I have configured Cognos with Cloudera Impaala, Apache Hive and Biginsight BigSQL. Does anyone come across connecting Cognos with Spark?
Quote from: Cognos8 on 25 Jan 2016 05:26:50 AM
I have configured Cognos with Cloudera Impaala, Apache Hive and Biginsight BigSQL. Does anyone come across connecting Cognos with Spark?
Hi,
I asked my secret IBM insider about this. Although currently Cognos supports 3 different interfaces to Hadoop data - Hive, Big SQL and Impala - the current versions of these three support only MapReduce and/or Tez as data processing engines, but not Spark. Once these support Spark, this will be how Cognos connects, but at the moment they don't. It's possible IBM might also start supporting the Spark SQL interface from Cognos at some point in the future, but that may or may not happen.
Cheers!
MF (via JP (http://www.cognoise.com/index.php?action=profile;u=93906)).
CA 11.0.8 starts to support spark sql.
To be more specific, connections to a SPARK SQL Thriftserver using the SIMBA SPARK JDBC driver are supported.
You would have to license that driver from SIMBA.
Apache Hive JDBC is not supported due to defects in Apach SPARK which prevents the Apache Hive driver from working correctly.