CoreOptions.WRITE_SEQUENCE_NUMBER_INIT_MODE;
import static org.assertj.core.api.Assertions.assertThat;
import static org.assertj.core.api.Assertions.assertThatThrownBy;

Expand Down Expand Up @@ -836,20 +843,64 @@ void testAlterTableAddColumnWhenPaimonSchemaNotMatch() throws Exception {
changes));
}

@Test
void testCreateTableSetsLakeStreamEnabledForCleanTable() throws Exception {
String database = "test_create_lakestream_db";
String tableName = "test_create_lakestream_table";
TablePath tablePath = TablePath.of(database, tableName);
Identifier identifier = Identifier.create(database, tableName);

// getTableDescriptor sets table.datalake.enabled=true, so the clean table advertises its
// LakeStream state to Paimon
flussPaimonCatalog.createTable(
tablePath, getTableDescriptor(FLUSS_SCHEMA), LAKE_CATALOG_CONTEXT);
@ParameterizedTest
@CsvSource({
"true,true,,,snapshot",
"true,true,false,,snapshot",
"true,true,true,,",
"true,false,false,,",
"false,true,false,,",
"false,true,true,,",
"true,true,false,scan,scan",
"true,true,true,snapshot,snapshot"
})
void testCreateTableOptions(
boolean primaryKey,
boolean historicalPartition,
@Nullable Boolean autoCompaction,
@Nullable String sequenceMode,
@Nullable String expectedSequenceMode)
throws Exception {
TablePath tablePath = TablePath.of("test_db", "test_create_options");

Schema.Builder schema = Schema.newBuilder().fromSchema(FLUSS_SCHEMA);
if (primaryKey) {
schema.primaryKey("id", "address");
}
TableDescriptor.Builder builder =
TableDescriptor.builder()
.schema(schema.build())
.partitionedBy("address")
.distributedBy(3)
.property(TABLE_DATALAKE_ENABLED, true)
.property(TABLE_DATALAKE_FORMAT.key(), "paimon")
.property(TABLE_DATALAKE_HISTORICAL_PARTITION_ENABLED, historicalPartition);
if (autoCompaction != null) {
builder.property(TABLE_DATALAKE_AUTO_COMPACTION, autoCompaction);
}
String option = "paimon." + WRITE_SEQUENCE_NUMBER_INIT_MODE.key();
if (sequenceMode != null) {
builder.customProperty(option, sequenceMode);
}
TableDescriptor descriptor = builder.build();
TestingLakeCatalogContext context = new TestingLakeCatalogContext(descriptor);
flussPaimonCatalog.createTable(tablePath, descriptor, context);

Table table = flussPaimonCatalog.getPaimonCatalog().getTable(identifier);
Table table = flussPaimonCatalog.getPaimonCatalog().getTable(toPaimon(tablePath));
assertThat(table.options()).containsEntry(LAKESTREAM_ENABLED_OPTION_KEY, "true");
assertThat(table.options().get(WRITE_SEQUENCE_NUMBER_INIT_MODE.key()))
.isEqualTo(expectedSequenceMode);

if ("snapshot".equals(expectedSequenceMode)) {
flussPaimonCatalog.alterTable(
tablePath, Collections.singletonList(TableChange.set(option, "scan")), context);
assertThat(
flussPaimonCatalog
.getPaimonCatalog()
.getTable(toPaimon(tablePath))
.options())
.containsEntry(WRITE_SEQUENCE_NUMBER_INIT_MODE.key(), "scan");
}
}

@Test
Expand Down
Loading
Loading