03-02-2015 11:01 AM
I already have a CDH 5.3.0 cluster and wanted to know how to try out Hive-on-Spark without having to install the Hive-on-spark beta parcel on the cluster:
03-02-2015 01:27 PM
I would like to use against the same dataset that the current Hive queries are running to compare the performance. Ideally get some sort of jar and use it on a single node of the cluster. How can I do that?