SparkRDD之coalesce和repartition

def coalesce ( numPartitions : Int , shuffle : Boolean = false ): RDD [T]

def repartition ( numPartitions : Int ): RDD [T]

coalesce和repartition将關聯資料合并到給定數量的分區中。 repartition對資料進行shuffle

java代碼如下：

package com.cb.spark.sparkrdd;

import java.util.Arrays;

import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;

public class CoalesceExample {
	public static void main(String[] args) {
		SparkConf conf = new SparkConf().setAppName("Coalesce").setMaster("local");
		JavaSparkContext jsc = new JavaSparkContext(conf);
		JavaRDD<Integer> rdd = jsc.parallelize(Arrays.asList(1, 2, 3, 4, 5, 6, 7, 8, 9, 10), 10);
		System.out.println(rdd.partitions().size());// 10
		JavaRDD<Integer> coalesceRDD = rdd.coalesce(2, false);// 将原來十個partition的資料合并到兩個partition當中
		System.out.println(coalesceRDD.partitions().size());// 2

		JavaRDD<Integer> repartitionRDD = rdd.repartition(5);
		System.out.println(repartitionRDD.partitions().size());// 5
		jsc.stop();
	}
}