Hi, Is there any performance difference in running jobs using hive client (hive -f option inside shell scripts) vs configuring hiverserver2 and running jobs using thrift services with java programs? . Which is the preferred option and why?
Thanks, Chandra
