spark源码阅读——shuffle读
2018-08-18 本文已影响19人
WJL3333
DAGScheduler在拆分任务的时候如果发现需要shuffle则会把之前RDD运算产生的结果输出到本地磁盘中（详细的会在以后的文章分析）。
紧接着就需要对Shuffle后的结果分别进行运算了（比如说count）
那么接着之前的RDD会有一个ShuffledRDD来处理shuffle之后的结果。
（实际上是一个新的Stage）
同样在这个Stage会把任务拆分成Task并发送给Executor
这里拆分成的Task是ResultTask实际上也很简单，任务反序列化之后执行ShuffledRDD.iterator -> ShuffledRDD.compute
override def compute(split: Partition, context: TaskContext): Iterator[(K, C)] = {
val dep = dependencies.head.asInstanceOf[ShuffleDependency[K, V, C]]
SparkEnv.get.shuffleManager.getReader(dep.shuffleHandle, split.index, split.index + 1, context)
.read()
.asInstanceOf[Iterator[(K, C)]]
}
从ShuffleManager获取一个Reader读取之前Shuffle输出的数据进行运算。
实际这个Reader是一个BlockStoreShuffleReader
这个类会做什么呢？
- 首先任务是计算我这个partition的结果，我需要知道之前依赖的partition的数据的位置（MapOutputTracker）
- 根据位置获取依赖的数据。（BlockManager）
- 如果需要combine则执行聚合逻辑
- 如果需要排序则排序（ExternalSorter）
/**
* Fetches and reads the partitions in range [startPartition, endPartition) from a shuffle by
* requesting them from other nodes' block stores.
*/
private[spark] class BlockStoreShuffleReader[K, C](
handle: BaseShuffleHandle[K, _, C],
startPartition: Int,
endPartition: Int,
context: TaskContext,
serializerManager: SerializerManager = SparkEnv.get.serializerManager,
blockManager: BlockManager = SparkEnv.get.blockManager,
mapOutputTracker: MapOutputTracker = SparkEnv.get.mapOutputTracker)
extends ShuffleReader[K, C] with Logging {
override def read(): Iterator[Product2[K, C]] = {
val wrappedStreams = new ShuffleBlockFetcherIterator(
context,
blockManager.shuffleClient,
blockManager,
mapOutputTracker.getMapSizesByExecutorId(handle.shuffleId, startPartition, endPartition),
serializerManager.wrapStream,
...
)
val serializerInstance = dep.serializer.newInstance()
// Create a key/value iterator for each stream
val recordIter = wrappedStreams.flatMap { case (blockId, wrappedStream) =>
...
serializerInstance.deserializeStream(wrappedStream).asKeyValueIterator
}
...
val interruptibleIter = new InterruptibleIterator[(Any, Any)](context, metricIter)
val aggregatedIter: Iterator[Product2[K, C]] = if (dep.aggregator.isDefined) {
if (dep.mapSideCombine) {
...
val combinedKeyValuesIterator = interruptibleIter.asInstanceOf[Iterator[(K, C)]]
dep.aggregator.get.combineCombinersByKey(combinedKeyValuesIterator, context)
} else {
val keyValuesIterator = interruptibleIter.asInstanceOf[Iterator[(K, Nothing)]]
dep.aggregator.get.combineValuesByKey(keyValuesIterator, context)
}
} else {
...
interruptibleIter.asInstanceOf[Iterator[Product2[K, C]]]
}
// Sort the output if there is a sort ordering defined.
dep.keyOrdering match {
case Some(keyOrd: Ordering[K]) =>
// Create an ExternalSorter to sort the data.
val sorter =
new ExternalSorter[K, C, C](context, ordering = Some(keyOrd), serializer = dep.serializer)
sorter.insertAll(aggregatedIter)
...
CompletionIterator[Product2[K, C], Iterator[Product2[K, C]]](sorter.iterator, sorter.stop())
case None =>
aggregatedIter
}
}
mapOutputTracker.getMapSizesByExecutorId(handle.shuffleId, startPartition, endPartition)
这个方法返回的参数类型是Seq[(BlockManagerId, Seq[(BlockId, Long)])]
也就是说在哪个Executor的哪个位置上保存着需要的数据信息，数据大小是多少。
（每个Executor都有一个SparkEnv，每个SparkEnv都包含一个BlockManager，Driver端的BlockManager是Master，Executor端的BlockManager是slave，集群内部的BlockManager构成了一个master-slave模式的集群，以后会说）
ShuffleBlockFetcherIterator
这个对象实际上就负责的从远程节点拉取所有数据的任务。
首先到了initialize方法
- 注册回调，方便任务结束后清理内存（ByteBuffer）
- 分离请求，有的数据实际可能保存在本地
- 获取远程数据块
- 获取本地数据块
private[this] def initialize(): Unit = {
context.addTaskCompletionListener(_ => cleanup())
...
val remoteRequests = splitLocalRemoteBlocks()
...
fetchRequests ++= Utils.randomize(remoteRequests)
...
fetchUpToMaxBytes()
...
fetchLocalBlocks()
...
}
这里将数据块信息拆分成获取任务的列表，这里有一个优化，为了加快获取速度，会将同一个文件拆分成多个请求同时获取。
private[this] def splitLocalRemoteBlocks(): ArrayBuffer[FetchRequest] = {
// Make remote requests at most maxBytesInFlight / 5 in length; the reason to keep them
// smaller than maxBytesInFlight is to allow multiple, parallel fetches from up to 5
// nodes, rather than blocking on reading output from one node.
val targetRequestSize = math.max(maxBytesInFlight / 5, 1L)
logDebug("maxBytesInFlight: " + maxBytesInFlight + ", targetRequestSize: " + targetRequestSize
+ ", maxBlocksInFlightPerAddress: " + maxBlocksInFlightPerAddress)
// Split local and remote blocks. Remote blocks are further split into FetchRequests of size
// at most maxBytesInFlight in order to limit the amount of data in flight.
val remoteRequests = new ArrayBuffer[FetchRequest]
// Tracks total number of blocks (including zero sized blocks)
var totalBlocks = 0
for ((address, blockInfos) <- blocksByAddress) {
totalBlocks += blockInfos.size
if (address.executorId == blockManager.blockManagerId.executorId) {
// Filter out zero-sized blocks
localBlocks ++= blockInfos.filter(_._2 != 0).map(_._1)
numBlocksToFetch += localBlocks.size
} else {
val iterator = blockInfos.iterator
var curRequestSize = 0L
var curBlocks = new ArrayBuffer[(BlockId, Long)]
while (iterator.hasNext) {
val (blockId, size) = iterator.next()
if (size > 0) {
curBlocks += ((blockId, size))
remoteBlocks += blockId
numBlocksToFetch += 1
curRequestSize += size
} else if (size < 0) {
throw new BlockException(blockId, "Negative block size " + size)
}
if (curRequestSize >= targetRequestSize ||
curBlocks.size >= maxBlocksInFlightPerAddress) {
remoteRequests += new FetchRequest(address, curBlocks)
curBlocks = new ArrayBuffer[(BlockId, Long)]
curRequestSize = 0
}
}
// Add in the final request
if (curBlocks.nonEmpty) {
remoteRequests += new FetchRequest(address, curBlocks)
}
}
}
remoteRequests
}
拆分完获取任务就要直接开始获取任务了。这里是有一个请求速率控制的机制在里面，分别是maxBytesInFlight和maxRequestInFlight，如果能发送请求则发送，否则放到延迟队列中，等待下一次调用这个方法的时候去发送请求。
private def fetchUpToMaxBytes(): Unit = {
// Send fetch requests up to maxBytesInFlight. If you cannot fetch from a remote host
// immediately, defer the request until the next time it can be processed.
// Process any outstanding deferred fetch requests if possible.
if (deferredFetchRequests.nonEmpty) {
for ((remoteAddress, defReqQueue) <- deferredFetchRequests) {
while (isRemoteBlockFetchable(defReqQueue) &&
!isRemoteAddressMaxedOut(remoteAddress, defReqQueue.front)) {
val request = defReqQueue.dequeue()
logDebug(s"Processing deferred fetch request for $remoteAddress with "
+ s"${request.blocks.length} blocks")
send(remoteAddress, request)
if (defReqQueue.isEmpty) {
deferredFetchRequests -= remoteAddress
}
}
}
}
// Process any regular fetch requests if possible.
while (isRemoteBlockFetchable(fetchRequests)) {
val request = fetchRequests.dequeue()
val remoteAddress = request.address
if (isRemoteAddressMaxedOut(remoteAddress, request)) {
logDebug(s"Deferring fetch request for $remoteAddress with ${request.blocks.size} blocks")
val defReqQueue = deferredFetchRequests.getOrElse(remoteAddress, new Queue[FetchRequest]())
defReqQueue.enqueue(request)
deferredFetchRequests(remoteAddress) = defReqQueue
} else {
send(remoteAddress, request)
}
}
def send(remoteAddress: BlockManagerId, request: FetchRequest): Unit = {
sendRequest(request)
numBlocksInFlightPerAddress(remoteAddress) =
numBlocksInFlightPerAddress.getOrElse(remoteAddress, 0) + request.blocks.size
}
def isRemoteBlockFetchable(fetchReqQueue: Queue[FetchRequest]): Boolean = {
fetchReqQueue.nonEmpty &&
(bytesInFlight == 0 ||
(reqsInFlight + 1 <= maxReqsInFlight &&
bytesInFlight + fetchReqQueue.front.size <= maxBytesInFlight))
}
// Checks if sending a new fetch request will exceed the max no. of blocks being fetched from a
// given remote address.
def isRemoteAddressMaxedOut(remoteAddress: BlockManagerId, request: FetchRequest): Boolean = {
numBlocksInFlightPerAddress.getOrElse(remoteAddress, 0) + request.blocks.size >
maxBlocksInFlightPerAddress
}
}
发送请求实际借助了ShuffleClient这个类，这个类会使用NettyBlockTransferService这个类向远程的BlockManager发起获取数据块请求。
整个过程是异步的，并在发请求的时候增加了回调。回调便是在获取结束之后把结果放到一个LinkedBlockingQueue里面。
之后便是获取本地的数据块，从本地的BlockManager直接获取即可。
但实际上这个类是一个Iterator，会不断被外部调用next()方法。
next方法实际是阻塞的，因为如果results这个队列是空的则一直阻塞在这里。
每次从队列中获取一个获取结果之后做相应的处理，包装成一个InputStream
每次调用next的时候都会获取一个BlockId和相应的文件流而无需考虑这个文件块是否是远程和本地，因为每次调用next的时候都会调用这个fetchUpToMaxBytes方法，保证远程数据可以一直被获取。
override def next(): (BlockId, InputStream) = {
if (!hasNext) {
throw new NoSuchElementException
}
numBlocksProcessed += 1
var result: FetchResult = null
var input: InputStream = null
while (result == null) {
...
result = results.take()
...
result match {
case r @ SuccessFetchResult(blockId, address, size, buf, isNetworkReqDone) =>
if (address != blockManager.blockManagerId) {
numBlocksInFlightPerAddress(address) = numBlocksInFlightPerAddress(address) - 1
...
}
bytesInFlight -= size
if (isNetworkReqDone) {
reqsInFlight -= 1
}
val in = try {
buf.createInputStream()
} catch {
case e: IOException =>
buf.release()
throwFetchFailedException(blockId, address, e)
}
input = streamWrapper(blockId, in)
...
}
}
fetchUpToMaxBytes() // <-----------------
}
currentResult = result.asInstanceOf[SuccessFetchResult]
(currentResult.blockId, new BufferReleasingInputStream(input, this))
}
流程梳理
ResultTask 在Executor上运行，调用ShuffledRDD的iterator方法。这个方法从ShuffleManager获取一个BlockStoreShuffleReader，这个Reader内部负责了获取远程Shuffle输出文件的任务，获取之后根据combine，排序等处理数据，完成后续的运算。
