Are there any plans to make Sparksql have the necessary enterprise grade features to join the ranks of Presto/HiveServer2?
Some key things missing from SparkSQL: Role Based Access control authorization (either Sql standard or Apache Ranger) LDAP user/group filters Data masking plugin DoAs impersonation Resource pool configuration so one query can't consume all memory leaving other queries starved Projects like https://github.com/yaooqinn/kyuubi have sprung up to overcome limitations of "Apache Spark's primordial Thrift JDBC/ODBC Server" as the above points don't seem to be on Spark's roadmap. -- Sent from: http://apache-spark-developers-list.1001551.n3.nabble.com/ --------------------------------------------------------------------- To unsubscribe e-mail: dev-unsubscr...@spark.apache.org
