spark countByKey用法详解

来源：互联网发布：mac 10.7.5 dmg 编辑：程序博客网时间：2024/06/15 22:23

统计每个key对应的value个数，需要注意的是rdd类型是pairRdd，即键值对的形式的rdd,详细代码如下：

private static void myCountByKey(){
       SparkConf conf=new SparkConf()
       .setMaster("local")
        .setAppName("myCountByKey");
       JavaSparkContext sc=new JavaSparkContext(conf);
       List<Tuple2<String,String>> studentList=Arrays.asList(new Tuple2<String,String>("c1","cai"),new Tuple2<String,String>("c2","niao")
               ,new Tuple2<String,String>("c1","feng"),new Tuple2<String,String>("c2","jin"),new Tuple2<String,String>("c2","niao"));
       JavaPairRDD<String, String> studentRdd= sc.parallelizePairs(studentList);
       Map<String, Object> studentCounts=studentRdd.countByKey();
       for(Map.Entry<String, Object> map:studentCounts.entrySet()){
           System.out.println("key:"+map.getKey()+",values:"+map.getValue());
       }
   }

运行结果：

key:c2,values:3
key:c1,values:2

0 0