pin_zhang created SPARK-21105:
---------------------------------
Summary: Useless empty files in hive table
Key: SPARK-21105
URL: https://issues.apache.org/jira/browse/SPARK-21105
Project: Spark
Issue Type: Improvement
Components: SQL
Affects Versions: 2.0.1
Reporter: pin_zhang
case class Base(v: Option[Double])
object EmptyFiles {
def main(args: Array[String]): Unit = {
val conf = new SparkConf().setAppName("scala").setMaster("local[12]")
val ctx = new SparkContext(conf)
val spark =
SparkSession.builder().enableHiveSupport().config(conf).getOrCreate()
val seq = Seq(Base(Some(1D)), Base(Some(1D)));
val rdd = ctx.makeRDD[Base](seq)
import spark.implicits._
rdd.toDS().write.format("json").mode(SaveMode.Append).saveAsTable("EmptyFiles")
}
}
// DataSet create many useless empty files for empty partition
// if insert small RDD into the table many times, which result in too many
empty files, which slow down the query.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
