流式处理新玩法，Esper和Storm的联合统计分析

一、首先是一个开源项目（storm-ester）：
https://github.com/tomdz/storm-esper

二、导入工程项目

自带的TwitterEsperSample.java例子好像不太好用，反正运行着没什么效果

自己写一个例子

总的来说就是spout模拟发送5次数据，EsperBolt接收后根据规则进行统计输出（这里是每两次计算平均值），然后最后一个bolt展示结果

1、首先是一个Spout：

public class Esper_Spout extends BaseRichSpout {    SpoutOutputCollector collector;    Map<Integer, String> toSend;    @Override    public void open(Map map, TopologyContext topologyContext, SpoutOutputCollector spoutOutputCollector) {        this.collector = spoutOutputCollector;        System.out.println(">Spout-->>open<<");        toSend = new HashMap<Integer, String>();        for (int i = 5; i < 10; i++) {            toSend.put(i, "test--" + i);        }    }    @Override    public void nextTuple() {        if (!toSend.isEmpty()) {            for (Map.Entry<Integer, String> transactionEntry : toSend.entrySet()) {                Integer transactionId = transactionEntry.getKey();                System.out.println(">Spout->send：" + transactionId);                //发送需要esper计算的数据，可多个                collector.emit(tuple(transactionId));            }            toSend.clear();        }    }    @Override    public void declareOutputFields(OutputFieldsDeclarer outputFieldsDeclarer) {        outputFieldsDeclarer.declare(new Fields("price"));    }}

2、然后写一个Bolt接收Esper处理后的结果输出

public class Esper_Bolt implements IRichBolt {    private OutputCollector collector;    @Override    public void prepare(Map map, TopologyContext topologyContext, OutputCollector outputCollector) {        this.collector = outputCollector;    }    @Override    public void execute(Tuple tuple) {        //这里就是结果啦        Object result = tuple.getValueByField("tps");        System.out.println("-->>bolt esper result--->"+result);    }    @Override    public void cleanup() {    }    @Override    public void declareOutputFields(OutputFieldsDeclarer outputFieldsDeclarer) {        outputFieldsDeclarer.declare(new Fields("word"));    }    @Override    public Map<String, Object> getComponentConfiguration() {        return null;    }}

3、最后来个main 方法运行起来

public class Esper_Sample {    public static void main(String[] args) {        Logger.getRootLogger().removeAllAppenders();        EsperBolt bolt = new EsperBolt.Builder()  .inputs().aliasStream("spout1").toEventType("Apple")                .outputs().onDefaultStream().emit("tps")                .statements().add("select avg(price) as tps from Apple.win:length_batch(2)")                .build();        TopologyBuilder builder = new TopologyBuilder();        builder.setSpout("spout1", new Esper_Spout());        builder.setBolt("get-categ", bolt) .shuffleGrouping("spout1");        builder.setBolt("get-result", new Esper_Bolt()) .shuffleGrouping("get-categ");        Config conf = new Config();        conf.setDebug(true);        LocalCluster cluster = new LocalCluster();        cluster.submitTopology("test", conf, builder.createTopology());//        Utils.sleep(3000);//        cluster.shutdown();    }}

然后就可以看到输出结果啦