Kafka生产者类不能实例化，需要包装成一个可实例化的类

import java.util.concurrent.Future

import org.apache.kafka.clients.producer.{KafkaProducer, ProducerRecord, RecordMetadata}

class KafkaSink[K,V](createProducer:()=> KafkaProducer[K,V]) extends Serializable{

  lazy val producer= createProducer()
  def send(topic:String, key:K, value:V):Future[RecordMetadata] =
    producer.send(new ProducerRecord[K,V](topic,key,value))

  def send(topic:String,value:V):Future[RecordMetadata] =
    producer.send(new ProducerRecord[K,V](topic,value))
}

object KafkaSink {
  import scala.collection.JavaConversions._

  def apply[K,V](config: Map[String,Object]):KafkaSink[K,V]= {
    val createProducerFunc = () => {
      val producer = new KafkaProducer[K,V](config)
      sys.addShutdownHook{
        producer.close()
      }
      producer
    }

    new KafkaSink(createProducerFunc)
  }

  def apply[K, V](config: java.util.Properties): KafkaSink[K,V] = apply(config.toMap)
}

注册为广播对象

"spark write kafka" should "be fine" in {

    val kafkaProducer:Broadcast[KafkaSink[String,String]]={
      val kafkaProducerConfig = {
        val p = new Properties();
        p.setProperty("bootstrap.servers","kafka1-c1:9092,kafka2-c1:9092,kafka3-c1:9092")
        p.put("key.serializer", "org.apache.kafka.common.serialization.StringSerializer")
        p.put("value.serializer", "org.apache.kafka.common.serialization.StringSerializer")
        p
      }
      sc.broadcast(KafkaSink[String,String](kafkaProducerConfig))
    }

    val seqList = for(i<- 0 until 100) yield (i.toString,(i*2).toString)

    val rdd = sc.makeRDD(seqList)
    rdd.foreachPartition(f=>{
      f.foreach(record => {
        kafkaProducer.value.send("topic",record._1,record._2)
      })
    })