Github user ruseel commented on the issue:
https://github.com/apache/zeppelin/pull/2565
What I want to accomplish eventually is providing a single spark cluster to
four people and interactive query they issued is run under a single pool, with
a fair scheduler. ---
