本文主要記錄使用SparkStreaming從Kafka里讀取數(shù)據(jù)，并使用Redis保存Offset。

相關(guān)文章：
1.Spark之PI本地
2.Spark之WordCount集群
3.SparkStreaming之讀取Kafka數(shù)據(jù)
4.SparkStreaming之使用redis保存Kafka的Offset
5.SparkStreaming之優(yōu)雅停止
6.SparkStreaming之寫數(shù)據(jù)到Kafka
7.Spark計算《西虹市首富》短評詞云

1.Redis Pool

object InternalRedisClient extends Serializable {

  @transient private var pool: JedisPool = null

  def makePool(redisHost: String, redisPort: Int, redisTimeout: Int,
               maxTotal: Int, maxIdle: Int, minIdle: Int): Unit = {
    makePool(redisHost, redisPort, redisTimeout, maxTotal, maxIdle, minIdle, true, false, 10000)
  }

  def makePool(redisHost: String, redisPort: Int, redisTimeout: Int,
               maxTotal: Int, maxIdle: Int, minIdle: Int, testOnBorrow: Boolean,
               testOnReturn: Boolean, maxWaitMillis: Long): Unit = {
    if (pool == null) {
      val poolConfig = new GenericObjectPoolConfig()
      poolConfig.setMaxTotal(maxTotal)
      poolConfig.setMaxIdle(maxIdle)
      poolConfig.setMinIdle(minIdle)
      poolConfig.setTestOnBorrow(testOnBorrow)
      poolConfig.setTestOnReturn(testOnReturn)
      poolConfig.setMaxWaitMillis(maxWaitMillis)
      pool = new JedisPool(poolConfig, redisHost, redisPort, redisTimeout, "root_jinkun")

      val hook = new Thread {
        override def run = pool.destroy()
      }
      sys.addShutdownHook(hook.run)
    }
  }

  def getPool: JedisPool = {
    assert(pool != null)
    pool
  }
}

2.KafkaRedisStreaming

object KafkaRedisStreaming {
  private val LOG = LoggerFactory.getLogger("KafkaRedisStreaming")

  def initRedisPool() = {
    // Redis configurations
    val maxTotal = 20
    val maxIdle = 10
    val minIdle = 1
    val redisHost = "127.0.0.1"
    val redisPort = 6379
    val redisTimeout = 30000
    InternalRedisClient.makePool(redisHost, redisPort, redisTimeout, maxTotal, maxIdle, minIdle)
  }

  /**
    * 從redis里獲取Topic的offset值
    *
    * @param topicName
    * @param partitions
    * @return
    */
  def getLastCommittedOffsets(topicName: String, partitions: Int): Map[TopicPartition, Long] = {
    if (LOG.isInfoEnabled())
      LOG.info("||--Topic:{},getLastCommittedOffsets from Redis--||", topicName)

    //從Redis獲取上一次存的Offset
    val jedis = InternalRedisClient.getPool.getResource
    val fromOffsets = collection.mutable.HashMap.empty[TopicPartition, Long]
    for (partition <- 0 to partitions - 1) {
      val topic_partition_key = topicName + "_" + partition
      val lastSavedOffset = jedis.get(topic_partition_key)
      val lastOffset = if (lastSavedOffset == null) 0L else lastSavedOffset.toLong
      fromOffsets += (new TopicPartition(topicName, partition) -> lastOffset)
    }
    jedis.close()

    fromOffsets.toMap
  }

  def main(args: Array[String]): Unit = {
    //初始化Redis Pool
    initRedisPool()

    val conf = new SparkConf()
      .setAppName("ScalaKafkaStream")
      .setMaster("local[2]")

    val sc = new SparkContext(conf)
    sc.setLogLevel("WARN")

    val ssc = new StreamingContext(sc, Seconds(3))

    val bootstrapServers = "hadoop1:9092,hadoop2:9092,hadoop3:9092"
    val groupId = "kafka-test-group"
    val topicName = "Test"
    val maxPoll = 20000

    val kafkaParams = Map(
      ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG -> bootstrapServers,
      ConsumerConfig.GROUP_ID_CONFIG -> groupId,
      ConsumerConfig.MAX_POLL_RECORDS_CONFIG -> maxPoll.toString,
      ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG -> classOf[StringDeserializer],
      ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG -> classOf[StringDeserializer]
    )

    // 這里指定Topic的Partition的總數(shù)
    val fromOffsets = getLastCommittedOffsets(topicName, 3)

    // 初始化KafkaDS
    val kafkaTopicDS =
      KafkaUtils.createDirectStream(ssc, LocationStrategies.PreferConsistent, ConsumerStrategies.Assign[String, String](fromOffsets.keys.toList, kafkaParams, fromOffsets))

    kafkaTopicDS.foreachRDD(rdd => {
      val offsetRanges = rdd.asInstanceOf[HasOffsetRanges].offsetRanges

      // 如果rdd有數(shù)據(jù)
      if (!rdd.isEmpty()) {
        val jedis = InternalRedisClient.getPool.getResource
        val p = jedis.pipelined()
        p.multi() //開啟事務

        // 處理數(shù)據(jù)
        rdd
          .map(_.value)
          .flatMap(_.split(" "))
          .map(x => (x, 1L))
          .reduceByKey(_ + _)
          .sortBy(_._2, false)
          .foreach(println)

        //更新Offset
        offsetRanges.foreach { offsetRange =>
          println("partition : " + offsetRange.partition + " fromOffset:  " + offsetRange.fromOffset + " untilOffset: " + offsetRange.untilOffset)
          val topic_partition_key = offsetRange.topic + "_" + offsetRange.partition
          p.set(topic_partition_key, offsetRange.untilOffset + "")
        }

        p.exec() //提交事務
        p.sync //關(guān)閉pipeline
        jedis.close()
      }
    })

    ssc.start()
    ssc.awaitTermination()
  }
}

程序運行結(jié)果如下：

redis中對應的Partition的key如下：

3.參考

http://lxw1234.com/archives/2018/02/901.html