Flink join（流流）详解（一）

最新推荐文章于 2025-04-10 10:57:59 发布

原创最新推荐文章于 2025-04-10 10:57:59 发布 · 2.5k 阅读

3 ·

CC 4.0 BY-SA版权

文章标签：

#flink

Flink 专栏收录该内容

26 篇文章

订阅专栏

本文基于Flink 1.11对两个或多个流的join进行测试。介绍了Flink join按窗口类型和join类型的分类，包括Tumbling Window Join、Sliding Window Join等。通过代码测试不同类型的join，还给出各类型join的示意图，并说明了一些注意事项，如session window触发较被动等。

本文基于flink 1.11进行测试。

前言

这里所说的join是两个或者多个流的join，涉及流批join的内容或者批批join会另写一篇文章专门说。

Flink的join按照窗口类型分可以分为：Tumbling Window Join、Sliding Window Join和Session Window Join。

按join类型分可以分为join和intervalJoin。前者类似RDBMS中的内连接，interval join使用一个公共键连接两个流的元素（我们现在称它们为A和B），其中流B的元素的时间戳与流A中元素的时间戳之间存在相对时间间隔。

inner join示意图：

interval join示意图：

本文会用到

Flink1.11 事件时间（event time）、watermark、watermarkstrategy使用详细案例文章的知识。
Flink window function及常用窗口算子文章的知识。

代码

创建用于join的两个数据源

package it.kenn.source;

import org.apache.flink.streaming.api.functions.source.SourceFunction;
import scala.Tuple3;

import java.util.Random;

public class ForJoinSource1 implements SourceFunction<Tuple3<String, Long, Double>> {
    boolean flag = true;

    @Override
    public void run(SourceContext<Tuple3<String, Long, Double>> ctx) throws Exception {
        Random random = new Random();
        while (flag) {
            int randInt = random.nextInt(100);
            ctx.collect(new Tuple3<>("S" + randInt, System.currentTimeMillis(), random.nextDouble() * 1000));
            Thread.sleep(30);
        }
    }

    @Override
    public void cancel() {
        flag = false;
    }
}
//-----------------------------------------------------------------------------------
package it.kenn.source;

import org.apache.flink.streaming.api.functions.source.SourceFunction;
import scala.Tuple3;

import java.util.Random;

public class ForJoinSource2 implements SourceFunction<Tuple3<String, Long, Double>> {
    boolean flag = true;

    @Override
    public void run(SourceContext<Tuple3<String, Long, Double>> ctx) throws Exception {
        Random random = new Random();
        while (flag) {
            int randInt = random.nextInt(110);
            ctx.collect(new Tuple3<>("S" + randInt, System.currentTimeMillis(), random.nextDouble() * 1000));
            Thread.sleep(20);
        }
    }

    @Override
    public void cancel() {
        flag = false;
    }
}

测试主程序

package it.kenn.join;

import it.kenn.source.ForJoinSource1;
import it.kenn.source.ForJoinSource2;
import org.apache.flink.api.common.eventtime.WatermarkStrategy;
import org.apache.flink.api.common.functions.JoinFunction;
import org.apache.flink.streaming.api.TimeCharacteristic;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.co.ProcessJoinFunction;
import org.apache.flink.streaming.api.windowing.assigners.SlidingEventTimeWindows;
import org.apache.flink.streaming.api.windowing.assigners.TumblingEventTimeWindows;
import org.apache.flink.streaming.api.windowing.time.Time;
import org.apache.flink.util.Collector;
import scala.Tuple3;

import java.time.Duration;

/**
 * 测试join
 */
//测试主程序
public static void main(String[] args) throws Exception {
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        StreamStreamJoinTest joinTest = new StreamStreamJoinTest();
        //设置事件时间
        env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
        DataStream<Tuple3<String, Long, Double>> source30 = env.addSource(new ForJoinSource1())
                //指定时间戳和watermark规则，注意要指定两个：forBoundedOutOfOrderness指定watermark生成策略，withTimestampAssigner指定那个字段是事件时间
                .assignTimestampsAndWatermarks(WatermarkStrategy.<Tuple3<String, Long, Double>>forBoundedOutOfOrderness(Duration.ofMillis(10)).withTimestampAssigner((e, ts) -> e._2()));
        DataStream<Tuple3<String, Long, Double>> source20 = env.addSource(new ForJoinSource2())
                .assignTimestampsAndWatermarks(WatermarkStrategy.<Tuple3<String, Long, Double>>forBoundedOutOfOrderness(Duration.ofMillis(10)).withTimestampAssigner((e, ts) -> e._2()));
        DataStream<Tuple3<String, Long, Double>> tumbJoinedStream = joinTest.tumbJoin(source20, source30);
        DataStream<Tuple3<String, Long, Double>> slidingJoinStream = joinTest.slidingJoin(source20, source30);
        DataStream<Tuple3<String, Long, Double>> intervalJoinStream = joinTest.intervalJoin(source20, source30);
        //对不同join进行测试
        tumbJoinedStream.print();
        env.execute();
    }

inner Tumbling Window Join代码测试

/**
     * inner Tumbling Window Join
     *
     * @param source20
     * @param source30
     * @return
     */
    public DataStream<Tuple3<String, Long, Double>> tumbJoin(DataStream<Tuple3<String, Long, Double>> source20, DataStream<Tuple3<String, Long, Double>> source30) {
        DataStream<Tuple3<String, Long, Double>> joinedStream = source20.join(source30)
                .where(e -> e._1().split("-")[1])//左流要join的字段
                .equalTo(e -> e._1().split("-")[1])//右侧流要join的字段
                .window(TumblingEventTimeWindows.of(Time.milliseconds(50)))//指定窗口类型和窗口大小
                //join函数，这里说是join但是跟数据库的join有一些区别，比如下面的逻辑并没有取两个流中的数据，而是比较两个流中数据的大小，只返回某个流中的数据
                .apply(new JoinFunction<Tuple3<String, Long, Double>, Tuple3<String, Long, Double>, Tuple3<String, Long, Double>>() {
                    @Override
                    public Tuple3<String, Long, Double> join(Tuple3<String, Long, Double> left, Tuple3<String, Long, Double> right) throws Exception {
                        return left._3() > right._3() ? left : right;
                    }
                });
        return joinedStream;
    }

下图是tumble window join的示意图，但是下面join结果有些歧义，像是笛卡尔积。其实只要在join的时候加上where条件就不可能会产生下面笛卡尔积的情况了。

下图还有一个信息点，在最后一个窗口的时候，只有橙色流中有数据，绿色流中并没有数据，那么这个窗口的计算不会被触发。

sliding Join 测试

/**
     * sliding Join 测试
     *
     * @param source20
     * @param source30
     * @return
     */
    public DataStream<Tuple3<String, Long, Double>> slidingJoin(DataStream<Tuple3<String, Long, Double>> source20, DataStream<Tuple3<String, Long, Double>> source30) {
        DataStream<Tuple3<String, Long, Double>> joinedStream = source20.join(source30)
                .where(e -> e._1().split("-")[1])//左流要join的字段
                .equalTo(e -> e._1().split("-")[1])//右侧流要join的字段
                .window(SlidingEventTimeWindows.of(Time.milliseconds(50), Time.milliseconds(30)))//指定窗口类型和窗口大小
                .apply(new JoinFunction<Tuple3<String, Long, Double>, Tuple3<String, Long, Double>, Tuple3<String, Long, Double>>() {
                    @Override
                    public Tuple3<String, Long, Double> join(Tuple3<String, Long, Double> left, Tuple3<String, Long, Double> right) throws Exception {
                        return left._3() > right._3() ? left : right;
                    }
                });
        return joinedStream;
    }

sliding join示意图：

interval join有一个需要注意的特点：有些事件可能在一个滑动窗口中没有被join但是在另外一个滑动窗口中去呗join了。比如上图橙色2号事件，在蓝色窗口中没有与绿色流join，但是在后面的绿色窗口中却与绿色3号join了。

Session Window Join测试

这种方式用到的场景好像不太多，如果哪天我用到了会在这里补上笔记的。

import org.apache.flink.api.java.functions.KeySelector;
import org.apache.flink.streaming.api.windowing.assigners.EventTimeSessionWindows;
import org.apache.flink.streaming.api.windowing.time.Time;
 
...

DataStream<Integer> orangeStream = ...
DataStream<Integer> greenStream = ...

orangeStream.join(greenStream)
    .where(<KeySelector>)
    .equalTo(<KeySelector>)
    //指定Gap大小
    .window(EventTimeSessionWindows.withGap(Time.milliseconds(1)))
    .apply (new JoinFunction<Integer, Integer, String> (){
        @Override
        public String join(Integer first, Integer second) {
            return first + "," + second;
        }
    });

session window join示意图：

上图是session window示意图。可以看到他的原理是通过两个流的间隔时间划分的窗口，这种窗口的数量非常不稳定。如果流中event间隔一直小于指定的GAP，那么窗口会一直不触发。换句话说，这种窗口的触发相比其他窗口而言比较被动，完全是数据驱动的触发，而不是时间驱动的触发。

interval Join 测试

/**
     * interval Join 测试
     *
     * @param source20
     * @param source30
     * @return
     */
    public DataStream<Tuple3<String, Long, Double>> intervalJoin(DataStream<Tuple3<String, Long, Double>> source20, DataStream<Tuple3<String, Long, Double>> source30) {
        SingleOutputStreamOperator<Tuple3<String, Long, Double>> intervalJoinedStream = source20.keyBy(e -> e._1().split("-")[1])
                .intervalJoin(source30.keyBy(e -> e._1().split("-")[1]))
                .between(Time.milliseconds(-12), Time.milliseconds(9))
                //默认情况下上面的between条件是包含边界的，如果不希望包含边界可以使用下面两个方法去除
                .lowerBoundExclusive()
                .upperBoundExclusive()
                .process(new ProcessJoinFunction<Tuple3<String, Long, Double>, Tuple3<String, Long, Double>, Tuple3<String, Long, Double>>() {
                    @Override
                    public void processElement(Tuple3<String, Long, Double> left, Tuple3<String, Long, Double> right, Context ctx, Collector<Tuple3<String, Long, Double>> out) throws Exception {
                        out.collect(left._3() > right._3() ? left : right);
                    }
                });
        return intervalJoinedStream;
    }

interval join示意图：