赞
踩
Spark 3.5.0
目前 Spark中的实现中,对于多分区的写入默认会先排序,这是没必要的。可以设置spark.sql.maxConcurrentOutputFileWriters 为大于0
来避免排序。
这部分主要分为三个部分:
一个是V1Writes规则的重改
;
另一个是FileFormatWriter中的dataWriter的选择
;
还有一个是Spark中为什么会加上Sort
这三部分是需要结合在一起分析讨论的
直接转到代码部分:
object V1Writes extends Rule[LogicalPlan] with SQLConfHelper {
import V1WritesUtils._
override def apply(plan: LogicalPlan): LogicalPlan = {
if (conf.plannedWriteEnabled) {
plan.transformUp {
case write: V1WriteCommand if !write.child.isInstanceOf[WriteFiles] =>
val newQuery = prepareQuery(write, write.query)
val attrMap = AttributeMap(write.query.output.zip(newQuery.output))
val writeFiles = WriteFiles(newQuery, write.fileFormat, write.partitionColumns,
write.bucketSpec, write.options, write.staticPartitions)
val newChild = writeFiles.transformExpressions {
case a: Attribute if attrMap.contains(a) =>
a.withExprId(attrMap(a).exprId)
}
val newWrite = write.withNewChildren(newChild :: Nil).transformExpressions {
case a: Attribute if attrMap.contains(a) =>
a.withExprId(attrMap(a).exprId)
}
newWrite
}
} else {
plan
}
}
其中 prepareQuery
是对满足条件的计划前加上Sort
逻辑排序,其中prepareQuery关键的代码
如下:
val requiredOrdering = write.requiredOrdering.map(_.transform {
case a: Attribute => attrMap.getOrElse(a, a)
}.asInstanceOf[SortOrder])
val outputOrdering = empty2NullPlan.outputOrdering
val orderingMatched = isOrderingMatched(requiredOrdering.map(_.child), outputOrdering)
if (orderingMatched) {
empty2NullPlan
} else {
Sort(requiredOrdering, global = false, empty2NullPlan)
}
write.requiredOrdering
中涉及到的类为InsertIntoHadoopFsRelationCommand
和InsertIntoHiveTable
,且这两个物理计划中的requiredOrdering
实现都是:
V1WritesUtils.getSortOrder(outputColumns, partitionColumns, bucketSpec, options)
getSortOrder
方法关键代码如下:
val sortColumns = V1WritesUtils.getBucketSortColumns(bucketSpec, dataColumns)
if (SQLConf.get.maxConcurrentOutputFileWriters > 0 && sortColumns.isEmpty) {
// Do not insert logical sort when concurrent output writers are enabled.
Seq.empty
} else {
// We should first sort by dynamic partition columns, then bucket id, and finally sorting
// columns.
(dynamicPartitionColumns ++ writerBucketSpec.map(_.bucketIdExpression) ++ sortColumns)
.map(SortOrder(_, Ascending))
}
所以说 如果 spark.sql.maxConcurrentOutputFileWriters为0
(默认值为0),则会加上Sort逻辑计划
,具体的实现可以参考SPARK-37287
如果spark.sql.maxConcurrentOutputFileWriters为0
(默认值为0)且 sortColumns为空
(大部分情况下为空,除非建表是partition加上bucket),则不会加上Sort逻辑计划
在 InsertIntoHadoopFsRelationCommand
和InsertIntoHiveTable
这两个物理计划中,最终写入文件/数据的时候,会调用到FileFormatWriter.write
方法,这里有个concurrentOutputWriterSpecFunc
函数变量的设置:
val concurrentOutputWriterSpecFunc = (plan: SparkPlan) => {
val sortPlan = createSortPlan(plan, requiredOrdering, outputSpec)
createConcurrentOutputWriterSpec(sparkSession, sortPlan, sortColumns)
}
val writeSpec = WriteFilesSpec(
description = description,
committer = committer,
concurrentOutputWriterSpecFunc = concurrentOutputWriterSpecFunc
)
executeWrite(sparkSession, plan, writeSpec, job)
设置concurrentOutputWriterSpecFunc
的代码如下:
private def createConcurrentOutputWriterSpec(
sparkSession: SparkSession,
sortPlan: SortExec,
sortColumns: Seq[Attribute]): Option[ConcurrentOutputWriterSpec] = {
val maxWriters = sparkSession.sessionState.conf.maxConcurrentOutputFileWriters
val concurrentWritersEnabled = maxWriters > 0 && sortColumns.isEmpty
if (concurrentWritersEnabled) {
Some(ConcurrentOutputWriterSpec(maxWriters, () => sortPlan.createSorter()))
} else {
None
}
}
如果 spark.sql.maxConcurrentOutputFileWriters为0
(默认值为0),则ConcurrentOutputWriterSpec
为None
如果 spark.sql.maxConcurrentOutputFileWriters大于0
且 sortColumns为空
(大部分情况下为空,除非建表是partition加上bucket),则为Some(ConcurrentOutputWriterSpec(maxWriters, () => sortPlan.createSorter())
其中executeWrite
会调用WriteFilesExec.doExecuteWrite
方法,从而调用FileFormatWriter.executeTask
,这里就涉及到dataWriter
选择:
val dataWriter =
if (sparkPartitionId != 0 && !iterator.hasNext) {
// In case of empty job, leave first partition to save meta for file format like parquet.
new EmptyDirectoryDataWriter(description, taskAttemptContext, committer)
} else if (description.partitionColumns.isEmpty && description.bucketSpec.isEmpty) {
new SingleDirectoryDataWriter(description, taskAttemptContext, committer)
} else {
concurrentOutputWriterSpec match {
case Some(spec) =>
new DynamicPartitionDataConcurrentWriter(
description, taskAttemptContext, committer, spec)
case _ =>
new DynamicPartitionDataSingleWriter(description, taskAttemptContext, committer)
}
}
这里其实会根据 concurrentOutputWriterSpec来选择不同的dataWriter
,默认情况下为DynamicPartitionDataSingleWriter
否则就会为DynamicPartitionDataConcurrentWriter
这两者的区别,见下文
至于Spark在写入文件的时候会加上Sort,这个是跟写入的实现有关的,也就是DynamicPartitionDataSingleWriter
和DynamicPartitionDataConcurrentWriter
的区别:
对于DynamicPartitionDataSingleWriter
会根据partition或者bucket作为最细粒度来作为writer的标准,如果相邻的两条记录所属不同的partition或者bucket,则会切换writer,所以说如果不根据partition或者bucket排序的话,会导致writer
频繁的切换,这会大大降低文件的写入速度。所以说需要根据partition或者bucket进行排序。
Copyright © 2003-2013 www.wpsshop.cn 版权所有,并保留所有权利。