Skip to content
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

[SPARK-21137][CORE] Spark reads many small files slowly #18441

Closed
wants to merge 1 commit into from
Closed

[SPARK-21137][CORE] Spark reads many small files slowly #18441

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
7 changes: 6 additions & 1 deletion core/src/main/scala/org/apache/spark/rdd/BinaryFileRDD.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ package org.apache.spark.rdd
import org.apache.hadoop.conf.{Configurable, Configuration}
import org.apache.hadoop.io.Writable
import org.apache.hadoop.mapreduce._
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat
import org.apache.hadoop.mapreduce.task.JobContextImpl

import org.apache.spark.{Partition, SparkContext}
Expand All @@ -35,8 +36,12 @@ private[spark] class BinaryFileRDD[T](
extends NewHadoopRDD[String, T](sc, inputFormatClass, keyClass, valueClass, conf) {

override def getPartitions: Array[Partition] = {
val inputFormat = inputFormatClass.newInstance
val conf = getConf
// setMinPartitions below will call FileInputFormat.listStatus(), which can be quite slow when
// traversing a large number of directories and files. Parallelize it.
conf.setIfUnset(FileInputFormat.LIST_STATUS_NUM_THREADS,
Runtime.getRuntime.availableProcessors().toString)
Copy link
Contributor

Choose a reason for hiding this comment

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

shall we use CPU_CORES_PER_EXECUTOR?

Copy link
Member

Choose a reason for hiding this comment

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

+1.
This code aggressively tries to use all of the cores on all of the Spark instances on a node.

Copy link
Member Author

Choose a reason for hiding this comment

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

@cloud-fan @kiszk this will happen on the driver, where it goes to assess the size of the data in order to compute the desired number of partitions. I don't know what the right value is here, to be honest. It won't actually use a ton of CPU because these threads will mostly be waiting for the external FS processes.

I didn't think it's worth yet another config, and didn't know what else to base it on. The number of driver cores maybe?

What is CPU_CORES_PER_EXECUTOR by the way, I don't see that in the code? I can check for spark.driver.cores in the Spark config here.

Copy link
Contributor

Choose a reason for hiding this comment

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

ah, if it's running on driver side, I think it's ok.

val inputFormat = inputFormatClass.newInstance
inputFormat match {
case configurable: Configurable =>
configurable.setConf(conf)
Expand Down
7 changes: 6 additions & 1 deletion core/src/main/scala/org/apache/spark/rdd/WholeTextFileRDD.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ package org.apache.spark.rdd
import org.apache.hadoop.conf.{Configurable, Configuration}
import org.apache.hadoop.io.{Text, Writable}
import org.apache.hadoop.mapreduce.InputSplit
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat
import org.apache.hadoop.mapreduce.task.JobContextImpl

import org.apache.spark.{Partition, SparkContext}
Expand All @@ -38,8 +39,12 @@ private[spark] class WholeTextFileRDD(
extends NewHadoopRDD[Text, Text](sc, inputFormatClass, keyClass, valueClass, conf) {

override def getPartitions: Array[Partition] = {
val inputFormat = inputFormatClass.newInstance
val conf = getConf
// setMinPartitions below will call FileInputFormat.listStatus(), which can be quite slow when
// traversing a large number of directories and files. Parallelize it.
conf.setIfUnset(FileInputFormat.LIST_STATUS_NUM_THREADS,
Runtime.getRuntime.availableProcessors().toString)
val inputFormat = inputFormatClass.newInstance
inputFormat match {
case configurable: Configurable =>
configurable.setConf(conf)
Expand Down