org.apache.storm.spout.SchemeAsMultiScheme Java Exaples

Source File: SingleTopology.java From nightwatch with GNU Lesser General Public License v3.0

6 votes

private static TopologyBuilder buildTopology() throws Exception {
        TopologyBuilder builder = new TopologyBuilder();
        String topicName = Configuration.getConfig().getString("rtc.mq.spout.topic");
        String groupName = Configuration.getConfig().getString("rtc.mq.spout.group");
        BrokerHosts hosts = new ZkHosts(Configuration.getConfig().getString("rtc.zk.hosts"));
        SpoutConfig spoutConfig = new SpoutConfig(hosts, topicName, "/consumers", groupName);
        spoutConfig.startOffsetTime = kafka.api.OffsetRequest.LatestTime();
        spoutConfig.zkServers = Arrays.asList(Configuration.getConfig().getString("rtc.storm.zkServers").split(","));
        spoutConfig.zkPort = Configuration.getConfig().getInt("rtc.storm.zkPort");
        spoutConfig.scheme = new SchemeAsMultiScheme(new StringScheme());
        KafkaSpout kafkaSpout = new KafkaSpout(spoutConfig);
        builder.setSpout("MQSpout", kafkaSpout, Configuration.getConfig().getInt("rtc.storm.spout.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.spout.task"));
        builder.setBolt("ExtractBolt", new ExtractBolt(), Configuration.getConfig().getInt("rtc.storm.extract.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.extract.bolt.task")).shuffleGrouping("MQSpout");
        builder.setBolt("Statistic", new StatisticBolt(), Configuration.getConfig().getInt("rtc.storm.statistic.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.statistic.bolt.task")).fieldsGrouping("ExtractBolt", new Fields(new String[]{"hashKeys"}));
//        builder.setBolt("Alarm", new AlarmBolt(), Configuration.getConfig().getInt("rtc.storm.alarm.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.alarm.bolt.task")).fieldsGrouping("Statistic", new Fields(new String[]{"EventName"}));
        return builder;
    }

Source File: ClusterTopology.java From nightwatch with GNU Lesser General Public License v3.0

6 votes

private static TopologyBuilder buildTopology() throws Exception {
        TopologyBuilder builder = new TopologyBuilder();
        String topicName = Configuration.getConfig().getString("rtc.mq.spout.topic");
        String groupName = Configuration.getConfig().getString("rtc.mq.spout.group");
        BrokerHosts hosts = new ZkHosts(Configuration.getConfig().getString("rtc.zk.hosts"));
        SpoutConfig spoutConfig = new SpoutConfig(hosts, topicName, "/consumers", groupName);

        spoutConfig.startOffsetTime = kafka.api.OffsetRequest.LatestTime();
        spoutConfig.zkServers = Arrays.asList(Configuration.getConfig().getString("rtc.storm.zkServers").split(","));
        spoutConfig.zkPort = Configuration.getConfig().getInt("rtc.storm.zkPort");
        spoutConfig.scheme = new SchemeAsMultiScheme(new StringScheme());
        KafkaSpout kafkaSpout = new KafkaSpout(spoutConfig);
        builder.setSpout("MQSpout", kafkaSpout, Configuration.getConfig().getInt("rtc.storm.spout.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.spout.task"));
        builder.setBolt("ExtractBolt", new ExtractBolt(), Configuration.getConfig().getInt("rtc.storm.extract.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.extract.bolt.task")).shuffleGrouping("MQSpout");
        builder.setBolt("Statistic", new StatisticBolt(), Configuration.getConfig().getInt("rtc.storm.statistic.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.statistic.bolt.task")).fieldsGrouping("ExtractBolt", new Fields(new String[]{"hashKeys"}));
//        builder.setBolt("Alarm", new AlarmBolt(), Configuration.getConfig().getInt("rtc.storm.alarm.bolt.parallelismHint")).setNumTasks(Configuration.getConfig().getInt("rtc.storm.alarm.bolt.task")).fieldsGrouping("Statistic", new Fields(new String[]{"EventName"}));
        return builder;
    }

Source File: KafkaSpoutBuilder.java From storm_spring_boot_demo with MIT License

6 votes

@Bean("kafkaSpout")
public KafkaSpout buildSpout() {
    super.setId("kafkaSpout");
    /**
     *
     * new ZkHosts(brokerZkStr):brokerZkStr逗号分隔，kafka的zookeeper集群
     * topic:storm订阅的topic，即从哪个topic读取消息
     * spout会根据config的zkRoot和id两个参数在zookeeper上为每个kafka分区创建保存kafka偏移量的路径，如：/zkRoot/id/partitionId。
     * {@link PartitionManager#committedPath()}
     * zkRoot:偏移量保存的zk根路径
     * id:如果重新运行，希望获取同样的偏移量，则设置为固定的ID
     * PS：kafka新版本已经不将偏移量保存在zookeeper了。而且也不推荐将offset写入zk（低效）。
     */
    SpoutConfig spoutConf = new SpoutConfig(new ZkHosts(brokerZkStr), topic, zkRoot, "kafkaSpout");
    spoutConf.scheme = new SchemeAsMultiScheme(new StringScheme());
    return new KafkaSpout(spoutConf);
}

Source File: KafkaStormWordCountTopology.java From Building-Data-Streaming-Applications-with-Apache-Kafka with MIT License

5 votes

public static void main(String[] args) throws Exception {

        String zkConnString = "localhost:2181";
        String topic = "words";
        BrokerHosts hosts = new ZkHosts(zkConnString);

        SpoutConfig kafkaSpoutConfig = new SpoutConfig(hosts, topic, "/" + topic,
                "wordcountID");
        kafkaSpoutConfig.startOffsetTime = kafka.api.OffsetRequest.EarliestTime();
        kafkaSpoutConfig.scheme = new SchemeAsMultiScheme(new StringScheme());

        TopologyBuilder topologyBuilder = new TopologyBuilder();
        topologyBuilder.setSpout("kafkaspout", new KafkaSpout(kafkaSpoutConfig));
        topologyBuilder.setBolt("stringsplit", new StringToWordsSpliterBolt()).shuffleGrouping("kafkaspout");
        topologyBuilder.setBolt("counter", new WordCountCalculatorBolt()).shuffleGrouping("stringsplit");

        Config config = new Config();
        config.setDebug(true);
        if (args != null && args.length > 1) {
            config.setNumWorkers(3);
            StormSubmitter.submitTopology(args[1], config, topologyBuilder.createTopology());
        } else {
            // Cap the maximum number of executors that can be spawned
            // for a component to 3
            config.setMaxTaskParallelism(3);
            // LocalCluster is used to run locally
            LocalCluster cluster = new LocalCluster();
            cluster.submitTopology("KafkaLocal", config, topologyBuilder.createTopology());
            // sleep
            try {
                Thread.sleep(10000);
            } catch (InterruptedException e) {
                // TODO Auto-generated catch block
                cluster.killTopology("KafkaToplogy");
                cluster.shutdown();
            }

            cluster.shutdown();
        }
    }

Source File: PirkTopology.java From incubator-retired-pirk with Apache License 2.0

5 votes

public static void runPirkTopology() throws PIRException
{
  // Set up Kafka parameters
  logger.info("Configuring Kafka.");
  String zkRoot = "/" + kafkaTopic + "_pirk_storm";
  BrokerHosts zkHosts = new ZkHosts(brokerZk);
  SpoutConfig kafkaConfig = new SpoutConfig(zkHosts, kafkaTopic, zkRoot, kafkaClientId);
  kafkaConfig.ignoreZkOffsets = forceFromStart;

  // Create conf
  logger.info("Retrieving Query and generating Storm conf.");
  Config conf = createStormConf();
  Query query = StormUtils.getQuery(useHdfs, hdfsUri, queryFile);
  conf.put(StormConstants.N_SQUARED_KEY, query.getNSquared().toString());
  conf.put(StormConstants.QUERY_INFO_KEY, query.getQueryInfo().toMap());

  // Configure this for different types of input data on Kafka.
  kafkaConfig.scheme = new SchemeAsMultiScheme(new PirkHashScheme(conf));

  // Create topology
  StormTopology topology = getPirkTopology(kafkaConfig);

  // Run topology
  logger.info("Submitting Pirk topology to Storm...");
  try
  {
    StormSubmitter.submitTopologyWithProgressBar(topologyName, conf, topology);
  } catch (AlreadyAliveException | InvalidTopologyException | AuthorizationException e)
  {
    throw new PIRException(e);
  }

}

Source File: KafkaStormIntegrationTest.java From incubator-retired-pirk with Apache License 2.0

5 votes

private SpoutConfig setUpTestKafkaSpout(Config conf)
{
  ZkHosts zkHost = new ZkHosts(zookeeperLocalCluster.getConnectString());

  SpoutConfig kafkaConfig = new SpoutConfig(zkHost, topic, "/pirk_test_root", "pirk_integr_test_spout");
  kafkaConfig.scheme = new SchemeAsMultiScheme(new PirkHashScheme(conf));
  logger.info("KafkaConfig initialized...");

  return kafkaConfig;
}

Source File: IPFraudDetectionTopology.java From Building-Data-Streaming-Applications-with-Apache-Kafka with MIT License

4 votes

public static void main(String[] args) throws AlreadyAliveException, InvalidTopologyException, AuthorizationException {
    Intialize(args[0]);
    logger.info("Successfully loaded Configuration ");


    BrokerHosts hosts = new ZkHosts(zkhost);
    SpoutConfig spoutConfig = new SpoutConfig(hosts, inputTopic, "/" + KafkaBroker, consumerGroup);
    spoutConfig.scheme = new SchemeAsMultiScheme(new StringScheme());
    spoutConfig.startOffsetTime = kafka.api.OffsetRequest.EarliestTime();
    KafkaSpout kafkaSpout = new KafkaSpout(spoutConfig);
    String[] partNames = {"status_code"};
    String[] colNames = {"date", "request_url", "protocol_type", "status_code"};

    DelimitedRecordHiveMapper mapper = new DelimitedRecordHiveMapper().withColumnFields(new Fields(colNames))
            .withPartitionFields(new Fields(partNames));


    HiveOptions hiveOptions;
    //make sure you change batch size and all paramtere according to requirement
    hiveOptions = new HiveOptions(metaStoreURI, dbName, tblName, mapper).withTxnsPerBatch(250).withBatchSize(2)
            .withIdleTimeout(10).withCallTimeout(10000000);

    logger.info("Creating Storm Topology");
    TopologyBuilder builder = new TopologyBuilder();

    builder.setSpout("KafkaSpout", kafkaSpout, 1);

    builder.setBolt("frauddetect", new FraudDetectorBolt()).shuffleGrouping("KafkaSpout");
    builder.setBolt("KafkaOutputBolt",
            new IPFraudKafkaBolt(zkhost, "kafka.serializer.StringEncoder", KafkaBroker, outputTopic), 1)
            .shuffleGrouping("frauddetect");

    builder.setBolt("HiveOutputBolt", new IPFraudHiveBolt(), 1).shuffleGrouping("frauddetect");
    builder.setBolt("HiveBolt", new HiveBolt(hiveOptions)).shuffleGrouping("HiveOutputBolt");

    Config conf = new Config();
    if (args != null && args.length > 1) {
        conf.setNumWorkers(3);
        logger.info("Submiting  topology to storm cluster");

        StormSubmitter.submitTopology(args[1], conf, builder.createTopology());
    } else {
        // Cap the maximum number of executors that can be spawned
        // for a component to 3
        conf.setMaxTaskParallelism(3);
        // LocalCluster is used to run locally
        LocalCluster cluster = new LocalCluster();
        logger.info("Submitting  topology to local cluster");
        cluster.submitTopology("KafkaLocal", conf, builder.createTopology());
        // sleep
        try {
            Thread.sleep(10000);
        } catch (InterruptedException e) {
            // TODO Auto-generated catch block
            logger.error("Exception ocuured" + e);
            cluster.killTopology("KafkaToplogy");
            logger.info("Shutting down cluster");
            cluster.shutdown();
        }
        cluster.shutdown();

    }

}

Source File: AdvertisingTopology.java From streaming-benchmarks with Apache License 2.0

4 votes

public static void main(String[] args) throws Exception {
    TopologyBuilder builder = new TopologyBuilder();

    Options opts = new Options();
    opts.addOption("conf", true, "Path to the config file.");

    CommandLineParser parser = new DefaultParser();
    CommandLine cmd = parser.parse(opts, args);
    String configPath = cmd.getOptionValue("conf");
    Map commonConfig = Utils.findAndReadConfigFile(configPath, true);

    String zkServerHosts = joinHosts((List<String>)commonConfig.get("zookeeper.servers"),
                                     Integer.toString((Integer)commonConfig.get("zookeeper.port")));
    String redisServerHost = (String)commonConfig.get("redis.host");
    String kafkaTopic = (String)commonConfig.get("kafka.topic");
    int kafkaPartitions = ((Number)commonConfig.get("kafka.partitions")).intValue();
    int workers = ((Number)commonConfig.get("storm.workers")).intValue();
    int ackers = ((Number)commonConfig.get("storm.ackers")).intValue();
    int cores = ((Number)commonConfig.get("process.cores")).intValue();
    int parallel = Math.max(1, cores/7);

    ZkHosts hosts = new ZkHosts(zkServerHosts);



    SpoutConfig spoutConfig = new SpoutConfig(hosts, kafkaTopic, "/" + kafkaTopic, UUID.randomUUID().toString());
    spoutConfig.scheme = new SchemeAsMultiScheme(new StringScheme());
    KafkaSpout kafkaSpout = new KafkaSpout(spoutConfig);

    builder.setSpout("ads", kafkaSpout, kafkaPartitions);
    builder.setBolt("event_deserializer", new DeserializeBolt(), parallel).shuffleGrouping("ads");
    builder.setBolt("event_filter", new EventFilterBolt(), parallel).shuffleGrouping("event_deserializer");
    builder.setBolt("event_projection", new EventProjectionBolt(), parallel).shuffleGrouping("event_filter");
    builder.setBolt("redis_join", new RedisJoinBolt(redisServerHost), parallel).shuffleGrouping("event_projection");
    builder.setBolt("campaign_processor", new CampaignProcessor(redisServerHost), parallel*2)
        .fieldsGrouping("redis_join", new Fields("campaign_id"));

    Config conf = new Config();

    if (args != null && args.length > 0) {
        conf.setNumWorkers(workers);
        conf.setNumAckers(ackers);
        StormSubmitter.submitTopologyWithProgressBar(args[0], conf, builder.createTopology());
    }
    else {

        LocalCluster cluster = new LocalCluster();
        cluster.submitTopology("test", conf, builder.createTopology());
        org.apache.storm.utils.Utils.sleep(10000);
        cluster.killTopology("test");
        cluster.shutdown();
    }
}

org.apache.storm.spout.SchemeAsMultiScheme Java Examples