spark.sql.shuffle.partitions和之间有什么区别spark.default.parallelism？
spark.sql.shuffle.partitions
spark.default.parallelism
我试图将它们都设置为SparkSQL,但第二阶段的任务编号始终为200.
SparkSQL
performance hadoop bigdata apache-spark apache-spark-sql
apache-spark ×1
apache-spark-sql ×1
bigdata ×1
hadoop ×1
performance ×1