Storm流之PartialKeyGrouping关键字分组

一、概述

       这种方式与按字段分组很相似，根据指定字段的值进行分组，不同的是，这种方式会考虑下游 bolt 数据处理的均衡性问题，在输入数据源关键字不平衡时会有更好的性能。 

二、代码

1.Spout

package com.test.csdn.partialkeygrouping;

import org.apache.storm.spout.SpoutOutputCollector;
import org.apache.storm.task.TopologyContext;
import org.apache.storm.topology.OutputFieldsDeclarer;
import org.apache.storm.topology.base.BaseRichSpout;
import org.apache.storm.tuple.Fields;
import org.apache.storm.tuple.Values;

import java.util.Map;
import java.util.Random;

/**
 * Created by Simon on 2018/8/1.
 */
public class PartialKeyGroupingSpout extends BaseRichSpout {
    private SpoutOutputCollector collector;
    private String[] str =  {"xiaomi","huawei","apple","oppo","vivo","lenovo","LG",
            "samsung","htc","honor","nokia","smartisan","Sony","BlackBerry","sharp"};
    @Override
    public void open(Map map, TopologyContext topologyContext, SpoutOutputCollector spoutOutputCollector) {
        this.collector=spoutOutputCollector;
    }
    @Override
    public void nextTuple() {
            try {
                Thread.sleep(1000);
                int i = new Random().nextInt(10);
                String string = str[i];
                boolean i1 = string.contains("i");
                collector.emit(new Values(i1,string));
            } catch (InterruptedException e) {
                e.printStackTrace();
            }
    }
    @Override
    public void declareOutputFields(OutputFieldsDeclarer outputFieldsDeclarer) {
        outputFieldsDeclarer.declare(new Fields("flag","string"));
    }
}

 

2.Bolt

package com.test.csdn.partialkeygrouping;

import org.apache.storm.topology.BasicOutputCollector;
import org.apache.storm.topology.OutputFieldsDeclarer;
import org.apache.storm.topology.base.BaseBasicBolt;
import org.apache.storm.tuple.Tuple;

/**
 * Created by Simon on 2018/8/1.
 */
public class PartialKeyGroupingBolt extends BaseBasicBolt {

    @Override
    public void execute(Tuple tuple, BasicOutputCollector basicOutputCollector) {
        System.out.println(Thread.currentThread().getName()+"___"+tuple.getValue(1));
    }

    @Override
    public void declareOutputFields(OutputFieldsDeclarer outputFieldsDeclarer) {

    }
}

 

3.Topo

package com.test.csdn.partialkeygrouping;

import com.test.csdn.nogrouping.NoGroupingBolt;
import com.test.csdn.nogrouping.NoGroupingSpout;
import org.apache.storm.Config;
import org.apache.storm.LocalCluster;
import org.apache.storm.topology.TopologyBuilder;
import org.apache.storm.tuple.Fields;
import org.apache.storm.utils.Utils;

/**
 * Created by Simon on 2018/8/1.
 */
public class PartialKeyGroupingTopo {
    public static void main(String[] args) {

        TopologyBuilder builder = new TopologyBuilder();

        builder.setSpout("spout", new PartialKeyGroupingSpout(), 2).setNumTasks(3);
        //指定3个，便于测试
        builder.setBolt("bolt", new PartialKeyGroupingBolt(), 3).setNumTasks(5).partialKeyGrouping("spout",new Fields("flag"));

        Config conf = new Config();
        conf.setDebug(false);

        LocalCluster cluster = new LocalCluster();
        cluster.submitTopology("PartialKeyGroupingTopo", conf, builder.createTopology());
        Utils.sleep(Long.MAX_VALUE);
        cluster.shutdown();

    }
}

 

三、运行输出