RDD repartition过程源码阅读

 

将rdd的分区数量由少变多，需要调用repartition算子。

  /**
   * reparition实际调用coalesce，并且shuffle=true
  /
  def repartition(numPartitions: Int)(implicit ord: Ordering[T] = null): RDD[T] = withScope {
    coalesce(numPartitions, shuffle = true)
  }

每条数据重新分区的计算规则：

def coalesce(numPartitions: Int, shuffle: Boolean = false,
               partitionCoalescer: Option[PartitionCoalescer] = Option.empty)
              (implicit ord: Ordering[T] = null)
      : RDD[T] = withScope {
    require(numPartitions > 0, s"Number of partitions ($numPartitions) must be positive.")
    if (shuffle) {
      /** Distributes elements evenly across output partitions, starting from a random partition. */
      val distributePartition = (index: Int, items: Iterator[T]) => {
        // 用分区号作为随机数的种子，随机取[0,numPartitions)范围内的一个整数
        var position = (new Random(index)).nextInt(numPartitions)
        items.map { t =>
          // Note that the hash code of the key will just be the key itself. The HashPartitioner
          // will mod it with the number of total partitions.
          // 用取得的随机数作为posittion的初始位置，将items数据集中的每条数据按照循序加1，形成带有位置信息的数据
          position = position + 1
          (position, t)
        }
      } : Iterator[(Int, T)]

      // include a shuffle step so that our upstream tasks are still distributed
      new CoalescedRDD(
        new ShuffledRDD[Int, T, T](mapPartitionsWithIndex(distributePartition),
        // 使用HashParititioner将数据再分配均匀分配到各个分区中
        new HashPartitioner(numPartitions)),
        numPartitions,
        partitionCoalescer).values
    } else {
      new CoalescedRDD(this, numPartitions, partitionCoalescer)
    }
  }

HashPartitioner

class HashPartitioner(partitions: Int) extends Partitioner {
  require(partitions >= 0, s"Number of partitions ($partitions) cannot be negative.")

  def numPartitions: Int = partitions

  def getPartition(key: Any): Int = key match {
    case null => 0
    // 取key的hashCode，然后对numPartitions取模
    case _ => Utils.nonNegativeMod(key.hashCode, numPartitions)
  }

  override def equals(other: Any): Boolean = other match {
    case h: HashPartitioner =>
      h.numPartitions == numPartitions
    case _ =>
      false
  }

  override def hashCode: Int = numPartitions
}


private[spark] object Utils extends Logging {

  def nonNegativeMod(x: Int, mod: Int): Int = {
    val rawMod = x % mod
    rawMod + (if (rawMod < 0) mod else 0)
  }
}