美文网首页大数据技术干货
spark RDD,reduceByKey vs groupBy

spark RDD,reduceByKey vs groupBy

作者: 大数据_zzzzMing | 来源:发表于2018-10-28 22:01 被阅读1次

    Spark 中有两个类似的api,分别是 reduceByKey 和 groupByKey 。这两个的功能类似,但底层实现却有些不同,那么为什么要这样设计呢?我们来从源码的角度分析一下。

    先看两者的调用顺序(都是使用默认的Partitioner,即defaultPartitioner)

    所用 spark 版本:spark 2.1.0

    先看reduceByKey

    Step1

      def reduceByKey(func: (V, V) => V): RDD[(K, V)] = self.withScope {
        reduceByKey(defaultPartitioner(self), func)
      }
    

    Setp2

      def reduceByKey(partitioner: Partitioner, func: (V, V) => V): RDD[(K, V)] = self.withScope {
        combineByKeyWithClassTag[V]((v: V) => v, func, func, partitioner)
      }
    

    Setp3

    def combineByKeyWithClassTag[C](
          createCombiner: V => C,
          mergeValue: (C, V) => C,
          mergeCombiners: (C, C) => C,
          partitioner: Partitioner,
          mapSideCombine: Boolean = true,
          serializer: Serializer = null)(implicit ct: ClassTag[C]): RDD[(K, C)] = self.withScope {
        require(mergeCombiners != null, "mergeCombiners must be defined") // required as of Spark 0.9.0
        if (keyClass.isArray) {
          if (mapSideCombine) {
            throw new SparkException("Cannot use map-side combining with array keys.")
          }
          if (partitioner.isInstanceOf[HashPartitioner]) {
            throw new SparkException("HashPartitioner cannot partition array keys.")
          }
        }
        val aggregator = new Aggregator[K, V, C](
          self.context.clean(createCombiner),
          self.context.clean(mergeValue),
          self.context.clean(mergeCombiners))
        if (self.partitioner == Some(partitioner)) {
          self.mapPartitions(iter => {
            val context = TaskContext.get()
            new InterruptibleIterator(context, aggregator.combineValuesByKey(iter, context))
          }, preservesPartitioning = true)
        } else {
          new ShuffledRDD[K, V, C](self, partitioner)
            .setSerializer(serializer)
            .setAggregator(aggregator)
            .setMapSideCombine(mapSideCombine)
        }
      }
    

    姑且不去看方法里面的细节,我们会只要知道最后调用的是 combineByKeyWithClassTag 这个方法。这个方法有两个参数我们来重点看一下,

    def combineByKeyWithClassTag[C](
          createCombiner: V => C,
          mergeValue: (C, V) => C,
          mergeCombiners: (C, C) => C,
          partitioner: Partitioner,
          mapSideCombine: Boolean = true,
          serializer: Serializer = null)
    

    首先是 partitioner 参数 ,这个即是 RDD 的分区设置。除了默认的 defaultPartitioner,Spark 还提供了 RangePartitioner 和 HashPartitioner 外,此外用户也可以自定义 partitioner 。通过源码可以发现如果是 HashPartitioner 的话,那么是会抛出一个错误的。

    然后是 mapSideCombine 参数 ,这个参数正是 reduceByKey 和 groupByKey 最大不同的地方,它决定是是否会先在节点上进行一次 Combine 操作,下面会有更具体的例子来介绍。

    然后是groupByKey

    Step1

      def groupByKey(): RDD[(K, Iterable[V])] = self.withScope {
        groupByKey(defaultPartitioner(self))
      }
    

    Step2

      def groupByKey(partitioner: Partitioner): RDD[(K, Iterable[V])] = self.withScope {
        // groupByKey shouldn't use map side combine because map side combine does not
        // reduce the amount of data shuffled and requires all map side data be inserted
        // into a hash table, leading to more objects in the old gen.
        val createCombiner = (v: V) => CompactBuffer(v)
        val mergeValue = (buf: CompactBuffer[V], v: V) => buf += v
        val mergeCombiners = (c1: CompactBuffer[V], c2: CompactBuffer[V]) => c1 ++= c2
        val bufs = combineByKeyWithClassTag[CompactBuffer[V]](
          createCombiner, mergeValue, mergeCombiners, partitioner, mapSideCombine = false)
        bufs.asInstanceOf[RDD[(K, Iterable[V])]]
      }
    

    Setp3

    def combineByKeyWithClassTag[C](
          createCombiner: V => C,
          mergeValue: (C, V) => C,
          mergeCombiners: (C, C) => C,
          partitioner: Partitioner,
          mapSideCombine: Boolean = true,
          serializer: Serializer = null)(implicit ct: ClassTag[C]): RDD[(K, C)] = self.withScope {
        require(mergeCombiners != null, "mergeCombiners must be defined") // required as of Spark 0.9.0
        if (keyClass.isArray) {
          if (mapSideCombine) {
            throw new SparkException("Cannot use map-side combining with array keys.")
          }
          if (partitioner.isInstanceOf[HashPartitioner]) {
            throw new SparkException("HashPartitioner cannot partition array keys.")
          }
        }
        val aggregator = new Aggregator[K, V, C](
          self.context.clean(createCombiner),
          self.context.clean(mergeValue),
          self.context.clean(mergeCombiners))
        if (self.partitioner == Some(partitioner)) {
          self.mapPartitions(iter => {
            val context = TaskContext.get()
            new InterruptibleIterator(context, aggregator.combineValuesByKey(iter, context))
          }, preservesPartitioning = true)
        } else {
          new ShuffledRDD[K, V, C](self, partitioner)
            .setSerializer(serializer)
            .setAggregator(aggregator)
            .setMapSideCombine(mapSideCombine)
        }
      }
    

    结合上面 reduceByKey 的调用链,可以发现最终其实都是调用 combineByKeyWithClassTag 这个方法的,但调用的参数不同。
    reduceByKey的调用

    combineByKeyWithClassTag[V]((v: V) => v, func, func, partitioner)
    

    groupByKey的调用

    combineByKeyWithClassTag[CompactBuffer[V]](
          createCombiner, mergeValue, mergeCombiners, partitioner, mapSideCombine = false)
    

    正是两者不同的调用方式导致了两个方法的差别,我们分别来看

    • reduceByKey的泛型参数直接是[V],而groupByKey的泛型参数是[CompactBuffer[V]]。这直接导致了 reduceByKey 和 groupByKey 的返回值不同,前者是RDD[(K, V)],而后者是RDD[(K, Iterable[V])]

    • 然后就是mapSideCombine = false 了,这个mapSideCombine 参数的默认是true的。这个值有什么用呢,上面也说了,这个参数的作用是控制要不要在map端进行初步合并(Combine)。可以看看下面具体的例子。

    <img src="https://img2018.cnblogs.com/blog/1011838/201810/1011838-20181027094209726-2067397752.png" width="65%" />

    <img src="https://img2018.cnblogs.com/blog/1011838/201810/1011838-20181027094214131-2054597375.png" width="65%" />

    从功能上来说,可以发现 ReduceByKey 其实就是会在每个节点先进行一次合并的操作,而 groupByKey 没有。

    这么来看 ReduceByKey 的性能会比 groupByKey 好很多,因为有些工作在节点已经处理了。那么 groupByKey 为什么存在,它的应用场景是什么呢?我也不清楚,如果观看这篇文章的读者知道的话不妨在评论里说出来吧。非常感谢!

    相关文章

      网友评论

        本文标题:spark RDD,reduceByKey vs groupBy

        本文链接:https://www.haomeiwen.com/subject/xcamtqtx.html