-
Notifications
You must be signed in to change notification settings - Fork 3.1k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Populate a nulls fraction when NDV is known in TableScanStatsRule
If a connector provides NDV but is missing nulls fraction statistic for a column (e.g. Delta Lake after "delta.dataSkippingNumIndexedCols" columns and MySql), populate a 10% guess value so that the CBO can still produce some estimates rather than failing to make any estimates due to lack of nulls fraction.
- Loading branch information
1 parent
b362da8
commit c2666d7
Showing
6 changed files
with
206 additions
and
40 deletions.
There are no files selected for viewing
This file contains 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
This file contains 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
122 changes: 122 additions & 0 deletions
122
core/trino-main/src/test/java/io/trino/cost/TestTableScanStatsRule.java
This file contains 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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,122 @@ | ||
| /* | ||
| * Licensed under the Apache License, Version 2.0 (the "License"); | ||
| * you may not use this file except in compliance with the License. | ||
| * You may obtain a copy of the License at | ||
| * | ||
| * http://www.apache.org/licenses/LICENSE-2.0 | ||
| * | ||
| * Unless required by applicable law or agreed to in writing, software | ||
| * distributed under the License is distributed on an "AS IS" BASIS, | ||
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
| * See the License for the specific language governing permissions and | ||
| * limitations under the License. | ||
| */ | ||
| package io.trino.cost; | ||
|
|
||
| import com.google.common.collect.ImmutableList; | ||
| import com.google.common.collect.ImmutableMap; | ||
| import io.trino.spi.connector.ColumnHandle; | ||
| import io.trino.spi.connector.TestingColumnHandle; | ||
| import io.trino.spi.statistics.ColumnStatistics; | ||
| import io.trino.spi.statistics.DoubleRange; | ||
| import io.trino.spi.statistics.Estimate; | ||
| import io.trino.spi.statistics.TableStatistics; | ||
| import io.trino.sql.planner.Symbol; | ||
| import org.testng.annotations.Test; | ||
|
|
||
| import static io.trino.spi.type.BigintType.BIGINT; | ||
| import static io.trino.spi.type.DoubleType.DOUBLE; | ||
| import static io.trino.spi.type.IntegerType.INTEGER; | ||
|
|
||
| public class TestTableScanStatsRule | ||
| extends BaseStatsCalculatorTest | ||
| { | ||
| @Test | ||
| public void testStatsForTableScan() | ||
| { | ||
| ColumnHandle columnA = new TestingColumnHandle("a"); | ||
| ColumnHandle columnB = new TestingColumnHandle("b"); | ||
| ColumnHandle columnC = new TestingColumnHandle("c"); | ||
| ColumnHandle columnD = new TestingColumnHandle("d"); | ||
| ColumnHandle columnE = new TestingColumnHandle("e"); | ||
| ColumnHandle unknownColumn = new TestingColumnHandle("unknown"); | ||
| tester() | ||
| .assertStatsFor(pb -> { | ||
| Symbol a = pb.symbol("a", BIGINT); | ||
| Symbol b = pb.symbol("b", DOUBLE); | ||
| Symbol c = pb.symbol("c", DOUBLE); | ||
| Symbol d = pb.symbol("d", DOUBLE); | ||
| Symbol e = pb.symbol("e", INTEGER); | ||
| Symbol unknown = pb.symbol("unknown", INTEGER); | ||
| return pb.tableScan( | ||
| ImmutableList.of(a, b, c, d, e, unknown), | ||
| ImmutableMap.of(a, columnA, b, columnB, c, columnC, d, columnD, e, columnE, unknown, unknownColumn)); | ||
| }) | ||
| .withTableStatisticsProvider(tableHandle -> TableStatistics.builder() | ||
| .setRowCount(Estimate.of(33)) | ||
| .setColumnStatistics( | ||
| columnA, | ||
| ColumnStatistics.builder().setDistinctValuesCount(Estimate.of(20)).build()) | ||
| .setColumnStatistics( | ||
| columnB, | ||
| ColumnStatistics.builder().setNullsFraction(Estimate.of(0.3)).setDistinctValuesCount(Estimate.of(23.1)).build()) | ||
| .setColumnStatistics( | ||
| columnC, | ||
| ColumnStatistics.builder().setRange(new DoubleRange(15, 20)).build()) | ||
| .setColumnStatistics( | ||
| columnD, | ||
| ColumnStatistics.builder().setDistinctValuesCount(Estimate.of(33)).build()) | ||
| .setColumnStatistics( | ||
| columnE, | ||
| ColumnStatistics.builder().setDistinctValuesCount(Estimate.of(31)).build()) | ||
| .setColumnStatistics(unknownColumn, ColumnStatistics.empty()) | ||
| .build()) | ||
| .check(check -> check | ||
| .outputRowsCount(33) | ||
| .symbolStats("a", assertion -> assertion | ||
| .distinctValuesCount(20) | ||
| // UNKNOWN_NULLS_FRACTION populated to allow CBO to use NDV for estimation | ||
| .nullsFraction(0.1)) | ||
| .symbolStats("b", assertion -> assertion | ||
| .distinctValuesCount(23.1) | ||
| .nullsFraction(0.3)) | ||
| .symbolStats("c", assertion -> assertion | ||
| .distinctValuesCountUnknown() | ||
| .nullsFractionUnknown() | ||
| .lowValue(15) | ||
| .highValue(20)) | ||
| .symbolStats("d", assertion -> assertion | ||
| .distinctValuesCount(33) | ||
| .nullsFraction(0)) | ||
| .symbolStats("e", assertion -> assertion | ||
| .distinctValuesCount(31) | ||
| .nullsFraction(0.06060606)) | ||
| .symbolStats("unknown", assertion -> assertion | ||
| .unknownRange() | ||
| .distinctValuesCountUnknown() | ||
| .nullsFractionUnknown() | ||
| .dataSizeUnknown())); | ||
| } | ||
|
|
||
| @Test | ||
| public void testZeroStatsForTableScan() | ||
| { | ||
| ColumnHandle columnHandle = new TestingColumnHandle("zero"); | ||
| tester() | ||
| .assertStatsFor(pb -> { | ||
| Symbol column = pb.symbol("zero", INTEGER); | ||
| return pb.tableScan( | ||
| ImmutableList.of(column), | ||
| ImmutableMap.of(column, columnHandle)); | ||
| }) | ||
| .withTableStatisticsProvider(tableHandle -> TableStatistics.builder() | ||
| .setRowCount(Estimate.zero()) | ||
| .setColumnStatistics( | ||
| columnHandle, | ||
| ColumnStatistics.builder().setDistinctValuesCount(Estimate.zero()).build()) | ||
| .build()) | ||
| .check(check -> check | ||
| .outputRowsCount(0) | ||
| .symbolStats("zero", assertion -> assertion.isEqualTo(SymbolStatsEstimate.zero()))); | ||
| } | ||
| } |
This file contains 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
This file contains 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
This file contains 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