diff options
Diffstat (limited to 'core')
3 files changed, 12 insertions, 0 deletions
diff --git a/core/src/main/scala/org/apache/spark/shuffle/FileShuffleBlockManager.scala b/core/src/main/scala/org/apache/spark/shuffle/FileShuffleBlockManager.scala index 660df00bc3..d0178dfde6 100644 --- a/core/src/main/scala/org/apache/spark/shuffle/FileShuffleBlockManager.scala +++ b/core/src/main/scala/org/apache/spark/shuffle/FileShuffleBlockManager.scala @@ -112,6 +112,7 @@ class FileShuffleBlockManager(conf: SparkConf) private val shuffleState = shuffleStates(shuffleId) private var fileGroup: ShuffleFileGroup = null + val openStartTime = System.nanoTime val writers: Array[BlockObjectWriter] = if (consolidateShuffleFiles) { fileGroup = getUnusedFileGroup() Array.tabulate[BlockObjectWriter](numBuckets) { bucketId => @@ -135,6 +136,9 @@ class FileShuffleBlockManager(conf: SparkConf) blockManager.getDiskWriter(blockId, blockFile, serializer, bufferSize, writeMetrics) } } + // Creating the file to write to and creating a disk writer both involve interacting with + // the disk, so should be included in the shuffle write time. + writeMetrics.incShuffleWriteTime(System.nanoTime - openStartTime) override def releaseWriters(success: Boolean) { if (consolidateShuffleFiles) { 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 fa2e617762..55ea0f17b1 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 @@ -63,6 +63,9 @@ private[spark] class SortShuffleWriter[K, V, C]( sorter.insertAll(records) } + // Don't bother including the time to open the merged output file in the shuffle write time, + // because it just opens a single file, so is typically too fast to measure accurately + // (see SPARK-3570). val outputFile = shuffleBlockManager.getDataFile(dep.shuffleId, mapId) val blockId = shuffleBlockManager.consolidateId(dep.shuffleId, mapId) val partitionLengths = sorter.writePartitionedFile(blockId, context, outputFile) 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 3262e670c2..b962c101c9 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 @@ -352,6 +352,7 @@ private[spark] class ExternalSorter[K, V, C]( // Create our file writers if we haven't done so yet if (partitionWriters == null) { curWriteMetrics = new ShuffleWriteMetrics() + val openStartTime = System.nanoTime partitionWriters = Array.fill(numPartitions) { // Because these files may be read during shuffle, their compression must be controlled by // spark.shuffle.compress instead of spark.shuffle.spill.compress, so we need to use @@ -359,6 +360,10 @@ private[spark] class ExternalSorter[K, V, C]( val (blockId, file) = diskBlockManager.createTempShuffleBlock() blockManager.getDiskWriter(blockId, file, ser, fileBufferSize, curWriteMetrics).open() } + // Creating the file to write to and creating a disk writer both involve interacting with + // the disk, and can take a long time in aggregate when we open many files, so should be + // included in the shuffle write time. + curWriteMetrics.incShuffleWriteTime(System.nanoTime - openStartTime) } // No need to sort stuff, just write each element out |