[SPARK-2503] Lower shuffle output buffer (spark.shuffle.file.buffer.kb) to 32KB.

This can substantially reduce memory usage during shuffle.

Author: Reynold Xin <rxin@apache.org>

Closes #1781 from rxin/SPARK-2503-spark.shuffle.file.buffer.kb and squashes the following commits:

104b8d8 [Reynold Xin] [SPARK-2503] Lower shuffle output buffer (spark.shuffle.file.buffer.kb) to 32KB.

4 files changed, 4 insertions, 4 deletions

diff --git a/core/src/main/scala/org/apache/spark/shuffle/sort/SortShuffleWriter.scala b/core/src/main/scala/org/apache/spark/shuffle/sort/SortShuffleWriter.scala
index 9a356d0dba..24db2f287a 100644
--- a/core/src/main/scala/org/apache/spark/shuffle/sort/SortShuffleWriter.scala
+++ b/core/src/main/scala/org/apache/spark/shuffle/sort/SortShuffleWriter.scala
@@ -40,7 +40,7 @@ private[spark] class SortShuffleWriter[K, V, C](
   private val ser = Serializer.getSerializer(dep.serializer.orNull)
 
   private val conf = SparkEnv.get.conf
-  private val fileBufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 100) * 1024
+  private val fileBufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 32) * 1024
 
   private var sorter: ExternalSorter[K, V, _] = null
   private var outputFile: File = null
diff --git a/core/src/main/scala/org/apache/spark/storage/ShuffleBlockManager.scala b/core/src/main/scala/org/apache/spark/storage/ShuffleBlockManager.scala
index 28aa35bc7e..f9fdffae8b 100644
--- a/core/src/main/scala/org/apache/spark/storage/ShuffleBlockManager.scala
+++ b/core/src/main/scala/org/apache/spark/storage/ShuffleBlockManager.scala
@@ -73,7 +73,7 @@ class ShuffleBlockManager(blockManager: BlockManager) extends Logging {
   val sortBasedShuffle =
     conf.get("spark.shuffle.manager", "") == classOf[SortShuffleManager].getName
 
-  private val bufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 100) * 1024
+  private val bufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 32) * 1024
 
   /**
    * Contains all the state related to a particular shuffle. This includes a pool of unused
diff --git a/core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala b/core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala
index cc0423856c..260a5c3888 100644
--- a/core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala
+++ b/core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala
@@ -101,7 +101,7 @@ class ExternalAppendOnlyMap[K, V, C](
   private var _memoryBytesSpilled = 0L
   private var _diskBytesSpilled = 0L
 
-  private val fileBufferSize = sparkConf.getInt("spark.shuffle.file.buffer.kb", 100) * 1024
+  private val fileBufferSize = sparkConf.getInt("spark.shuffle.file.buffer.kb", 32) * 1024
   private val keyComparator = new HashComparator[K]
   private val ser = serializer.newInstance()
 
diff --git a/core/src/main/scala/org/apache/spark/util/collection/ExternalSorter.scala b/core/src/main/scala/org/apache/spark/util/collection/ExternalSorter.scala
index 101c83b264..3f93afd57b 100644
--- a/core/src/main/scala/org/apache/spark/util/collection/ExternalSorter.scala
+++ b/core/src/main/scala/org/apache/spark/util/collection/ExternalSorter.scala
@@ -84,7 +84,7 @@ private[spark] class ExternalSorter[K, V, C](
 
   private val conf = SparkEnv.get.conf
   private val spillingEnabled = conf.getBoolean("spark.shuffle.spill", true)
-  private val fileBufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 100) * 1024
+  private val fileBufferSize = conf.getInt("spark.shuffle.file.buffer.kb", 32) * 1024
 
   // Size of object batches when reading/writing from serializers.
   //