Providing jars in HDFS

[David Klim]

Hello,
I have bee running Zeppelin in yarn-client mode, and I so far I was copying 
required jars to the folder specified by spark.home 
(/opt/zeppelin/interpreter/spark/) on each cluster node. Is it possible to 
specify some HDFS location to load the jars from there instead? How can I 
configure that
Thanks!