我有这个简单的火花程序。我想知道为什么所有数据都在一个分区中结束。
val l = List((30002,30000), (50006,50000), (80006,80000),
(4,0), (60012,60000), (70006,70000),
(40006,40000), (30012,30000), (30000,30000),
(60018,60000), (30020,30000), (20010,20000),
(20014,20000), (90008,90000), (14,0), (90012,90000),
(50010,50000), (100008,100000), (80012,80000),
(20000,20000), (30010,30000), (20012,20000),
(90016,90000), (18,0), (12,0), (70016,70000),
(20,0), (80020,80000), (100016,100000), (70014,70000),
(60002,60000), (40000,40000), (60006,60000),
(80000,80000), (50008,50000), (60008,60000),
(10002,10000), (30014,30000), (70002,70000),
(40010,40000), (100010,100000), (40002,40000),
(20004,20000),
(10018,10000), (50018,50000), (70004,70000),
(90004,90000), (100004,100000), (20016,20000))
val l_rdd = sc.parallelize(l, 2)
// print each item and index of the partition it belongs to
l_rdd.mapPartitionsWithIndex((index, iter) => {
iter.toList.map(x => (index, x)).iterator
}).collect.foreach(println)
// reduce on the second element of the list.
// alternatively you can use aggregateByKey
val l_reduced = l_rdd.map(x => {
(x._2, List(x._1))
}).reduceByKey((a, b) => {b ::: a})
// print the reduced results along with its partition index
l_reduced.mapPartitionsWithIndex((index, iter) => {
iter.toList.map(x => (index, x._1, x._2.size)).iterator
}).collect.foreach(println)
运行此操作时,您将看到数据(l_rdd
)分布在两个分区中。减少之后,结果RDD(l_reduced
)也有两个分区,但所有数据都在一个分区(索引0)中,另一个分区为空。即使数据很大(几GB),也会发生这种情况。不应该l_reduced
也分配到两个分区。
我使用Spark 1.6.1并且没有更改ShuffleManager。
答案 0 :(得分:3)
val l_reduced = l_rdd.map(x => {
(x._2, List(x._1))
}).reduceByKey((a, b) => {b ::: a})
参考上面的代码片段,您将按RDD的第二个字段进行分区。第二个字段中的所有数字都以0结尾。
当您致电HashPartitioner时,记录的分区号由以下function决定:
def getPartition(key: Any): Int = key match {
case null => 0
case _ => Utils.nonNegativeMod(key.hashCode, numPartitions)
}
Utils.nonNegativeMod定义为follows:
def nonNegativeMod(x: Int, mod: Int): Int = {
val rawMod = x % mod
rawMod + (if (rawMod < 0) mod else 0)
}
让我们看看当我们将上述两个逻辑应用于您的输入时会发生什么:
scala> l.map(_._2.hashCode % 2) // numPartitions = 2
res10: List[Int] = List(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
因此,所有记录最终都在分区0中。
您可以通过重新分区解决此问题:
val l_reduced = l_rdd.map(x => {
(x._2, List(x._1))
}).reduceByKey((a, b) => {b ::: a}).repartition(2)
给出:
(0,100000,4)
(0,10000,2)
(0,0,5)
(0,20000,6)
(0,60000,5)
(0,80000,4)
(1,50000,4)
(1,30000,6)
(1,90000,4)
(1,70000,5)
(1,40000,4)
或者,您可以创建custom partitioner。
答案 1 :(得分:0)
除非您另行指定,否则将根据相关密钥的哈希码完成分区,并假设哈希码将导致相对均匀的分布。在这种情况下,您的哈希码都是偶数,因此都将进入分区0。
如果这确实代表了您的数据集,那么reduceByKey
会出现过载,它会占用分区器和reduce函数。我建议为这样的数据集提供一种替代的分区算法。