It's not difficult in Kylin. In fact, for those queries unsupported by cubes, we do already have a way to route them to hive.(refer to HiveEnumerator in source code.)
However, routing queries to hive would inevitably ruin interactive dashboard expericence, so we've not even enabled it once for our customers. (in other words, it is not well tested) Still, hive is improving too, maybe the route-to-hive would be practical very soon. -- Regards, *Bin Mahone | 马洪宾* Apache Kylin: http://kylin.io Github: https://github.com/binmahone
