Skip to content

[SPARK-26601][SQL] Make broadcast-exchange thread pool configurable #23670

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
caneGuy wants to merge 11 commits into from
Closed

[SPARK-26601][SQL] Make broadcast-exchange thread pool configurable #23670

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
0cd3149
[SPARK-26601][SQL] Make broadcast-exchange thread pool configurable
caneGuy Jan 28, 2019
cc9faa6
Fix code style
caneGuy Jan 28, 2019
23ae9a0
Fix code style
caneGuy Jan 28, 2019
ba19ccc
Fix code style
caneGuy Jan 28, 2019
2117ee2
Remove suite
caneGuy Feb 15, 2019
27b0ec4
Refine comment
caneGuy Feb 18, 2019
8da84e3
Merge branch 'master' into zhoukang/make-broadcat-config
caneGuy Mar 26, 2019
a4523e9
Refine
caneGuy Apr 11, 2019
46f454f
Refine comment
caneGuy May 10, 2019
a10dac3
Refine
caneGuy May 12, 2019
40c3592
Refine
caneGuy May 13, 2019
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: 13 additions & 0 deletions sql/catalyst/src/main/scala/org/apache/spark/sql/internal/StaticSQLConf.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,19 @@ object StaticSQLConf {
.intConf
.createWithDefault(1000)

val BROADCAST_EXCHANGE_MAX_THREAD_THRESHOLD =
buildStaticConf("spark.sql.broadcastExchange.maxThreadThreshold")
.internal()
.doc("The maximum degree of parallelism to fetch and broadcast the table. " +
"If we encounter memory issue like frequently full GC or OOM when broadcast table " +
"we can decrease this number in order to reduce memory usage. " +
"Notice the number should be carefully chosen since decreasing parallelism might " +
"cause longer waiting for other broadcasting. Also, increasing parallelism may " +
"cause memory problem.")
.intConf
.checkValue(thres => thres > 0 && thres <= 128, "The threshold must be in [0,128].")
.createWithDefault(128)

val SQL_EVENT_TRUNCATE_LENGTH = buildStaticConf("spark.sql.event.truncate.length")
.doc("Threshold of SQL length beyond which it will be truncated before adding to " +
"event. Defaults to no truncation. If set to 0, callsite will be logged instead.")
Expand Down
5 changes: 3 additions & 2 deletions sql/core/src/main/scala/org/apache/spark/sql/execution/exchange/BroadcastExchangeExec.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,7 +32,7 @@ import org.apache.spark.sql.catalyst.plans.physical.{BroadcastMode, BroadcastPar
import org.apache.spark.sql.execution.{SparkPlan, SQLExecution}
import org.apache.spark.sql.execution.joins.HashedRelation
import org.apache.spark.sql.execution.metric.SQLMetrics
import org.apache.spark.sql.internal.SQLConf
import org.apache.spark.sql.internal.{SQLConf, StaticSQLConf}
import org.apache.spark.util.{SparkFatalException, ThreadUtils}

/**
Expand Down Expand Up @@ -157,5 +157,6 @@ case class BroadcastExchangeExec(

object BroadcastExchangeExec {
private[execution] val executionContext = ExecutionContext.fromExecutorService(
ThreadUtils.newDaemonCachedThreadPool("broadcast-exchange", 128))
ThreadUtils.newDaemonCachedThreadPool("broadcast-exchange",
Copy link
Member

@HyukjinKwon HyukjinKwon May 13, 2019

Choose a reason for hiding this comment

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

nit: the previous indentation was correct.

SQLConf.get.getConf(StaticSQLConf.BROADCAST_EXCHANGE_MAX_THREAD_THRESHOLD)))
}