Skip to content

[SPARK-26138][SQL] Pushdown limit through InnerLike when condition is empty #31567

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

Closed
wants to merge 4 commits into from
Closed

[SPARK-26138][SQL] Pushdown limit through InnerLike when condition is empty #31567

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
a0a0685
SPARK-26138: pushdown limit through InnerLike when condition is empty
wangyum Feb 15, 2021
57171fe
fix comments
wangyum Feb 16, 2021
a9a1955
fix
wangyum Feb 18, 2021
e0915b9
Merge remote-tracking branch 'upstream/master' into SPARK-26138
wangyum Feb 23, 2021
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
13 changes: 9 additions & 4 deletions sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -539,17 +539,22 @@ object LimitPushDown extends Rule[LogicalPlan] {
// pushdown Limit.
case LocalLimit(exp, u: Union) =>
LocalLimit(exp, u.copy(children = u.children.map(maybePushLocalLimit(exp, _))))
// Add extra limits below OUTER JOIN. For LEFT OUTER and RIGHT OUTER JOIN we push limits to
// the left and right sides, respectively. It's not safe to push limits below FULL OUTER
// JOIN in the general case without a more invasive rewrite.
// Add extra limits below JOIN. For LEFT OUTER and RIGHT OUTER JOIN we push limits to
// the left and right sides, respectively. For INNER and CROSS JOIN we push limits to
// both the left and right sides if join condition is empty. It's not safe to push limits
// below FULL OUTER JOIN in the general case without a more invasive rewrite.
// We also need to ensure that this limit pushdown rule will not eventually introduce limits
// on both sides if it is applied multiple times. Therefore:
// - If one side is already limited, stack another limit on top if the new limit is smaller.
// The redundant limit will be collapsed by the CombineLimits rule.
case LocalLimit(exp, join @ Join(left, right, joinType, _, _)) =>
case LocalLimit(exp, join @ Join(left, right, joinType, conditionOpt, _)) =>
val newJoin = joinType match {
case RightOuter => join.copy(right = maybePushLocalLimit(exp, right))
case LeftOuter => join.copy(left = maybePushLocalLimit(exp, left))
Copy link
Contributor

@c21 c21 Feb 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems that we can also push down limit into left side, for LEFT SEMI and LEFT ANTI join, right?
I can create a minor PR if it's not on your plan @wangyum , thanks.

Copy link
Member Author

@wangyum wangyum Feb 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems we can not pushdown LEFT SEMI JOIN, for example:

spark.range(20).selectExpr("id % 10 as id").repartition(1).write.saveAsTable("t1")
spark.range(5, 9, 1).repartition(1).write.saveAsTable("t2")
val df = spark.sql("select * from t1 LEFT SEMI JOIN t2 on t1.id = t2.id limit 3")
df.explain()
df.show

Current:

== Physical Plan ==
AdaptiveSparkPlan isFinalPlan=false
+- CollectLimit 3
   +- BroadcastHashJoin [id#10L], [id#11L], LeftSemi, BuildRight, false
      :- Filter isnotnull(id#10L)
      :  +- FileScan parquet default.t1[id#10L] Batched: true, DataFilters: [isnotnull(id#10L)], Format: Parquet, Location: InMemoryFileIndex(1 paths)[file:/Users/yumwang/spark/SPARK-28169/spark-warehouse/org.apache.spark..., PartitionFilters: [], PushedFilters: [IsNotNull(id)], ReadSchema: struct<id:bigint>
      +- BroadcastExchange HashedRelationBroadcastMode(List(input[0, bigint, false]),false), [id=#69]
         +- Filter isnotnull(id#11L)
            +- FileScan parquet default.t2[id#11L] Batched: true, DataFilters: [isnotnull(id#11L)], Format: Parquet, Location: InMemoryFileIndex(1 paths)[file:/Users/yumwang/spark/SPARK-28169/spark-warehouse/org.apache.spark..., PartitionFilters: [], PushedFilters: [IsNotNull(id)], ReadSchema: struct<id:bigint>


+---+
| id|
+---+
|  5|
|  6|
|  7|
+---+

Pushdown:

== Physical Plan ==
AdaptiveSparkPlan isFinalPlan=false
+- CollectLimit 3
   +- BroadcastHashJoin [id#10L], [id#11L], LeftSemi, BuildRight, false
      :- LocalLimit 3
      :  +- Filter isnotnull(id#10L)
      :     +- LocalLimit 3
      :        +- FileScan parquet default.t1[id#10L] Batched: true, DataFilters: [], Format: Parquet, Location: InMemoryFileIndex(1 paths)[file:/Users/yumwang/spark/SPARK-28169/spark-warehouse/org.apache.spark..., PartitionFilters: [], PushedFilters: [], ReadSchema: struct<id:bigint>
      +- BroadcastExchange HashedRelationBroadcastMode(List(input[0, bigint, false]),false), [id=#77]
         +- Filter isnotnull(id#11L)
            +- FileScan parquet default.t2[id#11L] Batched: true, DataFilters: [isnotnull(id#11L)], Format: Parquet, Location: InMemoryFileIndex(1 paths)[file:/Users/yumwang/spark/SPARK-28169/spark-warehouse/org.apache.spark..., PartitionFilters: [], PushedFilters: [IsNotNull(id)], ReadSchema: struct<id:bigint>


+---+
| id|
+---+
+---+

Copy link
Contributor

@c21 c21 Feb 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@wangyum - yes, you are right. My bad. LEFT OUTER/RIGHT OUTER join will output the left/right side rows anyway no matter of being matched or not. So they are safe for limit push down, but not LEFT SEMI/LEFT ANTI.

Copy link
Contributor

@c21 c21 Feb 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actually, what about LEFT SEMI / LEFT ANTI join without condition? They should be planned into physical operator BroadcastNestedLoopJoin instead.

case _: InnerLike if conditionOpt.isEmpty =>
join.copy(
left = maybePushLocalLimit(exp, left),
right = maybePushLocalLimit(exp, right))
case _ => join
}
LocalLimit(exp, newJoin)
Expand Down
20 changes: 19 additions & 1 deletion sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/LimitPushdownSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ import org.apache.spark.sql.catalyst.analysis.EliminateSubqueryAliases
import org.apache.spark.sql.catalyst.dsl.expressions._
import org.apache.spark.sql.catalyst.dsl.plans._
import org.apache.spark.sql.catalyst.expressions.Add
import org.apache.spark.sql.catalyst.plans.{FullOuter, LeftOuter, PlanTest, RightOuter}
import org.apache.spark.sql.catalyst.plans.{Cross, FullOuter, Inner, LeftOuter, PlanTest, RightOuter}
import org.apache.spark.sql.catalyst.plans.logical._
import org.apache.spark.sql.catalyst.rules._

Expand Down Expand Up @@ -194,4 +194,22 @@ class LimitPushdownSuite extends PlanTest {
LocalLimit(1, y.groupBy(Symbol("b"))(count(1))))).analyze
comparePlans(expected2, optimized2)
}

test("SPARK-26138: pushdown limit through InnerLike when condition is empty") {
Copy link
Member

@viirya viirya Feb 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shall we have a e2e test to check answer too?

Seq(Cross, Inner).foreach { joinType =>
val originalQuery = x.join(y, joinType).limit(1)
val optimized = Optimize.execute(originalQuery.analyze)
val correctAnswer = Limit(1, LocalLimit(1, x).join(LocalLimit(1, y), joinType)).analyze
comparePlans(optimized, correctAnswer)
}
}

test("SPARK-26138: Should not pushdown limit through InnerLike when condition is not empty") {
Seq(Cross, Inner).foreach { joinType =>
val originalQuery = x.join(y, joinType, Some("x.a".attr === "y.b".attr)).limit(1)
val optimized = Optimize.execute(originalQuery.analyze)
val correctAnswer = Limit(1, x.join(y, joinType, Some("x.a".attr === "y.b".attr))).analyze
comparePlans(optimized, correctAnswer)
}
}
}
17 changes: 15 additions & 2 deletions sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,14 +29,14 @@ import org.apache.spark.scheduler.{SparkListener, SparkListenerJobStart}
import org.apache.spark.sql.catalyst.expressions.GenericRow
import org.apache.spark.sql.catalyst.expressions.aggregate.{Complete, Partial}
import org.apache.spark.sql.catalyst.optimizer.{ConvertToLocalRelation, NestedColumnAliasingSuite}
import org.apache.spark.sql.catalyst.plans.logical.{Project, RepartitionByExpression}
import org.apache.spark.sql.catalyst.plans.logical.{LocalLimit, Project, RepartitionByExpression}
import org.apache.spark.sql.catalyst.util.StringUtils
import org.apache.spark.sql.execution.UnionExec
import org.apache.spark.sql.execution.adaptive.AdaptiveSparkPlanHelper
import org.apache.spark.sql.execution.aggregate.{HashAggregateExec, ObjectHashAggregateExec, SortAggregateExec}
import org.apache.spark.sql.execution.columnar.InMemoryTableScanExec
import org.apache.spark.sql.execution.command.FunctionsCommand
import org.apache.spark.sql.execution.datasources.SchemaColumnConvertNotSupportedException
import org.apache.spark.sql.execution.datasources.{LogicalRelation, SchemaColumnConvertNotSupportedException}
import org.apache.spark.sql.execution.datasources.v2.BatchScanExec
import org.apache.spark.sql.execution.datasources.v2.orc.OrcScan
import org.apache.spark.sql.execution.datasources.v2.parquet.ParquetScan
Expand Down Expand Up @@ -4021,6 +4021,19 @@ class SQLQuerySuite extends QueryTest with SharedSparkSession with AdaptiveSpark
}
}
}

test("SPARK-26138 Pushdown limit through InnerLike when condition is empty") {
withTable("t1", "t2") {
spark.range(5).repartition(1).write.saveAsTable("t1")
spark.range(5).repartition(1).write.saveAsTable("t2")
val df = spark.sql("SELECT * FROM t1 CROSS JOIN t2 LIMIT 3")
val pushedLocalLimits = df.queryExecution.optimizedPlan.collect {
case l @ LocalLimit(_, _: LogicalRelation) => l
}
assert(pushedLocalLimits.length === 2)
checkAnswer(df, Row(0, 0) :: Row(0, 1) :: Row(0, 2) :: Nil)
}
}
}

case class Foo(bar: Option[String])