如何在foreachPartition中使用SQLContext和SparkContext

Question

我想在里面使用SparkContext和SQLContext foreachPartition,但由于序列化错误而无法执行此操作.我知道这两个对象都不是可序列化的,但我认为它foreachPartition是在master上执行的,其中Spark Context和SQLContext都可用.

符号:

`msg -> Map[String,String]`
`result -> Iterable[Seq[Row]]`

这是我当前的代码(UtilsDM是一个对象extends Serializable).代码失败的部分开始从val schema =...,在那里我想写result的DataFrame,然后将其保存到拼花地板.也许我组织代码的方式效率低下,那么我想在此提出您的建议.谢谢.

// Here I am creating df from parquet file on S3
val exists = FileSystem.get(new URI("s3n://" + bucketNameCode), sc.hadoopConfiguration).exists(new Path("s3n://" + bucketNameCode + "/" + pathToSentMessages))
var df: DataFrame = null
if (exists) {
  df = sqlContext
    .read.parquet("s3n://bucket/pathToParquetFile")
}
UtilsDM.setDF(df)

// Here I process myDStream
myDStream.foreachRDD(rdd => {
  rdd.foreachPartition{iter =>
    val r = new RedisClient(UtilsDM.getHost, UtilsDM.getPort)
    val producer = UtilsDM.createProducer
    var df = UtilsDM.getDF
    val result = iter.map{ msg =>
        // ... 
        Seq(msg("key"),msg("value"))
    }

    // HERE I WANT TO WRITE result TO S3, BUT IT FAILS
    val schema = StructType(
                    StructField("key", StringType, true) ::
                    StructField("value", StringType, true)

    result.foreach { row =>
       val rdd = sc.makeRDD(row)
       val df2 = sqlContext.createDataFrame(rdd, schema)

       // If the parquet file is not created, then create it
       var df_final: DataFrame = null
       if (df != null) {
          df_final = df.unionAll(df2)
       } else {
          df_final = df2
       }
       df_final.write.parquet("s3n://bucket/pathToSentMessages)
}
  }
})

编辑:

我使用的是Spark 1.6.2和Scala 2.10.6.

Answer 1

这不可能.SparkContext,SQLContext并且SparkSession只能在驱动程序上使用.您可以在顶级使用sqlContext foreachRDD:

 myDStream.foreachRDD(rdd => {
     val df = sqlContext.createDataFrame(rdd, schema)
     ... 
 })

您无法在转换/操作中使用它:

myDStream.foreachRDD(rdd => {
     rdd.foreach { 
        val df = sqlContext.createDataFrame(...)
        ... 
     }
 })

你可能想要相当于:

myDStream.foreachRDD(rdd => {
   val foo = rdd.mapPartitions(iter => doSomethingWithRedisClient(iter))
   val df = sqlContext.createDataFrame(foo, schema)
   df.write.parquet("s3n://bucket/pathToSentMessages)
})