Spark中的Round Robin分区如何工作? [英] How does Round Robin partitioning in Spark work?

问题描述

我很难理解Spark中的Round Robin分区.考虑以下示例.我将大小为3的Seq分为3个分区:

I've trouble to understand Round Robin Partitioning in Spark. Consider the following exampl. I split a Seq of size 3 into 3 partitions:

val df = Seq(0,1,2).toDF().repartition(3)

df.explain

== Physical Plan ==
Exchange RoundRobinPartitioning(3)
+- LocalTableScan [value#42]

现在，如果我检查分区，我将得到:

Now if I inspect the partitions, I get:

df
  .rdd
  .mapPartitionsWithIndex{case (i,rows) => Iterator((i,rows.size))}
  .toDF("partition_index","number_of_records")
  .show

+---------------+-----------------+
|partition_index|number_of_records|
+---------------+-----------------+
|              0|                0|
|              1|                2|
|              2|                1|
+---------------+-----------------+

如果我对大小为8的Seq进行同样的操作并将其分成8个分区，则歪斜会更严重:

If I do the same with Seq of size 8 and split it into 8 partitions, I get even worse skew:

(0 to 7).toDF().repartition(8)
  .rdd
  .mapPartitionsWithIndex{case (i,rows) => Iterator((i,rows.size))}
  .toDF("partition_index","number_of_records")
  .show

+---------------+-----------------+
|partition_index|number_of_records|
+---------------+-----------------+
|              0|                0|
|              1|                0|
|              2|                0|
|              3|                0|
|              4|                0|
|              5|                0|
|              6|                4|
|              7|                4|
+---------------+-----------------+

有人可以解释这种行为.据我了解循环分区，所有分区的显示大小均相同.

Can somebody explain this behavior. As far as I understand round robin partitioning, all partitions show be ~same size.

推荐答案

(已检查Spark版本2.1-2.4)

据我从

As far as I can see from ShuffleExchangeExec code, Spark tries to partition the rows directly from original partitions (via mapPartitions) without bringing anything to the driver.

逻辑是从随机选择的目标分区开始，然后以循环方法将分区分配给行.请注意，为每个源分区都选择了开始"分区，并且可能会发生冲突.

The logic is to start with a randomly picked target partition and then assign partitions to the rows in a round-robin method. Note that "start" partition is picked for each source partition and there could be collisions.

最终分布取决于许多因素:源/目标分区的数量以及数据框中的行数.

The final distribution depends on many factors: a number of source/target partitions and the number of rows in your dataframe.

这篇关于Spark中的Round Robin分区如何工作?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！