Skip to content

[SPARK-20186][SQL] BroadcastHint should use child's stats #17504

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 1 commit into from
Closed

[SPARK-20186][SQL] BroadcastHint should use child's stats #17504

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion ...st/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -383,7 +383,7 @@ case class BroadcastHint(child: LogicalPlan) extends UnaryNode {

// set isBroadcastable to true so the child will be broadcasted
override def computeStats(conf: CatalystConf): Statistics =
super.computeStats(conf).copy(isBroadcastable = true)
child.stats(conf).copy(isBroadcastable = true)
}

/**
Expand Down
21 changes: 20 additions & 1 deletion .../test/scala/org/apache/spark/sql/catalyst/statsEstimation/BasicStatsEstimationSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,23 @@ class BasicStatsEstimationSuite extends StatsEstimationTestBase {
// row count * (overhead + column size)
size = Some(10 * (8 + 4)))

test("BroadcastHint estimation") {
val filter = Filter(Literal(true), plan)
val filterStatsCboOn = Statistics(sizeInBytes = 10 * (8 +4), isBroadcastable = false,
rowCount = Some(10), attributeStats = AttributeMap(Seq(attribute -> colStat)))
val filterStatsCboOff = Statistics(sizeInBytes = 10 * (8 +4), isBroadcastable = false)
checkStats(
filter,
expectedStatsCboOn = filterStatsCboOn,
expectedStatsCboOff = filterStatsCboOff)

val broadcastHint = BroadcastHint(filter)
checkStats(
broadcastHint,
expectedStatsCboOn = filterStatsCboOn.copy(isBroadcastable = true),
expectedStatsCboOff = filterStatsCboOff.copy(isBroadcastable = true))
}

test("limit estimation: limit < child's rowCount") {
val localLimit = LocalLimit(Literal(2), plan)
val globalLimit = GlobalLimit(Literal(2), plan)
Expand Down Expand Up @@ -97,8 +114,10 @@ class BasicStatsEstimationSuite extends StatsEstimationTestBase {
plan: LogicalPlan,
expectedStatsCboOn: Statistics,
expectedStatsCboOff: Statistics): Unit = {
assert(plan.stats(conf.copy(cboEnabled = true)) == expectedStatsCboOn)
// Invalidate statistics
plan.invalidateStatsCache()
assert(plan.stats(conf.copy(cboEnabled = true)) == expectedStatsCboOn)

plan.invalidateStatsCache()
assert(plan.stats(conf.copy(cboEnabled = false)) == expectedStatsCboOff)
}
Expand Down