spark2.x由浅入深深到底系列六之RDD java api详解三

得过且过

　　学习任何spark知识点之前请先正确理解spark，可以参考：正确理解spark
　　

　　本文详细介绍了spark key-value类型的rdd java api
　　

　　一、key-value类型的RDD的创建方式
　　1、sparkContext.parallelizePairs

JavaPairRDD javaPairRDD =
        sc.parallelizePairs(Arrays.asList(new Tuple2("test", 3), new Tuple2("kkk", 3)));
//结果：[(test,3), (kkk,3)]
System.out.println("javaPairRDD = " + javaPairRDD.collect());　　2、keyBy的方式
public class User implements Serializable {
    private String userId;
    private Integer amount;
    public User(String userId, Integer amount) {
        this.userId = userId;
        this.amount = amount;
    }
    @Override
    public String toString() {
        return "User{" +
                "userId='" + userId + '\'' +
                ", amount=" + amount +
                '}';
    }
}
JavaRDD userJavaRDD = sc.parallelize(Arrays.asList(new User("u1", 20)));
JavaPairRDD userJavaPairRDD = userJavaRDD.keyBy(new Function() {
    @Override
    public String call(User user) throws Exception {
        return user.getUserId();
    }
});
//结果：[(u1,User{userId='u1', amount=20})]
System.out.println("userJavaPairRDD = " + userJavaPairRDD.collect());　　3、zip的方式
JavaRDD rdd = sc.parallelize(Arrays.asList(1, 1, 2, 3, 5, 8, 13));
//两个rdd zip也是创建key-value类型RDD的一种方式
JavaPairRDD zipPairRDD = rdd.zip(rdd);
//结果：[(1,1), (1,1), (2,2), (3,3), (5,5), (8,8), (13,13)]
System.out.println("zipPairRDD = " + zipPairRDD.collect());　　4、groupBy的方式
JavaRDD rdd = sc.parallelize(Arrays.asList(1, 1, 2, 3, 5, 8, 13));
Function isEven = new Function() {
    @Override
    public Boolean call(Integer x) throws Exception {
        return x % 2 == 0;
    }
};
//将偶数和奇数分组，生成key-value类型的RDD
JavaPairRDD oddsAndEvens = rdd.groupBy(isEven);
//结果：[(false,[1, 1, 3, 5, 13]), (true,[2, 8])]
System.out.println("oddsAndEvens = " + oddsAndEvens.collect());
//结果：1
System.out.println("oddsAndEvens.partitions.size = " + oddsAndEvens.partitions().size());
oddsAndEvens = rdd.groupBy(isEven, 2);
//结果：[(false,[1, 1, 3, 5, 13]), (true,[2, 8])]
System.out.println("oddsAndEvens = " + oddsAndEvens.collect());
//结果：2
System.out.println("oddsAndEvens.partitions.size = " + oddsAndEvens.partitions().size());　　二、combineByKey
JavaPairRDD javaPairRDD =
        sc.parallelizePairs(Arrays.asList(new Tuple2("coffee", 1), new Tuple2("coffee", 2),
                new Tuple2("panda", 3), new Tuple2("coffee", 9)), 2);
//当在一个分区中遇到新的key的时候，对这个key对应的value应用这个函数
Function createCombiner = new Function() {
    @Override
    public Tuple2 call(Integer value) throws Exception {
        return new Tuple2(value, 1);
    }
};
//当在一个分区中遇到已经应用过上面createCombiner函数的key的时候，对这个key对应的value应用这个函数
Function2 mergeValue =
        new Function2() {
            @Override
            public Tuple2 call(Tuple2 acc, Integer value) throws Exception {
                return new Tuple2(acc._1() + value, acc._2() + 1);
            }
        };
//当需要对不同分区的数据进行聚合的时候应用这个函数
Function2 mergeCombiners =
        new Function2() {
            @Override
            public Tuple2 call(Tuple2 acc1, Tuple2 acc2) throws Exception {
                return new Tuple2(acc1._1() + acc2._1(), acc1._2() + acc2._2());
            }
        };
JavaPairRDD combineByKeyRDD =
        javaPairRDD.combineByKey(createCombiner, mergeValue, mergeCombiners);
//结果：[(coffee,(12,3)), (panda,(3,1))]
System.out.println("combineByKeyRDD = " + combineByKeyRDD.collect());　　combineByKey的数据流如下：

　　对于combineByKey的原理讲解详细见: spark core RDD api原理详解
　　三、aggregateByKey
JavaPairRDD aggregateByKeyRDD =
        javaPairRDD.aggregateByKey(new Tuple2(0, 0), mergeValue, mergeCombiners);
//结果：[(coffee,(12,3)), (panda,(3,1))]
System.out.println("aggregateByKeyRDD = " + aggregateByKeyRDD.collect());
//aggregateByKey是由combineByKey实现的，上面的aggregateByKey就是等于下面的combineByKeyRDD
Function createCombinerAggregateByKey =
        new Function() {
            @Override
            public Tuple2 call(Integer value) throws Exception {
                return mergeValue.call(new Tuple2(0, 0), value);
            }
        };
//结果是： [(coffee,(12,3)), (panda,(3,1))]
System.out.println(javaPairRDD.combineByKey(createCombinerAggregateByKey, mergeValue, mergeCombiners).collect());　　四、reduceByKey
JavaPairRDD reduceByKeyRDD = javaPairRDD.reduceByKey(new Function2() {
    @Override
    public Integer call(Integer value1, Integer value2) throws Exception {
        return value1 + value2;
    }
});
//结果：[(coffee,12), (panda,3)]
System.out.println("reduceByKeyRDD = " + reduceByKeyRDD.collect());
//reduceByKey底层也是combineByKey实现的，上面的reduceByKey等于下面的combineByKey
Function createCombinerReduce = new Function() {
    @Override
    public Integer call(Integer integer) throws Exception {
        return integer;
    }
};
Function2 mergeValueReduce =
        new Function2() {
            @Override
            public Integer call(Integer integer, Integer integer2) throws Exception {
                return integer + integer2;
            }
        };
//结果：[(coffee,12), (panda,3)]
System.out.println(javaPairRDD.combineByKey(createCombinerReduce, mergeValueReduce, mergeValueReduce).collect());　　五、foldByKey
JavaPairRDD foldByKeyRDD = javaPairRDD.foldByKey(0, new Function2() {
    @Override
    public Integer call(Integer integer, Integer integer2) throws Exception {
        return integer + integer2;
    }
});
//结果：[(coffee,12), (panda,3)]
System.out.println("foldByKeyRDD = " + foldByKeyRDD.collect());
//foldByKey底层也是combineByKey实现的，上面的foldByKey等于下面的combineByKey
Function2 mergeValueFold =
        new Function2() {
            @Override
            public Integer call(Integer integer, Integer integer2) throws Exception {
                return integer + integer2;
            }
        };
Function createCombinerFold = new Function() {
    @Override
    public Integer call(Integer integer) throws Exception {
        return mergeValueFold.call(0, integer);
    }
};
//结果：[(coffee,12), (panda,3)]
System.out.println(javaPairRDD.combineByKey(createCombinerFold, mergeValueFold, mergeValueFold).collect());　　六、groupByKey
JavaPairRDD groupByKeyRDD = javaPairRDD.groupByKey();
//结果：[(coffee,[1, 2, 9]), (panda,[3])]
System.out.println("groupByKeyRDD = " + groupByKeyRDD.collect());
//groupByKey底层也是combineByKey实现的，上面的groupByKey等于下面的combineByKey
Function createCombinerGroup = new Function() {
    @Override
    public List call(Integer integer) throws Exception {
        List list = new ArrayList();
        list.add(integer);
        return list;
    }
};
Function2 mergeValueGroup = new Function2() {
    @Override
    public List call(List integers, Integer integer) throws Exception {
        integers.add(integer);
        return integers;
    }
};
Function2 mergeCombinersGroup =
        new Function2() {
            @Override
            public List call(List integers, List integers2) throws Exception {
                integers.addAll(integers2);
                return integers;
            }
        };
//结果：[(coffee,[1, 2, 9]), (panda,[3])]
System.out.println(javaPairRDD.combineByKey(createCombinerGroup, mergeValueGroup, mergeCombinersGroup).collect());　　

　　对于api原理性的东西很难用文档说明清楚，如果想更深入，更透彻的理解api的原理，可以参考： spark core RDD api原理详解