ODIn/spark-instrumented-optimizer
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

Add number of bytes spilled to Web UI

Browse source
This commit is contained in:
Andrew Or 2014-01-10 21:40:55 -08:00
parent e6447152b3
commit bb8098f203
11 changed files with 75 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
core/src/main/scala/org/apache/spark/Aggregator.scala
View file

@ -32,9 +32,9 @@ case class Aggregator[K, V, C] (
mergeCombiners: (C, C) => C) {
private val sparkConf = SparkEnv.get.conf
private val externalSorting = sparkConf.getBoolean("spark.shuffle.externalSorting", true)
private val externalSorting = sparkConf.getBoolean("spark.shuffle.externalSorting", false)
def combineValuesByKey(iter: Iterator[_ <: Product2[K, V]]) : Iterator[(K, C)] = {
def combineValuesByKey(iter: Iterator[_ <: Product2[K, V]], context: TaskContext) : Iterator[(K, C)] = {
if (!externalSorting) {
val combiners = new AppendOnlyMap[K,C]
var kv: Product2[K, V] = null
@ -53,11 +53,12 @@ case class Aggregator[K, V, C] (
val (k, v) = iter.next()
combiners.insert(k, v)
}
combiners.registerBytesSpilled(context.attemptId)
combiners.iterator
}
}
def combineCombinersByKey(iter: Iterator[(K, C)]) : Iterator[(K, C)] = {
def combineCombinersByKey(iter: Iterator[(K, C)], context: TaskContext) : Iterator[(K, C)] = {
if (!externalSorting) {
val combiners = new AppendOnlyMap[K,C]
var kc: Product2[K, C] = null
@ -75,6 +76,7 @@ case class Aggregator[K, V, C] (
val (k, c) = iter.next()
combiners.insert(k, c)
}
combiners.registerBytesSpilled(context.attemptId)
combiners.iterator
}
}

3
core/src/main/scala/org/apache/spark/SparkEnv.scala
View file

@ -60,6 +60,9 @@ class SparkEnv private[spark] (
// All accesses should be manually synchronized
val shuffleMemoryMap = mutable.HashMap[Long, Long]()
// A mapping of task ID to number of bytes spilled by that task. This is mainly for book-keeping.
val bytesSpilledMap = mutable.HashMap[Long, Long]()
private val pythonWorkers = mutable.HashMap[(String, Map[String, String]), PythonWorkerFactory]()
// A general, soft-reference map for metadata needed during HadoopRDD split computation

4
core/src/main/scala/org/apache/spark/executor/Executor.scala
View file

@ -229,6 +229,7 @@ private[spark] class Executor(
m.executorRunTime = (taskFinish - taskStart).toInt
m.jvmGCTime = gcTime - startGCTime
m.resultSerializationTime = (afterSerialization - beforeSerialization).toInt
m.bytesSpilled = env.bytesSpilledMap.get(taskId).getOrElse(0)
}
val accumUpdates = Accumulators.values
@ -279,11 +280,12 @@ private[spark] class Executor(
//System.exit(1)
}
} finally {
// TODO: Unregister shuffle memory only for ShuffleMapTask
// TODO: Unregister shuffle memory only for ResultTask
val shuffleMemoryMap = env.shuffleMemoryMap
shuffleMemoryMap.synchronized {
shuffleMemoryMap.remove(Thread.currentThread().getId)
}
env.bytesSpilledMap.remove(taskId)
runningTasks.remove(taskId)
}
}

5
core/src/main/scala/org/apache/spark/executor/TaskMetrics.scala
View file

@ -48,6 +48,11 @@ class TaskMetrics extends Serializable {
*/
var resultSerializationTime: Long = _
/**
* The number of bytes spilled to disk by this task
*/
var bytesSpilled: Long = _
/**
* If this task reads from shuffle output, metrics on getting shuffle data will be collected here
*/

4
core/src/main/scala/org/apache/spark/rdd/CoGroupedRDD.scala
View file

@ -106,7 +106,8 @@ class CoGroupedRDD[K](@transient var rdds: Seq[RDD[_ <: Product2[K, _]]], part:
override val partitioner = Some(part)
override def compute(s: Partition, context: TaskContext): Iterator[(K, CoGroupCombiner)] = {
val externalSorting = sparkConf.getBoolean("spark.shuffle.externalSorting", true)
val externalSorting = sparkConf.getBoolean("spark.shuffle.externalSorting", false)
val split = s.asInstanceOf[CoGroupPartition]
val numRdds = split.deps.size
@ -150,6 +151,7 @@ class CoGroupedRDD[K](@transient var rdds: Seq[RDD[_ <: Product2[K, _]]], part:
map.insert(kv._1, new CoGroupValue(kv._2, depNum))
}
}
map.registerBytesSpilled(context.attemptId)
new InterruptibleIterator(context, map.iterator)
}
}

10
core/src/main/scala/org/apache/spark/rdd/PairRDDFunctions.scala
View file

@ -88,20 +88,22 @@ class PairRDDFunctions[K: ClassTag, V: ClassTag](self: RDD[(K, V)])
val aggregator = new Aggregator[K, V, C](createCombiner, mergeValue, mergeCombiners)
if (self.partitioner == Some(partitioner)) {
self.mapPartitionsWithContext((context, iter) => {
new InterruptibleIterator(context, aggregator.combineValuesByKey(iter))
new InterruptibleIterator(context, aggregator.combineValuesByKey(iter, context))
}, preservesPartitioning = true)
} else if (mapSideCombine) {
val combined = self.mapPartitions(aggregator.combineValuesByKey, preservesPartitioning = true)
val combined = self.mapPartitionsWithContext((context, iter) => {
aggregator.combineValuesByKey(iter, context)
}, preservesPartitioning = true)
val partitioned = new ShuffledRDD[K, C, (K, C)](combined, partitioner)
.setSerializer(serializerClass)
partitioned.mapPartitionsWithContext((context, iter) => {
new InterruptibleIterator(context, aggregator.combineCombinersByKey(iter))
new InterruptibleIterator(context, aggregator.combineCombinersByKey(iter, context))
}, preservesPartitioning = true)
} else {
// Don't apply map-side combiner.
val values = new ShuffledRDD[K, V, (K, V)](self, partitioner).setSerializer(serializerClass)
values.mapPartitionsWithContext((context, iter) => {
new InterruptibleIterator(context, aggregator.combineValuesByKey(iter))
new InterruptibleIterator(context, aggregator.combineValuesByKey(iter, context))
}, preservesPartitioning = true)
}
}

1
core/src/main/scala/org/apache/spark/ui/jobs/ExecutorSummary.scala
View file

@ -24,4 +24,5 @@ private[spark] class ExecutorSummary {
var succeededTasks : Int = 0
var shuffleRead : Long = 0
var shuffleWrite : Long = 0
var bytesSpilled : Long = 0
}

2
core/src/main/scala/org/apache/spark/ui/jobs/ExecutorTable.scala
View file

@ -48,6 +48,7 @@ private[spark] class ExecutorTable(val parent: JobProgressUI, val stageId: Int)
<th>Succeeded Tasks</th>
<th>Shuffle Read</th>
<th>Shuffle Write</th>
<th>Bytes Spilled</th>
</thead>
<tbody>
{createExecutorTable()}
@ -80,6 +81,7 @@ private[spark] class ExecutorTable(val parent: JobProgressUI, val stageId: Int)
<td>{v.succeededTasks}</td>
<td>{Utils.bytesToString(v.shuffleRead)}</td>
<td>{Utils.bytesToString(v.shuffleWrite)}</td>
<td>{Utils.bytesToString(v.bytesSpilled)}</td>
</tr>
}
}

7
core/src/main/scala/org/apache/spark/ui/jobs/JobProgressListener.scala
View file

@ -52,6 +52,7 @@ private[spark] class JobProgressListener(val sc: SparkContext) extends SparkList
val stageIdToTime = HashMap[Int, Long]()
val stageIdToShuffleRead = HashMap[Int, Long]()
val stageIdToShuffleWrite = HashMap[Int, Long]()
val stageIdToBytesSpilled = HashMap[Int, Long]()
val stageIdToTasksActive = HashMap[Int, HashSet[TaskInfo]]()
val stageIdToTasksComplete = HashMap[Int, Int]()
val stageIdToTasksFailed = HashMap[Int, Int]()
@ -78,6 +79,7 @@ private[spark] class JobProgressListener(val sc: SparkContext) extends SparkList
stageIdToTime.remove(s.stageId)
stageIdToShuffleRead.remove(s.stageId)
stageIdToShuffleWrite.remove(s.stageId)
stageIdToBytesSpilled.remove(s.stageId)
stageIdToTasksActive.remove(s.stageId)
stageIdToTasksComplete.remove(s.stageId)
stageIdToTasksFailed.remove(s.stageId)
@ -149,6 +151,7 @@ private[spark] class JobProgressListener(val sc: SparkContext) extends SparkList
Option(taskEnd.taskMetrics).foreach { taskMetrics =>
taskMetrics.shuffleReadMetrics.foreach { y.shuffleRead += _.remoteBytesRead }
taskMetrics.shuffleWriteMetrics.foreach { y.shuffleWrite += _.shuffleBytesWritten }
y.bytesSpilled += taskMetrics.bytesSpilled
}
}
case _ => {}
@ -184,6 +187,10 @@ private[spark] class JobProgressListener(val sc: SparkContext) extends SparkList
stageIdToShuffleWrite(sid) += shuffleWrite
totalShuffleWrite += shuffleWrite
stageIdToBytesSpilled.getOrElseUpdate(sid, 0L)
val bytesSpilled = metrics.map(m => m.bytesSpilled).getOrElse(0L)
stageIdToBytesSpilled(sid) += bytesSpilled
val taskList = stageIdToTaskInfos.getOrElse(
sid, HashSet[(TaskInfo, Option[TaskMetrics], Option[ExceptionFailure])]())
taskList -= ((taskEnd.taskInfo, None, None))

32
core/src/main/scala/org/apache/spark/ui/jobs/StagePage.scala
View file

@ -56,6 +56,8 @@ private[spark] class StagePage(parent: JobProgressUI) {
val hasShuffleRead = shuffleReadBytes > 0
val shuffleWriteBytes = listener.stageIdToShuffleWrite.getOrElse(stageId, 0L)
val hasShuffleWrite = shuffleWriteBytes > 0
val bytesSpilled = listener.stageIdToBytesSpilled.getOrElse(stageId, 0L)
val hasBytesSpilled = bytesSpilled > 0
var activeTime = 0L
listener.stageIdToTasksActive(stageId).foreach(activeTime += _.timeRunning(now))
@ -81,6 +83,12 @@ private[spark] class StagePage(parent: JobProgressUI) {
{Utils.bytesToString(shuffleWriteBytes)}
</li>
}
{if (hasBytesSpilled)
<li>
<strong>Bytes spilled: </strong>
{Utils.bytesToString(bytesSpilled)}
</li>
}
</ul>
</div>
@ -89,9 +97,10 @@ private[spark] class StagePage(parent: JobProgressUI) {
Seq("Duration", "GC Time", "Result Ser Time") ++
{if (hasShuffleRead) Seq("Shuffle Read") else Nil} ++
{if (hasShuffleWrite) Seq("Write Time", "Shuffle Write") else Nil} ++
{if (hasBytesSpilled) Seq("Bytes Spilled") else Nil} ++
Seq("Errors")
val taskTable = listingTable(taskHeaders, taskRow(hasShuffleRead, hasShuffleWrite), tasks)
val taskTable = listingTable(taskHeaders, taskRow(hasShuffleRead, hasShuffleWrite, hasBytesSpilled), tasks)
// Excludes tasks which failed and have incomplete metrics
val validTasks = tasks.filter(t => t._1.status == "SUCCESS" && (t._2.isDefined))
@ -153,13 +162,20 @@ private[spark] class StagePage(parent: JobProgressUI) {
}
val shuffleWriteQuantiles = "Shuffle Write" +: getQuantileCols(shuffleWriteSizes)
val bytesSpilledSizes = validTasks.map {
case(info, metrics, exception) =>
metrics.get.bytesSpilled.toDouble
}
val bytesSpilledQuantiles = "Bytes Spilled" +: getQuantileCols(bytesSpilledSizes)
val listings: Seq[Seq[String]] = Seq(
serializationQuantiles,
serviceQuantiles,
gettingResultQuantiles,
schedulerDelayQuantiles,
if (hasShuffleRead) shuffleReadQuantiles else Nil,
if (hasShuffleWrite) shuffleWriteQuantiles else Nil)
if (hasShuffleWrite) shuffleWriteQuantiles else Nil,
if (hasBytesSpilled) bytesSpilledQuantiles else Nil)
val quantileHeaders = Seq("Metric", "Min", "25th percentile",
"Median", "75th percentile", "Max")
@ -178,8 +194,7 @@ private[spark] class StagePage(parent: JobProgressUI) {
}
}
def taskRow(shuffleRead: Boolean, shuffleWrite: Boolean)
def taskRow(shuffleRead: Boolean, shuffleWrite: Boolean, bytesSpilled: Boolean)
(taskData: (TaskInfo, Option[TaskMetrics], Option[ExceptionFailure])): Seq[Node] = {
def fmtStackTrace(trace: Seq[StackTraceElement]): Seq[Node] =
trace.map(e => <span style="display:block;">{e.toString}</span>)
@ -205,6 +220,10 @@ private[spark] class StagePage(parent: JobProgressUI) {
val writeTimeReadable = maybeWriteTime.map{ t => t / (1000 * 1000)}.map{ ms =>
if (ms == 0) "" else parent.formatDuration(ms)}.getOrElse("")
val maybeBytesSpilled = metrics.map{m => m.bytesSpilled}
val bytesSpilledSortable = maybeBytesSpilled.map(_.toString).getOrElse("")
val bytesSpilledReadable = maybeBytesSpilled.map{Utils.bytesToString(_)}.getOrElse("")
<tr>
<td>{info.index}</td>
<td>{info.taskId}</td>
@ -234,6 +253,11 @@ private[spark] class StagePage(parent: JobProgressUI) {
{shuffleWriteReadable}
</td>
}}
{if (bytesSpilled) {
<td sorttable_customkey={bytesSpilledSortable}>
{bytesSpilledReadable}
</td>
}}
<td>{exception.map(e =>
<span>
{e.className} ({e.description})<br/>

13
core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala
View file

@ -77,7 +77,7 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
}
// Number of pairs in the in-memory map
private var numPairsInMemory = 0
private var numPairsInMemory = 0L
// Number of in-memory pairs inserted before tracking the map's shuffle memory usage
private val trackMemoryThreshold = 1000
@ -85,6 +85,9 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
// How many times we have spilled so far
private var spillCount = 0
// Number of bytes spilled in total
private var bytesSpilled = 0L
private val fileBufferSize = sparkConf.getInt("spark.shuffle.file.buffer.kb", 100) * 1024
private val syncWrites = sparkConf.getBoolean("spark.shuffle.sync", false)
private val comparator = new KCComparator[K, C]
@ -161,6 +164,14 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
shuffleMemoryMap(Thread.currentThread().getId) = 0
}
numPairsInMemory = 0
bytesSpilled += mapSize
}
/**
* Register the total number of bytes spilled by this task
*/
def registerBytesSpilled(taskId: Long) {
SparkEnv.get.bytesSpilledMap(taskId) = bytesSpilled
}
/**